Slashdot Mirror


Hotmail No Longer Accepts Long Passwords, Shortens Them For You

An anonymous reader writes "Microsoft doesn't like long passwords. In fact, the software giant not only won't let you use a really long one in Hotmail, but the company recently started prompting users to only enter the first 16 characters of their password. Let me rephrase that: if you have a password that has more than 16 characters, it will no longer work. Microsoft is making your life easier! You no longer have to input your whole password! Just put in the first 16 characters!" At least they warn you; I've run into some sites over the years that silently drop characters after an arbitrary limit.

497 comments

  1. 16 x 5 bits = 80 BIT !! by Anonymous Coward · · Score: 5, Funny

    That's enough for hotmail !!

    1. Re:16 x 5 bits = 80 BIT !! by sexconker · · Score: 2, Informative

      Where in the hell do you get 5 bits from?
      A-Za-z alone gets you past that (52), add in 0-9 and some symbols and you'll be well past 64 (2^6).

      My KeePass database lists my Hotmail address's password as having 99 bits of entropy.

    2. Re:16 x 5 bits = 80 BIT !! by Immerman · · Score: 2

      Fair argument, AC probably just miscounted, they were only off by one after all. On the other hand if you assume a non-random character distribution the actual bits-per-character are considerably lower - for example even a very crude non-predictive Huffman coding of the english language reduces common characters to three bits or less, more efficient encodings can reduce that even farther, I could see the average easily falling below 5 bits, and quite possibly below 4, even after factoring in the mangling that many people use on their passwords.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    3. Re:16 x 5 bits = 80 BIT !! by Anonymous Coward · · Score: 0

      Few use uppercase. Few use q and o and 0 and l and 1. That leaves 23 letters and 8 numbers. You have to go with the expected; leave the weirdos alone; go after the rich, fat cats so move that muscle and shake that fat.

      !!

    4. Re:16 x 5 bits = 80 BIT !! by RKBA · · Score: 3, Funny

      Where in the hell do you get 5 bits from?

      It's the old Baudot code young whippersnapper. Haven't you ever used TTYs or paper tape writer/readers?
      Get off my lawn!

    5. Re:16 x 5 bits = 80 BIT !! by DarwinSurvivor · · Score: 1

      AC probably only counted a-z (26 characters is fairly close to 32 which is 5 bits).

    6. Re:16 x 5 bits = 80 BIT !! by Guignol · · Score: 3, Funny

      I could swear I have seen this somewhere else:
      "16 chars ought to be enough for anyone"

    7. Re:16 x 5 bits = 80 BIT !! by Forty+Two+Tenfold · · Score: 1

      31.

      --
      Upward mobility is a slippery slope - the higher you climb the more you show your ass.
    8. Re:16 x 5 bits = 80 BIT !! by Anonymous Coward · · Score: 0

      By running Windows with the Baudot codepage

    9. Re:16 x 5 bits = 80 BIT !! by default+luser · · Score: 1

      Except two things:

      1. Baudot was case-insensitive. That adds another bit folks!

      2. Baudot was replaced by ASCII precisely because it was inefficient for mixed-use text/numbers/symbols. You had to send a distinct command word to change between symbol and letter tables, so a heavily-mixed data stream could use up to 10 bits per-character. They chose a constant 7 bits per-character over "maybe 5 bits, could be up to 10 bits."

      --

      Man is the animal that laughs.
      And occasionally whores for Karma.

    10. Re:16 x 5 bits = 80 BIT !! by Anonymous Coward · · Score: 0

      Well, use unicode passwords then. 16 characters only, but if some are Chinese . . .

    11. Re:16 x 5 bits = 80 BIT !! by Anonymous Coward · · Score: 0

      The real story is - How the fuck does Hotmail have the first 16 characters of my password? If they can shorten your password, it means they are storing it with ineffective encryption. No website should know what your password to log into that website it. Salt it, encrypt it, then throw away the original. Salt and encrypt any log-in attempts, and you know the password is correct if they match, even if you don't have the original password anymore.

    12. Re:16 x 5 bits = 80 BIT !! by Anonymous Coward · · Score: 0

      Nope, 32.

      When you count a quantity of things (characters), you generally start counting with "one" not "zero".

    13. Re:16 x 5 bits = 80 BIT !! by Guignol · · Score: 1

      Wow
      Is that true ?
      I was just joking above of course, but I assumed that the consequence was that your larger than 16 chars passwords would not work anymore, not that their truncated version would work as well
      Thanks for pointing that, I shall have a closer look at the (real) issue

    14. Re:16 x 5 bits = 80 BIT !! by Anonymous Coward · · Score: 0

      Be that as it may. The ability for MS to accept the first 16 characters implys that they are storing them in a recoverable format.

    15. Re:16 x 5 bits = 80 BIT !! by Joce640k · · Score: 1

      I would have hoped they're storing a hash of your password, making any limit completely arbitrary.

      I guess I was wrong.

      --
      No sig today...
  2. Hah! Take that, my bank! by Anonymous Coward · · Score: 2, Interesting

    12 letters, no special characters my ass.

    No, you may not know which bank I use.

    1. Re:Hah! Take that, my bank! by rwa2 · · Score: 4, Interesting

      At least they warn you; I've run into some sites over the years that silently drop characters after an arbitrary limit.

      Nah, they'd never do that at a reputable large financial institution... like, say, www.americanexpress.com

      Maybe they somehow figured out how to make money from handling fraud claims?

    2. Re:Hah! Take that, my bank! by Scarletdown · · Score: 1

      12 letters, no special characters my ass.

      No, you may not know which bank I use.

      Sounds like Bank of America. At least they let you use a mix of upper and lower case and numbers.

      --
      This space unintentionally left blank.
    3. Re:Hah! Take that, my bank! by Anrego · · Score: 3, Interesting

      Stupid as this whole thing is, Microsoft does make one good point.

      With the ease of phishing and harvesting passwords from other services where the user has used the same one.. who is gonna bother brute forcing a password.

      It's like if your car has a notoriously easy to pick lock.. but you park in a parking lot where no one else even bothers locking theirs (and some have even had their doors removed for even more convenience..)

    4. Re:Hah! Take that, my bank! by Stan92057 · · Score: 2

      How can brute force work on a web site sign in page? I would think banks code the site to stop brute force password input. im no programmer that's why i ask.

      --
      Jack of all trades,master of none
    5. Re:Hah! Take that, my bank! by Frosty+Piss · · Score: 1

      Sounds like Bank of America. At least they let you use a mix of upper and lower case and numbers.

      Really? My B of A password does indeed contain several "special" chars.

      And frankly, 16 character length including numbers, CAPs, lowers, and specials in a random string - is quite enough for most normal password use.

      I do have some 52 char random string passwords, but really, back to the main subject here, for Hotmail? Seriously, 16 chars is fine, if you use random strings.

      --
      If you want news from today, you have to come back tomorrow.
    6. Re:Hah! Take that, my bank! by Anrego · · Score: 3, Interesting

      It's kind of a back and forth game..

      You can't outright block access to an account after a certain number of tries because that creates an easy way to denial of service (someone can lock you out of your account just by entering a few bogus passwords). So you either block after a certain number of failed attempts (at which point botnets come into play) or install a captcha (at which point standard spam-level anti-captcha stuff comes into play.

      But my original point was that there are so many much easier ways to get accounts, why is anyone going to go through all that trouble.

      There is an argument for brute forcing when someone has broken into a server and stolen a list of hashed passwords (as then they can crank away at them all they want) so limiting to 16 chars kinda makes that a bit easier.. but I still think given hotmails user base they could easily just check against hashes for "password123" and get more than enough hits to make it not worth going further...

    7. Re:Hah! Take that, my bank! by roc97007 · · Score: 5, Interesting

      > Nah, they'd never do that at a reputable large financial institution... like, say, www.americanexpress.com

      Yeah. As you probably know, when you activate an AmEx corporate card, they require you to create a pin, and the voicemail says to use something you will remember, like the month/day of your mom's birthday.

      The automated system will actually REJECT a pin that is not a valid month/day. (Well cool. 366 total possibilities. That's not easy to brute-force at all.) I futzed with the system until I got a real person, and insisted I wanted to use a randomly generated number instead (which didn't happen to be a valid month/day). He said he couldn't do that, it had to be a date. He asked me for my mom's birthday and said he would set it to that. (My theory is that they do this to cut down on service calls.) I pointed out that this string could be uncovered by anyone with facebook access. He said that this is what it had to be. I went over his head. Eventually I found someone with the authority to set the pin to a string of my choosing. As far as I know, I'm the only AmEx card holder who has a pin set to something other than the customer's mother's birthday.

      This information (that AmEx has this requirement), could be of huge use to phishers were it ever, you know, published in a public forum.

      --
      Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
    8. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 5, Insightful

      The real question is how were they able to truncate your password if they used a hash?

    9. Re:Hah! Take that, my bank! by hobarrera · · Score: 2

      You've a pretty good bank. I've three bank accounts:
      1) 4 digits (yes, numeric only), account disabled after ONE failed attempt (need to re-enable it at an ATM).
      2) 8 letters/numbers, need to change it every 90days. Can't repeat old passwords.
      3) 4 digits, and the username is "secret" too. If you fail to log in 3 times, your account is disabled and you need to pick a new password AND username.

      While the latter offers a wierd but somewhat better form of security than the rest, you can be pretty sure that 12 letters is pretty good, since banks have extremely lousy standards!

    10. Re:Hah! Take that, my bank! by hobarrera · · Score: 1

      Someone paying attention while I type my password with a good memory might remember 16 digits. The same does not apply for 50 digits.

    11. Re:Hah! Take that, my bank! by metalmonkey · · Score: 4, Interesting

      americanexpress was the worst, the 'Set password' page input field was limited to the maximum number of characters however the 'Login' page was not.

      So if my password is: 'myreallylongpassword', it would appear accept my password. But it would be only only use 'myreallylong' as input.
      When I go to login and enter 'myreallylongpassword' it took the whole password as input and denied me access, since it didn't equal to 'myreallylong'.

      I went through quite a few password resets before I figured this out.

    12. Re:Hah! Take that, my bank! by number11 · · Score: 1

      It's kind of a back and forth game..

      You can't outright block access to an account after a certain number of tries because that creates an easy way to denial of service (someone can lock you out of your account just by entering a few bogus passwords).

      You can't? Somebody better tell my bank, and at least one of my credit card companies, and a mutual fund or two. "We're sorry, but someone has attempted to access your account with a wrong password, and access has been disabled. Please call Customer Service between 9 and 9 Eastern to have access restored."

    13. Re:Hah! Take that, my bank! by maz2331 · · Score: 3, Funny

      I'll bet that their db servers and Windows domain passwords are "sa" and "admin" as well.

    14. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      You have 12 letters on your ass, but you're too conservative to add special characters?

    15. Re:Hah! Take that, my bank! by Cinder6 · · Score: 1

      Thank you! I had to go through several AmEx passwords before I gave up and used a shorter one. Now I know I wasn't crazy.

      --
      If you can't convince them, convict them.
    16. Re:Hah! Take that, my bank! by shutdown+-p+now · · Score: 1

      This information (that AmEx has this requirement), could be of huge use to phishers were it ever, you know, published in a public forum.

      No way. They'll never guess that you have to key in the year first!

    17. Re:Hah! Take that, my bank! by amiller2571 · · Score: 1

      Banks are about the only site that do that, most email site and such only disable login for a few minutes.

    18. Re:Hah! Take that, my bank! by Immerman · · Score: 4, Insightful

      Exactly. The fact that they can do this practically screams "We haven't bothered to implement even the most basic security precautions on our password database!" I mean come on - wasn't it established that storing recoverable passwords was a bad idea back in the text-only mainframe days? I could kind of understand it if it was some backwater site created by a high-school computer wiz, but Microsoft? Sigh. Yeah *sure* I'll trust your security software to keep my home PC safe - after all you're the company that did such a great job on the OS itself that running separate security software is practically mandatory.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    19. Re:Hah! Take that, my bank! by sgt+scrub · · Score: 2

      Oh sure. After I finished trying all 366 possible numbers you tell everyone your pin isn't a date! Just for that I'm going to rob someone else.

      --
      Having to work for a living is the root of all evil.
    20. Re:Hah! Take that, my bank! by Aaron+B+Lingwood · · Score: 3, Interesting

      The only Australian bank that I use has the following setup-
      Login: Primary Account Number
      Password: 5 characters, A-Z 0-9 (no lower case)

      Account locks after 3 incorrect attempts.

      As a measure against the key-logger, the password is entered by clicking on a virtual keyboard which repositions itself on the screen randomly after each click.
      Can not login without Javascript enabled. This measure is useless on mobile devices though as the virtual keyboard fills the entire screen and thus can not be repositioned. If an attacker found a chink in the armour that would allow multiple password attempts without locking the account (likely, as this appears to be done in script), a brute-force will likely succeed in a very short amount of time.

      On the plus-side, I am informed of ALL login attempts and transactions via SMS and must login and enter a one-time-pad to authorize larger transactions. I am also given the previous login time and date upon login and a separate code is required to add a new payee or for overseas transfers.

      --
      [Rent This Space]
    21. Re:Hah! Take that, my bank! by 1u3hr · · Score: 4, Insightful

      The real question is how were they able to truncate your password if they used a hash?

      Maybe they always truncated the password, just didn't tell you.

    22. Re:Hah! Take that, my bank! by gman003 · · Score: 2

      How do they even do that?

      Anyone doing a password system is going to take the input string and hash it. Hashes always come out with the same length (for a given algorithm, at least), so there's no reason to have *any* limit on password length. And even if you truncate it on entry, why would you not either warn the user (as Wachovia did back when I used them, with their 9-12 character passwords only), or at least truncate on password check as well.

      The only logical explanation is that they're storing passwords as plain text, which I knew not to do before I even graduated high school.

    23. Re:Hah! Take that, my bank! by sg_oneill · · Score: 1

      Oh trust me. That lock out is enforced server side. Banks are not THAT stupid. Mine actually locks my account l after 4 attempts and I need to phone the bank and provide a tonne of personal details to unlock it.

      --
      Excuse the Unicode crap in my posts. That's an apostrophe, and slashdot is busted.
    24. Re:Hah! Take that, my bank! by mabhatter654 · · Score: 2

      That was my first question too.

      A proper hash setup should flag only part of a password as wrong with no way of knowing if you were close or not. This is clearly screwed up. Either the hash is deliberately weak, or some idiot admin cracked the whole damn table and made it just 16 characters ... Which is even WORSE!

    25. Re:Hah! Take that, my bank! by mabhatter654 · · Score: 1

      I have paranoids. I kick my users out after three tries... No time delay to automatically unlock.

      If only that "Customer Service" number didn't ring at home at 2am!!!

    26. Re:Hah! Take that, my bank! by mabhatter654 · · Score: 1

      But then your attempt to use the XKCD method doesn't really work.

    27. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      Thank you! I had to go through several AmEx passwords before I gave up and used a shorter one. Now I know I wasn't crazy.

      Yes, that IS what you would like to think...

    28. Re:Hah! Take that, my bank! by metalmonkey · · Score: 1

      Well they technically did give feedback, the text input field in html had the character limit.
      I didn't notice it was truncated when typing - easy to miss as text box had characters beyond the max displayable characters. They probably were hashing probably a designer just put the limit on the text box.
      The input field in the login page did not have the same limit.

    29. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 1

      The real question is how were they able to truncate your password if they used a hash?

      No, the real real question is why is anyone still using Hotmail anyway?

    30. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      As far as I know, I'm the only AmEx card holder who has a pin set to something other than the customer's mother's birthday.

      You're not the only one. They rep I spoke to was rather befuddled and had to get a supervisor when I informed her that I did not know my mother's birthday, or my mother, and could not possibly find out the required information since I was abandoned as an infant in Vietnam and my parents were never tracked down. After some long, awkward moments (too many of them really) they let me select another date, though they told me it had to be a valid date.

    31. Re:Hah! Take that, my bank! by CTU · · Score: 1

      Well instead of a lockout, how about a delay between attempts. It could be 2-5 seconds whcih won't really effect most normal people, but would stop people from trying to bruit force a password by making it take way to long to try.

    32. Re:Hah! Take that, my bank! by Tastecicles · · Score: 1

      Bruteforcing databases does not require webforms, it requires direct access to the database (usually through some sort of SQL injection exploit), beyond which point you don't even need an open browser.

      --
      Operation Guillotine is in effect.
    33. Re:Hah! Take that, my bank! by Tastecicles · · Score: 1

      Are there any Hotmail staff on hand to field this one??

      --
      Operation Guillotine is in effect.
    34. Re:Hah! Take that, my bank! by holostarr · · Score: 2

      I'm not sure, but I would assume right after you attempt to login, they would take the password you just posted, truncate it to 16 chars, MD5/SHA1 it, update their records, and inform you that your password has been truncated. That would be the only possible way other than them storing your password in plain text.

    35. Re:Hah! Take that, my bank! by happy_smile · · Score: 2

      The real question is how were they able to truncate your password if they used a hash?

      It's not impossible. Let's see one of possible scenarios when a user sign in using his/her long password:
      (1) On the server side, generate the hash from user's input and check if it is matched with the stored hash in the database.
      (2) If it is matched, continue the sign in process, and truncate the long password: take the first 16 chars of the long password input by the user, generate a new hash, update the old one, set a certain flag if necessary so that in the future only 16 chars are used to generate the hash to validate the login.

    36. Re:Hah! Take that, my bank! by Bert64 · · Score: 2

      If you block access to an account, you create a denial of service opportunity. The idea of account lockouts is thus utterly ridiculous.

      And you do nothing to someone who takes the opposite approach - try thousands of accounts with a single password (where on a system as large as hotmail, someone will have "Password1" or similar.

      Instead you really should block the source address of any obvious attacks, which while obviously not perfect (botnets, proxies etc) is at least moderately more effective.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    37. Re:Hah! Take that, my bank! by Bert64 · · Score: 1, Informative

      And you expected anything better from MS? The same company who's flagship OS not only uses an unsalted hash for storing user passwords, but actually allows you to authenticate using just the hash without ever knowing the original plaintext, thus making the hash itself the plaintext password?

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    38. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      Seems possible to me. I am not sure if somebody else posted this but if you just create a new hash of the first characters, after checking the existing hash, it should be possible.

      So
      mylongpassword > to server

      check hash of mylongpassword = ok

      create new hash of mylongp

      ???

      profit

    39. Re:Hah! Take that, my bank! by Bert64 · · Score: 1

      What happens when someone manages to guess a large number of your usernames, and then intentionally locks them all out?
      Or what happens when someone takes the aforementioned userlist and tries each account with 2 different but very common passwords, what are the odds that on a large system at least one user has qwerty1 or password1 as their password?

      Account lockouts are a stupid idea...

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    40. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      Note to self: roc97007's AmEx password is numeric, most likely 4 or 6 digits and NOT a valid date.

    41. Re:Hah! Take that, my bank! by DarwinSurvivor · · Score: 1

      I have a 2000 word dictionary of words around 4-8 characters. Randomly generated you could easily fit 3 words which is 2000^3 == 32 bits of entropy. Not great, but still difficult to brute-force on a system that undoubtedly tracks (and responds to) failed login attempts.

    42. Re:Hah! Take that, my bank! by Sir_Sri · · Score: 1

      How can brute force work on a web site sign in page?

      Interception of the packets themselves. Remember hackers will have hotmail accounts themselves where they know what is being transmitted, so they might be able to reverse engineer the packets, they could also just compromise the database.

      Brute force doesn't need to be done on the site itself if you can get an encrypted copy of password you can brute force that at your leisure.

      The thing with hotmail, and windows live in general is that the system has so many different potential points of failure it's really hard to know where an attack came from. There's the web form, the web security, general security of the OS, hotmail is hugely popular so all of the various phishing scams centred around hotmail, then there are the various messenger clients (some of which might connect to microsoft messenger and not be from microsoft itself), all of the various versions of games for windows live, god knows how it's handled internally but each department might have control over their own mechanism to access the database (so if you signed in through an xbox site you'd be passed through something managed by the xbox team etc.).

      I don't really envy anyone trying to secure systems like this. Too many things that can go wrong in too many different places, and isolating everything down to a single point of failure is still, well, a single point of failure.

    43. Re:Hah! Take that, my bank! by Havenwar · · Score: 1

      If someone is able to pay attention while you type your password, then your problem is physical security - not password length. Using that as an argument is arguing for security through obscurity, which is generally not the best choice. Better then to remove the asshole that's shoulder-surfing, or wait to access your account until you're in a private location.

    44. Re:Hah! Take that, my bank! by Havenwar · · Score: 1

      SOME banks have extremely lousy standards. Maybe even MOST banks. Personally I chose my bank based on this since I almost exclusively interact with my bank online, and I made sure to tell the banks that failed to live up to basic usability standards when I tested them exactly why I was closing my accounts.

      It might be inconvenient to switch banks, but except for a small amount of fringe cases it's definitely doable. Put that one time effort in comparison to the increased risk and headaches of constantly being exposed to a crappy security/login system, and I think you'll notice as I did that over the long run you're better off making the switch.

    45. Re:Hah! Take that, my bank! by Serious+Callers+Only · · Score: 1

      Form -> truncate -> hash -> compare

      As long as they have always dropped any characters after 16 before hashing, and continue to do so, they could easily still be salting and hashing those first 16. For the vast majority of customers it will make no difference, and for those who use phrases it will only be slightly less secure.

      If they can actually recover the pass phrase however that would be a different matter.

    46. Re:Hah! Take that, my bank! by skovnymfe · · Score: 1

      Block for a set time after certain attempts. Even if you've got the biggest botnet in the world, if you can only try 5 passwords per hour, you're screwed. Granted the denial of service point is still an issue, but as it is now, there's no easy way to get around it.

    47. Re:Hah! Take that, my bank! by Gorath99 · · Score: 1

      As far as I know, I'm the only AmEx card holder who has a pin set to something other than the customer's mother's birthday.

      You're not, or at least you wouldn't be if I still had the corporate AmEx card from my previous job. When the customer support person from AmEx asked me about my mother's birthday I told him that I'm terrible with dates (which is true) and I don't know it by heart. After that he asked me some other dates to which I gave the same reply. He then just let me have whatever 4-digit number I liked, and I picked one that isn't a valid date.

      Honestly, the process was pretty painless. Apparently this is not universally true, but the AmEx helpdesk in my region (The Netherlands) has always been polite and helpful to me.

    48. Re:Hah! Take that, my bank! by hairyfeet · · Score: 1

      I agree you just gotta look at the userbase. Working here at the shop 6 days a week i get to see pretty much what folks use and I can honestly say i haven't seen a person under 60 years old use Hotmail in fricking years. See a lot of younger Yahoo and Gmail users but NEVER Hotmail, they are always folks that have been on the thing since forever and if you got into their Hotmail you'd find a lot of recipes, emails from their Aunt Mildred, any of the data there would be pretty much useless.

      Hell I doubt even 10% of their aging userbase even uses 12 characters and I frankly wouldn't be surprised if a good 30%+ are using either their SSNs or their medicare numbers, old folks seem to love to use those numbers.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    49. Re:Hah! Take that, my bank! by DaveGod · · Score: 1

      By treating a login as a password reset.

    50. Re:Hah! Take that, my bank! by TheLink · · Score: 1

      For throwaway email accounts? It's not like I care that much about my hotmail accounts.

      And now it looks like Microsoft is telling hotmail users that they shouldn't care either.

      --
    51. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      "The only logical explanation is that they're storing passwords as plain text, which I knew not to do before I even graduated high school."

      Or they are taking the input password and cut it down before its sent to the login processor. Doesn't mean it's stored plaintext.

      Although I have to say I've ran into this a fee times and also had failed logins because of it. The set password let me Tyler as long of one as I wanted, but when I hit submit it cut it down to whatever length they had set and then encrypted and stored it. When you went to login, some of the sites would have an input limit on the password box which would auto stop at the limit and would therefore match and allow you in. Other sites though, would use the fully entered password and compare to the stored and fail to login because they didn't match. That was a pain in the ass

    52. Re:Hah! Take that, my bank! by Celarent+Darii · · Score: 1

      You mean like unix has done like forever ?

    53. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      When gp is entering his password on the set password screen, it's probably a password field, so the characters are all bullets. If the field's width is less than the width of the max chars as bullets, then it's not going to be perceptible when you are typing and characters are no longer appearing in the field. They're all bullets, it's not like you're going to verify the 'M' you just typed. Set up a password field, set the size to 8 and maxlength to 12, and see for yourself.

    54. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      And you expected anything better from MS?

      No, but I do expect better from any of my banks, and I don't get it. Their password policies are inexcusable. I have passwords at any number of sites where I don't much care if they get cracked, but if any snotty-nosed 11-year-old can theoretically crack my banking accounts, I care a lot.

    55. Re:Hah! Take that, my bank! by msauve · · Score: 1
      You didn't read the summary, did you?

      the company recently started prompting users to only enter the first 16 characters of their password.

      So, how exactly does your step 1 work, where they hash the entered 16 characters, and have it match an existing hash created from a longer password?

      Either they've been truncating to 16 characters for a minimum of one "forced password change" cycle (do they force changes?), or they don't use hashes.

      --
      "National Security is the chief cause of national insecurity." - Celine's First Law
    56. Re:Hah! Take that, my bank! by The+Atog+Lord · · Score: 1

      How can brute force work on a web site sign in page? I would think banks code the site to stop brute force password input. im no programmer that's why i ask.

      There are two types of attacks one can make against passwords: online and offline. In an online attack, the attacker just goes to the website itself and starts entering passwords.The website can just lock him out after several failed attempts; even if there is a password-reset option, this can still be very time-consuming. However, if more pernicious attack that is an offline attack. In this attack, the attacker has stolen the hashed password file, and he can spend an arbitrary amount of time breaking its passwords, limited only by the number of cycles on his computer.

      What makes a change in policy to a maximum of 16 characters absurd is that the strength of passwords really does matter when it comes to how long they will endure these offline attacks. In fact, there is evidence that using a password of at least 16 characters leads to a password that is more difficult for attackers to break.

      http://www.cylab.cmu.edu/research/techreports/2011/tr_cylab11008.html

    57. Re:Hah! Take that, my bank! by fisted · · Score: 1

      What? Client-side cannot perform the truncation before hashing or whut? +5 Insightful for such a crap comment, yay /.

    58. Re:Hah! Take that, my bank! by rollingcalf · · Score: 1

      If they've found the password file with the hashes, they can run the brute forcing on their own system without any tracking of failed login attempts.

      --
      ---------
      There is inferior bacteria on the interior of your posterior.
    59. Re:Hah! Take that, my bank! by sootman · · Score: 1

      For online banking at my bank, you pick a password that is numbers and letters only, and it is case-insensitive. (The reason for this is so you can punch it in on a phone's numeric keypad.) So, 36 possible characters. Happy cracking!

      --
      Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
    60. Re:Hah! Take that, my bank! by DarkOx · · Score: 2

      There is a way to do it securely (well as secure are max 16 char passwords are).

      1. Collect plaintext from the user
      2. hash the plaintext and validate against the old password db
      3. If success, then truncate the plain text and store the new hash.
      4. overwrite the old hash with some flag value, like say all NULLs
      5. Wait until the old password database is all NULL and dump it.

      --
      Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
    61. Re:Hah! Take that, my bank! by crossmr · · Score: 1

      No, the worst was a website here in Korea.
      they asked you to enter a password of 6-10 characters, but the text field didn't have a limit for some reason.
      when you put in a password longer than 10 characters it took it, and created your account
      when you tried to login, it would generate a DB error if you tried to input your longer than 10 character password. if you input only the first 10 characters it would fail to login
      couple this with the fact that 99.9% of websites in Korea, for some god awful reason, give you no means of contacting support without logging in, and that under Korea's real name system you were basically allowed 1 account per person..

      you were effectively locked out of the site.

    62. Re:Hah! Take that, my bank! by colinrichardday · · Score: 1

      Especially since user names are often guessable from real names, such as first initial last name..

    63. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      Of course they could still be hashing the password:

      1) User enters full >16 character password
      2) Check against existing hash
      3) If (2) passes, hash first 16 characters of password & store, overwriting existing hash

      But of course, this is slashdot, so the basement dwellers mod anything anti-Microsoft +5, while I'll be relegated to the slums of -1

    64. Re:Hah! Take that, my bank! by roc97007 · · Score: 1

      I'm sure it's different from place to place. For instance, it sounds like you didn't have to break out of an automatic validation system to get a human. That's challenging to do in my area.

      But what this also tells me is that the policy of picking mother's birthday is worldwide, not just in the western united states. That's very interesting.

      --
      Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
    65. Re:Hah! Take that, my bank! by qvatch · · Score: 1

      TD bank in canada: 5-8 letters, no symbols. Sure you have to answer a few (random 3 of 5) quiz-of-your-life-history questions after, but really. http://www.tdcanadatrust.com/products-services/banking/electronic-banking/faq-idplus.jsp Passwords must: - be 5 to 8 characters in length - not contain spaces or special characters (e.g. #, &, @)

    66. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      So why would a password system on a popular email site be deliberately weak? Everybody needs to think about this. Microsoft is not populated by idiots. They know how to do this right (ok, they fail at it sometimes, but they know). So what's the point? To allow law enforcement to snoop on you? They could already do that--they own the system.

      Speculation: since a lot of people dumbly use the same password for all their acounts, this would provide a convenient way for, let's call them "state actors", to gain your credentials by subpoena or agreement and use them on other sites without going through all that password cracking stuff and such.

      No proof of this--just speculating.

    67. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      Get access to the server through this week's vulnerability, steal the password database, and have at it.

      Whatever the "strength" of the password, do not reuse passwords!

    68. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      Just like it's always trimmed your password to 8 characters?

    69. Re:Hah! Take that, my bank! by Kharny · · Score: 1

      wow, pathetic bankaccount protection in the us and aus then....
      My (old 2000ish) dutch bank had an account nr. with a randomised number generated with a special digipass that used the bankcard, pin and time with a random seed.
      My current finnish bank uses a accountnr, pincode and a onetime pad code to access and confirm

      --
      Make a man a fire and he will be warm for a day, set a man on fire and he will be warm for the rest of his life
    70. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      That was one thing I liked about Hotmail and simultaneously dispised about my bank.

      Bank: 12 characters, limited special characters.
      Hotmail: whatever the fuck you want.

      I was kind of hoping the pasword rules would become similar because my bank would raise the bar instead of Microsoft lowering it. *sigh* Is it really that hard to accept an arbitrary string, hash it, and store the result?

    71. Re:Hah! Take that, my bank! by CTU · · Score: 1

      Don't know much about unix, but I guess...so how long will it take to try 8,000,000 passwords or what not possible with two seconds between each try...besides that many files should get someone seeing a problem long before it gets hacked.

    72. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      americanexpress was the worst, the 'Set password' page input field was limited to the maximum number of characters however the 'Login' page was not.

      It's even worse when they do it the other way round - you can set a long password, but you can't ever log in with it....

    73. Re:Hah! Take that, my bank! by skelly33 · · Score: 1

      While it sounds stupid, they could have been storing multiple hashes for every password all along. Like if they had a minimum password length of 8, they could store N hashes to get every variation from 8 to N+7 password lengths. When you create the account, log in, or change the password, they could regenerate the entire hash set at any time. They could have been preempting this move for years, and collected statistics on accounts with > 16 length passwords (I bet there are not that many, comparatively). A little advance planning here and they would not have ever had to store the plain text password.

    74. Re:Hah! Take that, my bank! by Scarletdown · · Score: 1

      Sounds like Bank of America. At least they let you use a mix of upper and lower case and numbers.

      Really? My B of A password does indeed contain several "special" chars.

      And frankly, 16 character length including numbers, CAPs, lowers, and specials in a random string - is quite enough for most normal password use.

      I do have some 52 char random string passwords, but really, back to the main subject here, for Hotmail? Seriously, 16 chars is fine, if you use random strings.

      Hmmm. They must have changed it in the recent past then. Guess next time I log in to do a transfer or whatever, I should test and see if I can give a new password with special characters.

      --
      This space unintentionally left blank.
    75. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      The only Australian bank that I use has the following setup-

      Whilst we're in Australia...

      The DSD (an approx. Aussie equivalent of the American NSA) recommends the following password complexity:

                a minimum length of 12 alphabetic characters with no complexity requirement; or
                a minimum length of nine characters, consisting of at least three of the following character sets:
                      lowercase alphabetic characters (a-z)
                      uppercase alphabetic characters (A-Z)
                      numeric characters (0-9)
                      special characters.

      That's for "regular" security. For Top Secret, it increases to 15 and 10 characters.

      Source: http://www.dsd.gov.au/infosec/ism/index.htm and in "Filter by section", choose "Access Control, Identification and Authorisation". Directly applies to Australian Government properties, and good reading for everybody else too.

    76. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      sooo 9999-366 is lower hanging fruit then just 366?

    77. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      So, you now have one more bit of entropy than everyone else...

    78. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      This is simple. Follow their rule about the month/day of your mom's birthday, but give a false answer. *YOU* know what you answered, and nobody is going to glean it from Facebook.

      That said, limiting this 4-digit field to 366 total possibilities is rather silly.

      Same for all other "standard security questions" for which someone who knows you could guess the legitimate answer. Give a false one -- only you know what it was.

    79. Re:Hah! Take that, my bank! by roc97007 · · Score: 1

      True. I'm not happy at the fixed number of digits and the requirement that it be numeric only. But one does what one can.

      --
      Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
    80. Re:Hah! Take that, my bank! by hobarrera · · Score: 1

      So I shouldn't use mobile devices outside the privacy of my house?

    81. Re:Hah! Take that, my bank! by Havenwar · · Score: 1

      That's quite a non sequitur there. Whether the device is mobile or not is irrelevant, as is whether you are home or not. What matters is whether people can see you type in your password. Someone shouldersurfing in your living room is just as bad as someone doing it at the office, or in the subway, or in a library.

      We all know you shouldn't log in to a sensitive site from an internet café or similar due to the risk of unfriendly intercepts, and mobile devices have their own security risk (low physical security - they can snatch it while you're logged in and bypass your security all together). This means that if you DO have to log in to a sensitive site from a mobile device it's a security risk to do so around people anyway; totally unrelated to the length of your password.

      I'd recommend excusing yourself and heading to the restroom and doing it there, if you can't just find a gap between people and put yourself with your back to a wall. If you can't deal with this then that's your problem, but a longer password does nothing to protect you from the REAL dangers of logging in at a physically insecure location.

    82. Re:Hah! Take that, my bank! by Hork_Monkey · · Score: 1

      In their defense (sarcasm), Hotmail existed for a number of years prior to being bought out by Microsoft.

      I'll leave another sarcastic defense on why it wasn't fixed since the purchase.

      In all honesty, I was really upset when MS bought them.

    83. Re:Hah! Take that, my bank! by DarwinSurvivor · · Score: 1

      Oh of course, but if they've gotten to the hashes, chances are they've gotten to everything else already anyways. As long as you aren't using that same passphrase on other systems you should be fine.

    84. Re:Hah! Take that, my bank! by Anonymous Coward · · Score: 0

      Like slashd... oh.. wait..

  3. Clearly by Narnie · · Score: 2, Informative

    Somebody hasn't read the relevant xkcd.

    --
    greed@All_Evils:~#
    1. Re:Clearly by Anonymous Coward · · Score: 3, Informative

      http://xkcd.com/936/

    2. Re:Clearly by UnknownSoldier · · Score: 4, Interesting

      I've posted about this in the past http://slashdot.org/comments.pl?sid=3001279&cid=40757735

      > Inconsistent password policies for length, characters and expiry date.

      We _really_ need standards for passwords & passphrases: minimum LENGTH and SYMBOLS included.

      If you site can't handles passwords / passphrases around ~ 96 characters long with the characters (space) 0x20 - 0x7E, your site is *broken*.

      The same crap with usernames. Stop limiting me to a max username length of 12 characters A-Z,a-z because your shitty architect / programmer / DB guy doesn't have a clue about security.

      I propose a multi-tiered system with a schema like:
                  NAME#@%
                  PASS#@%

      Where
          # is the max length allowed * 16
          @ represents which glyphs are allowed to be. Higher is better, which each level including the characters from the previous set
      A = A-Z (0x41-0x5A)
      B = a-z (0x61-0x7A)
      C = 0-9 (0x30-0x39)
      D = space,!-/ (0x20-0x2F)
      E = :-@ (0x3A-0x40)
      F = [-` (0x5B-0x60)
      G = {-~ (0x7B-0x7E)
      % is the number of months the password is valid for.

      Examples:
      NAME1C0 is 16 characters, in range: A-Z,a-z,0-9, 0 = never expires
      PASS6G3 is 6*16 = 96 characters, in range 0x20 .. 0x7E, expires in 3 months

      Then we flame & shame the idiots, er sites, that use crappy username and password polices.

      Maybe time for RFC ?

    3. Re:Clearly by tepples · · Score: 1

      From the transcript: "(You can add a few more bits to account for the fact that this is only one of a few common formats.)" But three more bits for eight common formats pales in comparison to the 16 more bits that switching to a word-salad passphrase buys you.

    4. Re:Clearly by Sarten-X · · Score: 4, Interesting

      You are not the only one, and that's sad.

      While the comic has a pretty poor explanation, the theory is sound: A four-word phrase offers more options (and therefore more protection against a brute-force attack) than an 8-character password, for even a very small dictionary. In fact, the number of options is so drastically larger that it more than compensates for any alterations to the password, like adding numbers or misspelling.

      Of course, the number of options is expanded even further when the password may not actually be a phrase. Maybe the password is really just a 30-digit section of pi. An attacker must try that (and every other number combination) too, so the brute-force strength of a long phrase password is still higher than a shorter random password.

      For a comic, the strip is perfectly valid. A longer (though simpler) password is vastly stronger against brute-force attacks than a shorter one, even though the shorter one looks weirder.

      Note, though, that the strip does not account for attacks other than brute-force, but phrases are still usually better. An attacker physically standing in your office will quickly recognize that a jumbled mess of letters and punctuation taped to your monitor is a password, but an obscure quote attributed to someone he's never heard of is just another office decoration. Even against phishing attacks or plaintext storage hacks, a long phrase is no less secure than a shorter password, since it's not the password that's being attacked.

      --
      You do not have a moral or legal right to do absolutely anything you want.
    5. Re:Clearly by mabhatter654 · · Score: 1

      It works because brute force checking is not going to go 25 characters out. They'll move on. The main weakness at this point is somebody getting the hashed list from a server somewhere. Once they have that it's easy to crack.. That's why youre not supposed to reuse passwords on different sites... Again using the 936 method makes generating them easier.

    6. Re:Clearly by mikkelm · · Score: 1

      That's the real reason to pick a long password, yes, but I have issues with the way the entropy is calculated in that strip. It's based on assumptions, and the entropy seems to be calculated with no regard to any of the many other common permutations.

    7. Re:Clearly by mikkelm · · Score: 1

      That's a rather offensive opening for a post that doesn't really do anything to corroborate the methods used to calculate the entropy. I'm well aware of the exponential benefits of longer passwords.

    8. Re:Clearly by mikkelm · · Score: 1

      I read the transcript. "A few more bits" would actually be a good few more bits. There are many, many more feasible and realistically common permutations. Enough, in my opinion, to bring the total entropy to a level way beyond what's necessary to use on any system that allows a thousand queries per second.

    9. Re:Clearly by Anonymous Coward · · Score: 0

      Yeah that's almost brilliant. I'm working on a project that deals with passwords, and I have the same problem across the board. Inconsistent password policies just drive me nuts. Your idea would almost help, but all sites should have standardize password policies, or at very least a base limit of acceptable, and that is light years beyond 16 alpha numeric.

    10. Re:Clearly by CTU · · Score: 1

      That is only if they know the sites and account names you use. So using the same password for /. as you do for a garden club forums won't really matter more so if one username is cumputergod and the garden one is flowerpower as there is no link to them

    11. Re:Clearly by Anonymous Coward · · Score: 0

      There's one attack that long passphrases are more vulnerable to than a short cryptic password with the same entropy: shoulder-surfing. If I glance at a passphrase, I've memorised it; if I glance at a cryptic password, I just remember that it was a load of gibberish, and maybe a few of the characters. Of course, this is just a side-effect of the fact that long passphrases are easier to remember.

    12. Re:Clearly by houghi · · Score: 2

      Where I work we recently had a password requirement change.
      Programs we use must have 8 character passwords (at least 1 upper/lower case, one number) and that is how the network password was as well.

      They changed it for the network password. It must be at least 10 characters. Contain two numbers, one special character and one upper/lower case.

      Previously you needed to change the password every 40 days or so. They have changed that to 30 days. Now the downside of all this (besides the fact that they did not tell us that this new requirement was in place)

      1) You need two passwords instead of 1. Having it done all automatically will cause a security issue, so no, that is not an option.
      2) People who, like me and the rest of my department, change it every month can't anymore, because months have 31 days and it must be changed after 30.
      3) We still have no idea what the special characters are. I have no idea if there are some that are not allowed.
      4) I have seen people writing them down and I have adviced people to use their 8 character password and add +1 to it.

      T his all means that the system has become LESS secure, but sure, the IT department can tell they did their job, as they only look at the theoretical part and not the practical part.

      Too many IT departments when looking at password security do not vector in the users who are actually will be using it. If you do not do that, it will be less secure. If you do not use all parameters available, your outcome will be wrong.

      It also has increased the out time for people who are waiting for a password reset and therefore the workload for the IT department.

      At another company I once calculated that changing the time a password is valid from 30 to 90 days would save my department two FTE days per month as we were the ones they called and not IT. That was denied. People could not call IT due to language problems.

      --
      Don't fight for your country, if your country does not fight for you.
    13. Re:Clearly by Havenwar · · Score: 1

      Sure. But 99% of users will use a COMMON pattern, which means adding even a few more bits would account for the vast majority if users. It's educating those people to use a better system that would drastically increase the overall web security.

      Think of it like this... if we convince some guy who uses Pin3c0ne34! as a password to use curtains carrot lollipop analbeads instead, then that's not going to make as much of a difference as convincing someone that uses password as a password to use hippo nice boss awful. If you make someone that has a meh password get a great password, 10 out of 10 horrible passwords will still be cracked in the first second, and the majority of hackers won't even spend the computing power go after the meh, because they wouldn't need to.

      If you shift a bunch of the bad password users to good passwords, then the hackers would have to attack the meh to get the same level of results, thus increasing their time investment drastically!

      So sure, if you use an obscure variant of a simple scheme, you're likely quite safe. It's like using a foreign front door lock - no local thief is likely to be able to bump it, because they don't carry blanks for that lock since it's so unusual. But while that might work for your personal security it does nothing for the overall security state of the neighbourhood.

    14. Re:Clearly by Anonymous Coward · · Score: 1

      So... what you're saying is that you're aware the comic is right, but you're arguing because he doesn't properly show his work?

      Let me guess, you're an elementary school teacher?

      He's not a student of your, he's not a scientist, he's a nerd, a geek, a webcomic writer. He's not required to show his work, but he'd probably be more than happy to do so. I wouldn't be surprised if his exact calculations are found in the xkcd forums, or somewhere else on the internet, because he's like that. But he served up the summary and the correct answer in a comic form, easy to digest and easy to understand.

      Because that's the right thing to do.

      Now shoo, you lazy fuck, go and find those calculations for yourself if you're so interested... and if not, then stop arguing just for the sake of it. You bring shame on yourself and your autistic brethren.

    15. Re:Clearly by UnknownSoldier · · Score: 1

      > 2) People who, like me and the rest of my department, change it every month can't anymore, because months have 31 days and it must be changed after 30.

      Agree that is annoying. For this, I use this trick: append the month number as 2 digits on the end of your password.
      i.e.
      Let's sat is Feb which is month 2, then the password would be:
      MySuperDuperPassword02

      That way you should only be at most 1 number off.

      > It also has increased the out time for people who are waiting for a password reset and therefore the workload for the IT department.
      That's another area where people screw up security. If you are going to require frequent password changes, then the turn-around-time MUST be extremely short, else you are just going to frustrate your users.

    16. Re:Clearly by Anonymous Coward · · Score: 0

      Stop limiting me to a max username length of 12 characters A-Z,a-z because your shitty architect / programmer / DB guy doesn't have a clue about security

      Usernames are meant to be public. They show up on your profile page, etc.

      I'd be happier allowing unlimited usernames on my site if I could come up with a way to ensure they'd display properly on every browser with every user's choice of font (eg a CSS "widthOf(string)" option for "width" type styles) then I wouldn't have to worry so much about someone fucking up the display list for everyone just because they've named themselves ||||||||||||(100 more abbreviated for the sake of slashdot's ascii art filter)|||||.

    17. Re:Clearly by mikkelm · · Score: 1

      XKCD is a comic that involves a lot of math and science, and complains about dishonesty in academia on a frequent basis. Yet here the author is being excessively generous to one concept that he favours, and excessively pessimistic regarding another that he's trying to dispel. That's what irks me.

      But if you don't understand the concept, and would rather write a post full of meaningless insults and devoid of reason and sensibility, then there's nothing anyone can do to stop you. Knock yourself out.

    18. Re:Clearly by Anonymous Coward · · Score: 0

      Assuming there are 10000 common words, (10^4), having four of them is only 10^8 combinations, which is far less than the advertised entropy.

    19. Re:Clearly by Anonymous Coward · · Score: 0

      Having a 'password change policy' makes the system less secure - period. Most people are not able to come up with and remember a new difficult password every month. Much better to never change passwords, but have a long complicated one that is hopeless to guess.

      The every-month crowd all have easily-guessed passwords after some time.

    20. Re:Clearly by Sarten-X · · Score: 1

      Four words from a 10^4-word dictionary is (10^4)^4, which is 10,000,000,000,000,000 (or 10^16, not 10^8) possible combinations, or about 53 bits.

      The dictionary used for the comic is only 2000 words.

      --
      You do not have a moral or legal right to do absolutely anything you want.
    21. Re:Clearly by Sarten-X · · Score: 1

      So perhaps do a bit of research on your own? You clearly also don't understand the concept, but you sure are quick to take offense.

      The entropy of a system is simply the number of options for that scheme, expressed in bits. Sure, there are assumptions in the comic but they are applied equally to both sides. The password uses only one particular pattern (and offers a few more bits to account for picking other patterns), while the phrase is restricted to using exactly four words. While the phrase is restricted to a 2000-word (11 bit) dictionary, the password has a 16-bit dictionary to choose from (covering a vocabulary of 65,000 words).

      The comic isn't meant to be an academic paper, but rather to make accessible the results of decades of study into information theory.

      --
      You do not have a moral or legal right to do absolutely anything you want.
    22. Re:Clearly by PuZZleDucK · · Score: 1

      ...Maybe the password is really just a 30-digit section of pi...

      Well, due to Pi being never ending and never repeating, all passwords are actually n-digit sectins of pi (encoded into ascii where n=8*num-chars for example, but you could probably use unicode if you want :p).

      --
      Can a person program a new solution to a problem? Why should anyone be able to stop such a thing? -Richard Stallman
    23. Re:Clearly by bingoUV · · Score: 1

      Have you figured out the proof for this ?

      --
      Bingo Dictionary - Pragmatist, n. A myopic idealist.
    24. Re:Clearly by PuZZleDucK · · Score: 1

      Far from it... my mathematical credentials consist of one term of statistics and a couple of episodes of Numberphile :P

      --
      Can a person program a new solution to a problem? Why should anyone be able to stop such a thing? -Richard Stallman
  4. AOL Used to.... by Anonymous Coward · · Score: 4, Interesting

    Along time ago I had a 10 character password that ended with some numbers for an AOL account. I fumbled the numbers at the end of the password once, aware of such, but hit login anyway and it still let me in. I tested and confirmed it not to care what numbers were at the end of the password. Later it was revealed that AOL was just making a Hash of the first 8 characters of the users password, so it really didn't matter what you entered past the 8th char because it would be trimmed before computing the hash....

    1. Re:AOL Used to.... by _merlin · · Score: 3, Insightful

      UNIX operating systems used to do that, too. This was happening as recently as early releases of OSX. Only eight characters of passwords were significant.

    2. Re:AOL Used to.... by jamstar7 · · Score: 1

      Heh. I modded some BBS software back when I still ran my board on OS9 & a CoCo 3. I had the password database set to 4 bytes: a 3 byte CRC of the passphrase and up to 256 characters in said passphrase. Pretty safe, IIRC.

      --
      Understanding the scope of the problem is the first step on the path to true panic.
    3. Re:AOL Used to.... by Anonymous Coward · · Score: 1

      Windows used LM hash until Windows NT and had it enabled by default until Vista.

      It had 14 byte password limit (zero padded if shorter) - OK, 256^14, or 2^112, should be strong enough...

      Well, nope, ASCII only - OK, 95^14 ~= 2^90 still doesn't sound too bad...

      Nope, we'll also force all letters to uppercase - well, it's 69^14 ~= 2^86, less than initial 112 bit, but still quite good, eh?

      Nope, fuck you, now we'll split those 14 characters in two 7 byte strings and hash them separately without salting, how you like that 2^44 search space?

    4. Re:AOL Used to.... by Cinder6 · · Score: 1

      There was also that Amazon bug a few years ago that made it so capital/lowercase in a password was treated the same.

      http://www.ghacks.net/2011/01/31/amazon-login-may-accept-password-variants/

      They also used to drop anything after the 8th character:

      http://news.softpedia.com/news/Eight-Character-Password-Bug-Identified-on-Amazon-181109.shtml

      --
      If you can't convince them, convict them.
    5. Re:AOL Used to.... by GigsVT · · Score: 1

      Not really. If someone got your DB they could collide (i.e. crack, albiet not necessarily getting the original password, but one that would work for logging in) all the passwords in a trivial amount of time.

      Finding a collision in a 24 bit output space, even with brute force and a fixed length requirement, could probably be done in under a second.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    6. Re:AOL Used to.... by Anonymous Coward · · Score: 0

      With BBS-era hardware? (IE a 486-33mhz with 4MB RAM?) on DOS.

    7. Re:AOL Used to.... by Anonymous Coward · · Score: 1

      CRCs are checksums, not cryptographic hashes.

      Cryptographic hash assumes malicious intentions and should be a) slow enough to limit bruteforcing, b) practically impossible to arbitrarily find a collision.
      Checksum assumes random transmission faults and should be a) fast, b) reasonably preventing random collisions.

      CRC is extremely fast, so you could bruteforce it on a 486. There are also algorithms (can't remember were they developed back then, though) that allow you to create a message with any chosen CRC value.

    8. Re:AOL Used to.... by Anonymous Coward · · Score: 0

      Indeed! This was the first thing I thought of when I read this headline. Maybe they are using a modified NTLM method and just tossed the rest of their (computationally barely relevant) segments? I was horrified to discover how that worked when I used a demo of L0phtcrack to recover an NT password in 1998 or something. Even using the naive brute-force attack on a Pentium II, it didn't take long...

    9. Re:AOL Used to.... by Bert64 · · Score: 2

      It's largely pointless anyway, because the windows networking protocols let you authenticate just using the hash (yes even today with lanman turned off)...

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    10. Re:AOL Used to.... by thogard · · Score: 1

      This predates a 486 by many years. OS9 on a coco 3 ran at about 1.8 MHz and had to bank switch memory since it could address 64k bytes at a time since the 6809 had a 16 bit address bus. Microware's OS could support as many users as it could support serial ports and screens and there were hacks to allow it to have up to 2 mbytes of ram.

      OS9 would CRC check each modules as it was loaded to make sure it wasn't corrupt. It had a list it could check so you could list banned software only only approved modules. It is a feature that that Apples OS X now does as of its latest version nearly 3 decades later.

    11. Re:AOL Used to.... by PCM2 · · Score: 1

      Later it was revealed that AOL was just making a Hash of the first 8 characters of the users password, so it really didn't matter what you entered past the 8th char because it would be trimmed before computing the hash....

      Believe it or not, I've noticed the same thing at Bank of America ATMs. I have a 6-digit PIN on my ATM card (or so I thought). I haven't tried it lately, but the last time I did, I found I could enter some number of digits of the actual PIN (maybe the first 4 or 5), and then after that I could enter any number of additional digits that I wanted. I could enter six digits or ten ... the PIN would always work if I got the first few numbers right.

      --
      Breakfast served all day!
  5. Ummm, nothing new here.... by Anonymous Coward · · Score: 5, Informative

    Umm, TFA says that Hotmail has never accepted passwords longer than 16 characters - it used to silently truncate them. The only thing that's changed is that Hotmail is now letting you know that it's truncating the password.

    1. Re:Ummm, nothing new here.... by Anonymous Coward · · Score: 0

      Umm, TFA says that Hotmail has never accepted passwords longer than 16 characters - it used to silently truncate them. The only thing that's changed is that Hotmail is now letting you know that it's truncating the password.

      Thankyou.
      My immediate thought upon reading the summary was "What, did they always truncate them, or are they storing the actual password instead of a hash?"

      But this does raise the question of why can't you enter a longer password? Do they have a problem with simply truncating your input?

    2. Re:Ummm, nothing new here.... by Anonymous Coward · · Score: 0

      Yes, And it really surprises me that people didn't know about. I thought it was common knowledge that hotmail ignores anything after 16 charters since the beginning.

  6. Huh. by jd · · Score: 5, Informative

    Well, in the Bad Old Days, Unix passwords could only be 8 characters, later extended to 16. Less concerned with the original scheme, more with the fact that Microsoft may be using password algorithms from the 1980s.

    --
    It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
    1. Re:Huh. by Anonymous Coward · · Score: 0

      No idea why but the Linux machines at my college that we use remotely are configured like that. And when you enter a password longer than the original in the ssh it works.

    2. Re:Huh. by eamonman · · Score: 1

      The main solaris server back in college was kind of like that. I had used 9 character PWs for most of college (I figured one more made it safer :P) , and it was only finally during senior year did I notice that that last character didn't matter. In fact, you could type in anything after the first 8 and it still worked (this led to me showing off that I could still log in after mashing the keyboard)

      But yeah, I mean those were the wild wild west days where you telnet-ed in (ssh-ing came around later), ytalk wasn't blocked, and finger wasn't blocked either (hence many girls got creepy ytalk requests from the outside world).

      --
      0- Eamonman Proud member of DNRC
    3. Re:Huh. by Darinbob · · Score: 1

      I definitely ran into a problem with this under an early BSD. Entered a longer password, entered it a second time, they matched, it was accepted. Then I could no longer log in the next day. Reset password and try again. Then the next day same problem. Eventually it was figured out and the admin patched a file and fixed it.

      I think there was also a problem for awhile with user names where additional letters were ignored by some tools.

    4. Re:Huh. by Anonymous Coward · · Score: 0

      Ugh, at my college the Linux labs have still have an 8 character limit to passwords.

    5. Re:Huh. by Anonymous Coward · · Score: 0

      Less concerned with the original scheme, more with the fact that Microsoft may be using password algorithms from the 1980s.

      Wait, shouldn't we be happy that M$ is finally using 80's algorithms?

  7. How were they storing the passwords before? by halexists · · Score: 5, Informative

    RTFA and you learn that they've only been storing the first 16 characters for years, letting you type away in vain. Otherwise they'd have to produce new hashes for the "shorter" passwords that they expect users to use now. (There's no such thing as reading the first 16 digits of a hashed password).

    1. Re:How were they storing the passwords before? by VTI9600 · · Score: 2, Interesting

      From TFA:

      [...] this can only mean one of two things, according to Kaspersky:

              Store full plaintext passwords in their database and then compare the first 16 chars only.
              Calculate the hash only on the first 16 and ignore the rest.

      I’m fairly certain Microsoft isn’t stupid enough to go with the first option. Storing passwords in clear text would be a disaster,

      I wouldn't doubt for a second that MS would go with the first option. They are, after all, competing with Yahoo :-) Also, wasn't it Microsoft that came up with the oxymoronical term "reversible encryption"?

      On the other hand, Hotmail was originally built on FreeBSD by non-MS types, so who knows? To this day I still find it amusing to think of all the difficulty they must have had porting the platform to Windows.

    2. Re:How were they storing the passwords before? by TheNinjaroach · · Score: 1

      If they wanted to convert, they can always wait until the next time you have a successful login. Then they can hash whatever part of the plaintext password you provide.

      Or, maybe they've had a plaintext copy somewhere all along.

      --
      I went to eat some animal crackers and the box said, "Do not eat if seal is broken." I opened the box and sure enough..
    3. Re:How were they storing the passwords before? by Anonymous Coward · · Score: 1

      they do, messenger protocol needs you to send the password in plain text, not the hash

    4. Re:How were they storing the passwords before? by moderatorrater · · Score: 3, Insightful

      That doesn't say anything about how it's stored in the database.

    5. Re:How were they storing the passwords before? by blueg3 · · Score: 4, Insightful

      Also, wasn't it Microsoft that came up with the oxymoronical term "reversible encryption"?

      You're perhaps thinking of hashing. Reversibility is pretty much a requirement for encryption.

    6. Re:How were they storing the passwords before? by Anonymous Coward · · Score: 0

      No.

      http://technet.microsoft.com/en-us/library/cc784581%28v=ws.10%29.aspx

    7. Re:How were they storing the passwords before? by Anonymous Coward · · Score: 0

      That doesn't say anything about how it's stored in the database.

      No, it certainly doesn't. Your observance of the obvious is notable.

      What it does say, is that it no longer fucking matters how it's stored in the database. Now the attacker can go for something easier than the database!

    8. Re:How were they storing the passwords before? by Anonymous Coward · · Score: 0

      Exactly this. The company I work for decided to completely redo how we stored passwords ( for the better ). We gave all of our members 6 months where we used both login schemas ( it would try to log you in using the new, if it failed, it tried the old ). When you first logged in after the change, it would automatically change your password to the new system & remove the hash from the old system.

      After the 6 months, if you logged in with the wrong password, it required you to fill out a form with your username & email address. A new, random, password was emailed to the member. Very simple way of switching methods of storing passwords/hashes.

    9. Re:How were they storing the passwords before? by Anonymous Coward · · Score: 0

      Mod up. I can't believe how many idiots there are in this thread who don't understand this simple fact: you give them the plaintext password every time you login successfully. That's how you seamlessly upgrade from md5 to sha256 or sha512: just wait for the next login.

    10. Re:How were they storing the passwords before? by AK+Marc · · Score: 1

      Also, wasn't it Microsoft that came up with the oxymoronical term "reversible encryption"?

      Oxymoron? Looks more like redundant to me. Encryption should be reversible. Otherwise it is a hash. "Reversible hash" would be an oxymoron.

    11. Re:How were they storing the passwords before? by Anonymous Coward · · Score: 0

      please dont use big words like reversability when you don't know what they mean kthnx late

    12. Re:How were they storing the passwords before? by Anonymous Coward · · Score: 0

      How does that link prove that "reversible encryption" is an oxymoron?

  8. When this happens... by rbprbp · · Score: 2

    Whenever I see any website that rejects passwords longer than X characters, I turn away and go somewhere else. My smallest password those days is 20 characters with numbers and special characters. I expect pretty much any decent website to accept those.

    --
    They're there in their room. You're on your own.
    1. Re:When this happens... by Anonymous Coward · · Score: 1

      I used to have the same approach. However then I joined the shit hole ISP called internode in Australia, for something as important as your ISP password they don't even accept special characters.

    2. Re:When this happens... by notknown86 · · Score: 1

      Why?

      Say the total number of characters upper case + lower case + numbers + special characters is somewhere around 80. And a password is, say, 10 characters on average.

      Is someone really going to issue 10737418240000000000 requests to a publically exposed web server to break your password?

      Or, even in the worst case - they manage to access the password hashes directly and don't need the requests to do it - aren't you basically fucked anyway? If they can access protected areas on a service you trust?

    3. Re:When this happens... by Stiletto · · Score: 5, Insightful

      The question that should be asked is, "What's a 'Special Character' and why shouldn't it be allowed in a password?"

      I had this argument with a developer the other day.

      Him: "What characters should be allowed in this text field?"
      Me: "Um, How about all of them, at least the printable ones."
      Him: "What about special characters?"
      Me: "Give me an example."
      Him: "The ! sign"
      Me: "What's so special about that? I can type it? I use it at the end of some sentences when I'm angry. Why would you not allow it?"
      Him: "What about non-latin characters?"
      Me: "What, are they special too?"
      Him: "You need to specify a list of every character that is allowed in the text field, otherwise I cannot program it."
      Me: [Facepalm]

      etc..

      There doesn't seem to be any compelling security reason to exclude certain characters from eligibility for use in a password.

    4. Re:When this happens... by queazocotal · · Score: 2

      If you are using various devices to login with.
      Sure, your normal keyboard may have an unusual keystroke that yields .
      But, other OSs, or key maps are unlikely to either support the requisite keystroke ('But my phone doesn't even have alt-gr") or it is going to be in a place only findable With extreme effort.

    5. Re:When this happens... by Obfuscant · · Score: 4, Interesting

      Him: "You need to specify a list of every character that is allowed in the text field, otherwise I cannot program it." Me: [Facepalm]

      The developer is right. You are trying to enforce an ambiguous requirement. "All of them, at least the printable ones" is not specific. "Printable" assumes a font. In the symbol font (as found on Winders) there are a lot of "printable" characters that don't show up on a keyboard. Since they are mapped into the same binary values, how do you differentiate?

      "My password has a an "upside down A" but you are accepting a double quote and letting me log in. It's broke!"

      This is not a trivial issue. It appears that someone has had the same kind of conversation with some web developers regarding proper email addresses.

      Him: "What characters should be allowed in an email address?"

      Boss: "Anything that is in an email address."

      Him: "Hmmm, ok, all I've ever seen are A-Za-z0-9.- and one '@'. That's what I'll code.

      Me: "Hey, your website it broken, it doesn't accept valid email addresses! Don't you idiots bother to read the RFC for internet messaging when you program this stuff?"

      Him: "It works fine with my address."

      Me: "It's broken AND HERE IS HOW TO CHANGE THE JAVASCRIPT CODE TO FIX IT."

      Him: "How did you get ahold of our proprietary javascript code?"

      Do you see the problem?

    6. Re:When this happens... by drkstr1 · · Score: 2

      If you do not restrict your input (at least to some extent), you open up the crypto library on your system as a potential attack vector. Granted, its not very likely to come across an exploitable flaw of this nature, but even the best programmers make mistakes. Bottom line, opening up a pipe (completely unrestricted input) to a low level system service is a bad idea. I think a sane limit on password input (maybe 256 - 512 printable latin ascii characters) is reasonable.

      --
      Fanboy Status: Apache Flex, C#, Eclipse, KDE, Pirate Party, Ron Paul, Slackware, Windows 7
    7. Re:When this happens... by Stiletto · · Score: 1

      It's a good point about length. It's probably fine to specify an upper AND lower length limit. Is there actually some standard or best practice out there regarding allowed characters in passwords?

    8. Re:When this happens... by Rob+Kaper · · Score: 1

      The question that should be asked is, "What's a 'Special Character' and why shouldn't it be allowed in a password?"

      I had this argument with a developer the other day.

      Any character that cannot be hashes or escaped before sending it to the storage backend. (Which, agreed, on a modern platform are none.)

      When I started web programming in 1995 some sites had little more than CSV files for storing data. Input filtering then suddenly makes a lot of sense, especially because handy utility methods for hashing and escaping weren't as widely available in all languages as they are now. Any developer still opting for such requirements is obviously an old-timer who hasn't updated his or her skills, or was trained by one.

    9. Re:When this happens... by Rob+Kaper · · Score: 1

      Whenever I see any website that rejects passwords longer than X characters, I turn away and go somewhere else.

      Allow me to nitpick: all websites do for sufficient values of X. Most browsers have a built-in maximum for the length of input fields (32-bit unsigned int for Webkit, 65535 characters) and most webservers have a maximum size configured for HTTP POST requests. ;-)

    10. Re:When this happens... by 2fuf · · Score: 1

      > Me: "Um, How about all of them, at least the printable ones."

      Okay, starting with Chinese...

    11. Re:When this happens... by Anonymous Coward · · Score: 0

      Same thing with GMail, alphabetics and numbers only.

    12. Re:When this happens... by D'Sphitz · · Score: 1

      There doesn't seem to be any compelling security reason to exclude certain characters from eligibility for use in a password.

      I can't see any compelling reason to have any restrictions at all, other than to store people's passwords in plain text or some decryptable format.

      From the hosts perspective, there should be no difference between an 8 character alpha-numeric password, a tab-delimmitted list of non-printable characters, and the entire japanese translation of War and Peace. They all hash to the same length.

    13. Re:When this happens... by blueg3 · · Score: 3, Informative

      you open up the crypto library on your system as a potential attack vector.

      If your crypto library cannot hash an arbitrarily-long string of arbitrary binary data, then it's a very bad crypto library. Or, more likely, you are using it stupidly.

    14. Re:When this happens... by blueg3 · · Score: 2

      Arguably, the designation "printable character" does not imply a font at all. What the symbol font has is glyphs, which map to character code points -- ones that are already defined to mean something different and printable. "Printable" is more a property of the character (code point) and not of the glyph. Fortunately, there's this whole Unicode thing that (a) is a widely-accepted standard with easily-obtained libraries, (b) is supported by web browsers, Windows, Mac OS, etc., (c) will tell you if a character is printable or not.

    15. Re:When this happens... by Immerman · · Score: 1

      Am I missing something? What great harm is caused by your ISP account being hacked? Sure if you use the probably-provided email account attackers will be able to access it, and they'll be able to get 'net access on your dime, but beyond that? They're not going to be able to listen in on your own traffic unless they manage to convince you to connect to them rather than your ISP (in which case they don't care what your password is, they can just accept anything you send as valid), so I'm not really seeing the harm.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    16. Re:When this happens... by Immerman · · Score: 2

      Really that's plenty - uppercase, lowercase, and numbers = 62 states per digit, or roughly 6 bits. To add just one more bit per character would require another 64 characters worth of punctuation and special characters - at that point you're making it dramatically more difficult to remember and type your password without actually increasing the security substantially - adding a couple more characters will be just as secure while being a lot easier to enter.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    17. Re:When this happens... by Immerman · · Score: 1

      Having just argued in another post that upper/lower/numbers was perfectly sufficient from a security standpoint, I'm compelled to agree with you from a convenience perspective. However there are a couple other factors worth considering:

      1) "special" characters are harder to remember and more prone to errors when being entered - allowing them translates into more forgotten/repeatedly mis-entered passwords and the associated increase in worldwide frustration and support costs
      2) many "special" characters can't be readily entered on non-qwerty/non-english keyboards. Allowing region-specific characters will potentially cause serious problems for a user trying to access their account from a computer in another region.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    18. Re:When this happens... by gman003 · · Score: 1

      Him: "You need to specify a list of every character that is allowed in the text field, otherwise I cannot program it."

      Okay, here's your bloody list of all allowed characters. Assuming you use UTF-8, of course. If you use UTF-16, you'll need to add the Supplementary Multilingual Plane, the Supplementary Ideographic Plane, and the Supplementary Special-Purpose Plane.

    19. Re:When this happens... by gman003 · · Score: 3, Informative

      Look at an ASCII table sometime.

      The first 0x20 characters, plus 0x7F, are "non-printable" or "control" characters, having no visual representation in any "standard" font, instead having some effect on the system - NUL, start-of-header, start-of-text, end-of-text, enquiry, acknowledge, bell, backspace, tab, line feed, vertical tab, form feed, carriage return, shift out, shift in, data link escape, device codes 1-4, and a few others I can't remember. The other 0x5F are "printable" - they actually show some character on the screen. That includes everything from space to ~, literally.

      Those are official terms. ISO encodings and Unicode add more printing and non-printing characters, but they all have the same base. And I suppose EBCDIC has its own set of control characters, incompatible with ASCII et al (although if you're basing your password system on "what EBCIDIC allows", you fail on at least a dozen levels already).

    20. Re:When this happens... by Rary · · Score: 1

      This is how the conversation should've ended.

      Him: "You need to specify a list of every character that is allowed in the text field, otherwise I cannot program it."
      You: "Unicode".

      --

      "You cannot simultaneously prevent and prepare for war." -- Albert Einstein

    21. Re:When this happens... by Tumalu · · Score: 1

      When people suggest that non-Latin characters should be allowed in passwords, I sometimes wonder how familiar they are with the common methods for inputting these characters.

      Consider the Hanzi/Kanji used in Chinese and Japanese writing. Because it isn't practical to have a separate key for each possible character, an IME is used to make it possible to type these characters on a standard keyboard. Chinese and Japanese keyboards to generally have a few extra keys for switching between IMEs and/or between modes within an IME, but this doesn't really affect my point.

      Entering a given character often involves a series of key strokes followed by a conversion command (invoked either by a special key or sometimes just with space/enter). The problem with using these characters in passwords is that when it's time to do the conversion, there's usually more than one candidate string to chose from. The order of the candidate strings doesn't even necessarily stay the same between invocations of the conversion command, so if you're entering text into a password field and are unable to see what you're typing, it would become difficult to enter the desired characters.

      Of the GUI toolkits that I've used recently which have a special control for entering passwords, none of them even allow IMEs to operate while entering text into the password field.

    22. Re:When this happens... by Tastecicles · · Score: 1

      I don't see what's ambiguous about "all printable characters". Maybe I'm missing something.

      If it were me, I would specify "every character in the Unicode space".

      Jobbed.

      If your website does not accept Unicode input on your forms, your website is broken.

      Slashdot, I'm looking directly at YOU.

      --
      Operation Guillotine is in effect.
    23. Re:When this happens... by Bert64 · · Score: 1

      I used to have a password with tab and control characters in it, it worked extremely well on the linux console but was pretty much unusable in most gui based environment.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    24. Re:When this happens... by DarwinSurvivor · · Score: 1

      Or switch you to a $300/month internet+tv+phone+telepath plan and stick you with the bill.

    25. Re:When this happens... by DarwinSurvivor · · Score: 1

      That's a valid reason to not USE a password with certain characters, not a valid reason to not SUPPORT them. If a users chooses to put a character in their password they won't be able to type again (like using a computer they won't have further access to, etc), that's their OWN damned fault.

    26. Re:When this happens... by Anonymous Coward · · Score: 0

      Back when euro currency came, I tried to include the euro sign "€" in my password for a linux account.... I never could log in through ssh with it!

    27. Re:When this happens... by karlm · · Score: 1

      Same thing with GMail, alphabetics and numbers only.

      Your post is ambiguous, but it seems you're asserting that GMail does also not allow symbols in passwords. I'll bite. My GMail password contains one or more symbols. Have fun with your 1-bit head start on cracking my 80+ bit GMail login.

      --
      Copyright Violation:"theft, piracy"::Anti-Trust Violation:"thermonuclear price terrorism"<-Overly dramatic language.
    28. Re:When this happens... by drkstr1 · · Score: 1

      You are missing the point entirely.

      --
      Fanboy Status: Apache Flex, C#, Eclipse, KDE, Pirate Party, Ron Paul, Slackware, Windows 7
    29. Re:When this happens... by swilver · · Score: 1

      It should just be any character. I don't care what the character is, accept anything, punctuation, spaces, kanji, hiroglyphs, political incorrect characters, characters that give offense to religious groups, the drawing some 3 year old made... ANYTHING.

      You then just hash those characters... all of them... and store that number.

      I hope I made it clear to the developer.

    30. Re:When this happens... by Anonymous Coward · · Score: 0

      Allow me to nitpick: 32-bit unsigned int would be 4294967295 characters.

    31. Re:When this happens... by bloodhawk · · Score: 1

      They get access to your email accounts, they get the right to create more email accounts under your name. They get the ability with internode to disable certain firewall protections or alter your ADSL profile. They can actually do quite a lot of damage.

    32. Re:When this happens... by Anonymous Coward · · Score: 0

      Despite your low user id, you must be a young whippersnapper. The reason for not allowing "special characters" is that there is not one true character map. Different encodings screw up special characters much more often than the ASCII printable characters. For example, the algorithm which creates the actual WPA key from the passphrase often leads to different results on different systems if you use out of the ordinary characters, but works cross platform if you only use ASCII printable characters.

      There's a human interface version of this problem: Try touch-typing a password that contains a "z" on a QWERTZ or AZERTY keyboard if you're used to QWERTY keyboards (or vice versa). Another similar problem arises if you have to have users type in printed codes (activation, serial numbers, etc.): Choose only one character from these sets to encode information: {0, O, D}, {l, 1, |, 7}, {(, "{", [, <}, {), "}", ], >}. Treat characters from these sets (and other sets of characters which humans can interpret as synonymous) as representatives of the set. Discard " and '. Make your code case-insensitive. Your help desk will not thank you for it, but they won't burn effigies of you like they otherwise would.

      The sane thing to do is not to pretend that a password system can provide a level of security that would warrant an amount of entropy which can't be provided by an eight to sixteen character password from a limited character set.

    33. Re:When this happens... by Anonymous Coward · · Score: 0

      Someone uses your internet connection to download lots of child porn.

      A few weeks later, a SWAT team busts down your door at 5AM, points guns at your family, and you're thrown in jail.

    34. Re:When this happens... by colinrichardday · · Score: 1

      I'd pay $300 a month to be telepathic. Or does "telepath" not mean what I believe it means?

    35. Re:When this happens... by Immerman · · Score: 1

      The problem with this scenario is that actually *using* your 'net connection requires a physical data-link, so the logs would (hopefully) show that when all this child porn was downloaded the data link wasn't actually connected to your home, which gives you at least the beginnings of a pretty solid case, as well as a clue to who the perpetrators actually were.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    36. Re:When this happens... by Obfuscant · · Score: 1

      Look at an ASCII table sometime.

      Ahhh, so the specification has an unstated requirement for "ASCII" encoding.

      Yes, I know what 'ASCII nonprinting" characters are, so thanks for your condescending lecture. I've been doing this for more than thirty years now. I know that you can get nitpicky and talk character coding and glyphs and several levels of abstraction that are irrelevant to the point being made. You can throw about terms like UTF-8 and UTF-16 and ISO and Unicode and whatever, and the point is that none of that was specified. Just saying "all of them, at least the printing ones" is an ambiguous, incomplete specification, and you're making the developer guess what you want.

      A poor developer will he happy using his own standards when you don't care enough to tell him what you want, or don't know enough to know you need to tell him. A good developer will ask, just to avoid surprises and possibly costly code fixes later on.

      A poor developer will ignore the existing standards and provide code to "validate" an email address (or password, or whatever) that excludes many of the valid characters, because he's never seen them used and the boss didn't tell him he needed to do it right. That was my example of this problem.

      The fact that the person who was supposed to be providing the specification did a "facepalm" when asked for a more formal definition than just "all of them, at least the printing ones" shows that there is a bit of ignorance floating about, and it wasn't the developer's.

    37. Re:When this happens... by Stiletto · · Score: 1

      I'll admit that I left out the beginning of the conversation:

      Me: "Users are reporting that we are silently dropping punctuation characters from this text field." ... now try this, go back and re-read the conversation in that light, and notice how that one line changes everything. We now have a developer trying to weasel away from his bug (turns out he added un-asked-for logic to filter out non-alpha-numeric characters), pointing to the spec claiming that the full allowed character set should have been specified if that's what I wanted.

      Having been recently a developer (I'm only just now starting to grow my pointy hair), I definitely have mixed feelings about the "Must have complete 5000 page requirements document in order to write any code" mentality. It's extremely important to have clear, unambiguous requirements for the major software features, important business logic, etc. But some developers take it too far: They can't take a dump without a detailed spec and written use cases. Other developers take a lack of specification to mean "Insert whatever goofy logic I feel like", like the above (otherwise, very talented) developer did in this case. In my documents, I tend to stick to specifying what is required. If I also have to list what is NOT required, you're going to find yourself with a 5000-page document for an iPhone app.

      My intention in the post was not to launch a "1-pager vs. full-specification argument" although it's an interesting discussion. It was merely to point out how ingrained into developers' heads the idea of a "special" character has become, for whatever reason. I've been asked way too often "should we allow 'special' characters here?"

      And yes, I understand the distinction between printable and non-printable becomes less clear when unicode comes into play.

    38. Re:When this happens... by Obfuscant · · Score: 1

      Me: "Users are reporting that we are silently dropping punctuation characters from this text field." ... now try this, go back and re-read the conversation in that light, and notice how that one line changes everything.

      The fact that you think it changes everything shows that you aren't really aware of what it takes to write specifications. If you didn't tell him the first time what you wanted, then you got what he wanted. That you continued the "specification" by expecting him to guess what you wanted again, knowing that it failed the first time, is fascinating.

      In fact, the fact that you were already having a discussion about a failure to specify fully what you wanted and were making him fix what you call "his bug" clarifies why he was telling you he wanted it in writing. Your 'facepalm' should have been "you're right, I didn't tell you exactly what I wanted the first time, here is a list."

      Having been recently a developer (I'm only just now starting to grow my pointy hair), I definitely have mixed feelings about the "Must have complete 5000 page requirements document in order to write any code" mentality.

      Excessive use of hyperbole noted. It doesn't take 5000 pages to give a specific list of what characters you want to accept in a text field. It takes more than "all of them, at least the printing ones", however. For example: you seem to be saying that he can exclude tabs. And linefeeds/CR. Would you go back to him a third time and complain that some of your users were complaining to you that they were trying to enter something in nicely formatted columns and the tabs didn't work? You did say "at least".

      And being asked to clarify one piece of the unwritten specification after it becomes clear that you didn't get what you wanted doesn't mean you have to go back and write a 5000 page spec for the entire project.

      If you care to browse the RFCs sometimes, you will find that they are able to define character lists in about ten or fifteen lines. Given that you could have referred to any RFC, as in "RFC5322 atext and specials" and been precise... no "at least" that leaves things open to interpretation. No "at least" that leaves the developer wondering just what must be there and when you'll be back wanting something he didn't think you wanted.

      It's extremely important to have clear, unambiguous requirements for the major software features, important business logic, etc.

      And your discussion shows that it is also important to have clear, unambiguous requirements for features that have already been programmed in a way that you didn't want and you were expecting to be fixed to meet your ambiguous exacting standards.

      I'd probably do the same thing your developer did, given a lack of a clear definition of what was wanted the first time, and a customer that came back wanting me to fix what I'd done based on the lack of specification, who couldn't be bothered to be specific the second time.

    39. Re:When this happens... by Stiletto · · Score: 1

      I guess we're going to have to agree to disagree then. To me, requirements document what the program is and does, not what it isn't or doesn't do. It makes as much sense to write in a spec "Do not put any character constraints on this text field," as it does to write "Do not display a purple elephant in this window."

    40. Re:When this happens... by DarwinSurvivor · · Score: 1

      So would I if there wasn't a monthly limit of 3 thoughts and a $400/thought overage fee.

    41. Re:When this happens... by Anonymous Coward · · Score: 0

      Anecdotal point. I support a certain, fairly well known statistical package on Windows. A couple of years ago our IT guys added complexity requirements to the domain password rules, which increased the likelihood of symbol characters being used in passwords.

      As a result, we found that some users were suddenly unable to authenticate to the server upon which the statistical app was running in certain cases. These were related to some symbol characters being used in passwords - at least one of the slashes, and either the single- or double-quote were affected.

      Now I'm not entirely certain what the root cause of the problem is, but it seems related to inaccurate escaping of password characters which results in the pass (or at least the derived hash) being mangled in some manner.

  9. Let's give this a shot by Ol+Biscuitbarrel · · Score: 3, Funny

    hunte

    1. Re:Let's give this a shot by Anonymous Coward · · Score: 0

      Error: ***** is unacceptable. Password must contain both letters and numbers.

    2. Re:Let's give this a shot by steelfood · · Score: 1

      Looks good. Shows up as ***** on my screen.

      --
      "If a nation expects to be ignorant and free in a state of civilization, it expects what never was and never will be."
  10. more crackable by harvey+the+nerd · · Score: 0

    Presumably someone from the NSA or IRS wants to know...

    1. Re:more crackable by Pinhedd · · Score: 3, Insightful

      Most website authentication systems use a hash to store passwords. The unhashed string is formed from a salt, some unchanging record information (such as the user's username, or date of registration), and the user's plaintext password. During the hashing process, all of this gets distilled down to a fixed length string regardless of the complexity of the password. Thus, a lengthy password is not necessarily more secure than a short but sufficiently complex password. Any site worth their salt (pun intended) will lock an account after a number of failed logins anyway. The majority of compromised accounts come from successful phishing and social engineering, not from randomly guessing passwords. Now, encryption on the other hand should use a very strong and long password.

    2. Re:more crackable by Anonymous Coward · · Score: 0

      Presumably someone from the NSA or IRS wants to know...

      Presumably the NSA or IRS is full of dumbshits then if that is their requirement.

      Give it another 20 years...at that point you'll probably hear one of them say "Rainbow tables? No, never heard of them...they must be new, right?"

    3. Re:more crackable by osu-neko · · Score: 3, Insightful

      Any site worth their salt (pun intended) will lock an account after a number of failed logins anyway. The majority of compromised accounts come from successful phishing and social engineering, not from randomly guessing passwords.

      That second fact is the reason why your first sentence is incorrect. Locking an account after a certain number of failed login attempts introduces a kind of denial of service attack on the site (at least, denying that particular user access) while not actually stopping any feasible attack vector. It's the kind of security flaw you see implemented by coders that don't really understand security. Preventing too many attempts in too short a time is a security feature. Locking an account after too many attempts is a security flaw. You might as well just give hackers an input field where they can type in the name of any legitimate user they want to lock out of a system illegitimately.

      --
      "Convictions are more dangerous enemies of truth than lies."
    4. Re:more crackable by Immerman · · Score: 1

      True, but there are lots of situations where a DOS attack is largely pointless, while a brute-force attack offers significant benefits. Online banking for example - hack my account and you can steal my money, an appealing target for any thief - but only a petty ass is going to try to lock me out of my account, especially since a quick call to the bank will fix the problem for me. Handled properly there's even some good PR in it for the bank "There was a hacking attempt on your account, but we stopped it. Sorry for the minor inconvenience it caused you" - people like knowing the folks guarding their money are actually paying attention.

      I say this as a guy that ran some university gateway servers for several years - short-term lockouts radically reduced the number of hacking attempts, and despite the higher than average density of petty ass-holery (hey, it's college) I only had a handful of complaints in all that time.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
  11. Hash???? by NinePenny · · Score: 2

    Hmm... Why wouldn't they just store a 16 char hash of whatever password you want?

    Usually you only see this when someone is doing something wrong from a security standpoint.

    1. Re:Hash???? by tepples · · Score: 0, Offtopic

      Because hash has been illegal since 1937, except for a 17-month period in 1969-1970.

  12. So? by jd2112 · · Score: 5, Insightful

    Who in their right mind would trust anything sensitive enough to require a 16 character password to Hotmail?

    --
    Any insufficiently advanced magic is indistinguishable from technology.
    1. Re:So? by Anonymous Coward · · Score: 0

      My web site is super secure. I accept 4096 character passwords. I then store all passwords as a hash.

      It's amazing how many people use the same password!

      Oh, it's a 1-byte hash to save storage space. Does that matter?

    2. Re:So? by Anonymous Coward · · Score: 0

      Pretty much every single one of my non-tech friends. They're also the ones who's entire list of bookmarks is Google, Hotmail, Facebook, YouTube, and maybe The Chive. Anything outside of that is scary and they refuse to learn.

    3. Re:So? by shutdown+-p+now · · Score: 1

      It becomes more important when you get all those single sign-ons within the ecosystem. Like Google password securing your email, IM, social network, cloud storage, and a dozen other things. MS accounts are not quite the same yet, but it'll be more pronounced for people who use Win8 and use their MS account to log in (which you're not required to do, but the system certainly nudges you there).

      That said, 16 case-sensitive alphanumeric characters is still 62^16 combinations - that's, what, 96 bits?

    4. Re:So? by Anonymous Coward · · Score: 0

      Maybe people find it easier to make an easy-to-remember medium-strength password if it's longer. For example, a series of common words rather than a sequence of random characters. (The usual reference.)

    5. Re:So? by Anonymous Coward · · Score: 0

      Some people use passphrases retard.

      Passphrases can easily be superior to passwords, *unless* the system truncates them for absolutely no rational reason.

    6. Re:So? by Anonymous Coward · · Score: 0

      Just a random point.. I believe it's the US website to get a passport (the main site where the 3 different types of "passports" can be selected... I believe it was that site... requires your password to be between two set lengths, if I recall correctly letters and numbers only, and your password *must* start with a number.
      I apologize if it wasn't their site that has this limitation, I'll have to check my email and see who I emailed the letter to pitching about how reduced the security was from requiring a password like this

    7. Re:So? by PCM2 · · Score: 1

      Who in their right mind would trust anything sensitive enough to require a 16 character password to Hotmail?

      It's not just Hotmail. To login to Hotmail these days, you use your Microsoft account. So it applies to anything you use a Microsoft account for, including Hotmail (or Outlook.com), SkyDrive storage, Office Web apps, calendaring, etc.

      Also, some people like to use long password not just because they're more secure, but because they're easier to remember (for example, "My dog has fleas" is easier to remember than "2MnG7_0z").

      --
      Breakfast served all day!
  13. dario90 by Anonymous Coward · · Score: 0

    Security -- the microsoft fail of all life

  14. isn't this backwards by lc_overlord · · Score: 1

    I though you where supposed to enforce longer passwords instead
    The math is clear, if a 8 character alphanumeric password takes a second to break then a 20 char password takes about 15.000.000.000.000 years to crack or 110 times the age of the universe.

    --
    - "There is nothing quite like an ineffective solution to an nonexistant problem"
    1. Re:isn't this backwards by Obfuscant · · Score: 1

      The math is clear, if a 8 character alphanumeric password takes a second to break

      If you have software/hardware that can brute force an 8 character alphanumeric password in one second, the NSA would love to talk to you. Probably already is. It was nice knowing you, hope it turns out well.

    2. Re:isn't this backwards by hobarrera · · Score: 1

      I'm pretty sure we'll be able to crack 20 char passwords in a matter of seconds in less than 1k years.

    3. Re:isn't this backwards by Tastecicles · · Score: 1

      that's only 268 million tries (assuming 28-bit entropy). Entirely possible with only one Pentium-class core. In fact, my forensic box, for several years, was a Pentium Pro with two processors - huge overkill but it gave me a very quick system with which I could recover data - if it still worked I'd still be using it. The average Windows password fell before that in way less than two minutes.

      --
      Operation Guillotine is in effect.
    4. Re:isn't this backwards by DarwinSurvivor · · Score: 1

      And hopefully you'll have changed your password by then.

    5. Re:isn't this backwards by Obfuscant · · Score: 1

      that's only 268 million tries (assuming 28-bit entropy).

      There are 26 characters a-z, 26 more A-Z. My 108 key standard keyboard has another 42 characters (0-9 and symbols). That's 94 total. 94^8 is 6,095,689,385,410,816. (I think that's something like 12 BILLION times the number you gave.)

      You can make assumptions about what a password is likely to contain, and if you run into morons who enforce rules about what you must have and cannot have in your passwords, you can save some tries, but a generic brute-force password cracker will require the use of all available characters. Oops, I forgot "space". That's 95^8.

    6. Re:isn't this backwards by Tastecicles · · Score: 1

      yes, I pulled a number out of my arse, but you get the point I was trying to make (such as evidenced by your complete lack of an attempt to argue against the rest of my post. Thank you).

      --
      Operation Guillotine is in effect.
    7. Re:isn't this backwards by Obfuscant · · Score: 1

      yes, I pulled a number out of my arse, but you get the point I was trying to make (such as evidenced by your complete lack of an attempt to argue against the rest of my post. Thank you).

      Yes, I failed to argue against the rest of your post because I recognized a number pulled out of your ass and knew that you weren't trying to have a serious discussion. The fact you missed was that your number was a factor of about 24 million too small, which you then used to argue that it could be done in a very short amount of time. Wow, you can do 268 million in two seconds. Now try the real number, which would be done, using your super fast, l33t hardware, in 286 days.

      As I said, if you have hardware that can brute force an 8 character password in two seconds, the NSA would love to talk to you. Whether that hardware might someday exist is irrelevant, we're talking about today. Present tense.

  15. Re:You need more than 16 char by sexconker · · Score: 0

    If you're protecting against Sky-Net SAC-NORAD missile launches I can see it, otherwise it's overkill.

    Unfortunately you need a lot more when people listen to the terrible advice given out by terrible comics and start using passphrases consisting entirely of dictionary words. "incorrect equine cell affixer" becomes "incorrect equine" when truncated to 16 characters.

  16. Why have such short limits? by Paradigm_Complex · · Score: 5, Interesting

    As fun as it is to bash Microsoft, they're not the only ones who do this. Presumably there is some technical reason why this is done, but I am at a loss for what this would be. Would someone be able to explain to me the reason why such limits are put in place?

    It seems with modern computer capability that absurdly long passwords would be trivial. The hashed password length would be the same irrelevant, so I can't see storage space as the issue. The only other idea which comes to my mind is the computational difficulty of hashing the passwords, but even that has to be trivial by today's standards, even with millions of users hitting the servers. Why not go overboard and just allow several kilobytes worth of password?

    --
    "A witty saying proves nothing." - Voltaire
    1. Re:Why have such short limits? by Anonymous Coward · · Score: 1

      There is no point storing a password that is longer than the hash as there is then multiple string that resolve to the same hash.

    2. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      You are assuming they hash the passwords at all. I hope they do, but wouldn't be that surprised if they didn't.

    3. Re:Why have such short limits? by Volanin · · Score: 2

      Commenting here, as my finger slipped and I wrongly modded this as Troll. Gosh, I am a human, give me an option to remoderate my miskates, you silly slashdot!

      --
      If I clone myself, can I call it a thread?
      If a girl winks to us, can I call it a race condition?
    4. Re:Why have such short limits? by Anonymous Coward · · Score: 1

      It makes even less sense when you consider that they should never be storing the actual password anyways, but instead they should be storing a salted hash of it. The original password length is irrelevant when they are only storing the hash, since all hashes will be the same lenght.

    5. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      I'm guessing they use MS SQL server in their backend...

    6. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      I believe it's legacy reasons. A long time ago, the data was read into a 16-byte buffer. Someone made the decision to truncate passwords instead of disallowing them. If they were to remedy the problem and expand the max allowed password size, everyone with an old "greater than 16 character" password would lose access to their accounts

    7. Re:Why have such short limits? by bertok · · Score: 5, Interesting

      Every time I see any kind of password length limit somewhere, I instinctively know that somewhere behind the scenes there is this table column:

          user_password VARCHAR(16) NOT NULL

      It's the same sinking feeling I get when I see the "the following special characters cannot be used in the password field" error message, which just tells me immediately that the code that submits the password field looks like:

          $cmd = "UPDATE ... user_password='" + $password + "' ... "

      There really, really needs to be a "guild of programmers" or somesuch, along the lines of the Bar Association, so that anybody who writes code like the above can be summarily ejected from it.

    8. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      NVARCHAR(16)

    9. Re:Why have such short limits? by John+Bokma · · Score: 1

      I am afraid that would eject a large number of programmers (aka "programmers"). Wouldn't surprise me if this would eject 80%. But I am all for it.

    10. Re:Why have such short limits? by Darinbob · · Score: 1

      Because people don't think. That is all.

      Seriously, the people who design the web UI front end and the people who save the data to the databases on the back end are not security experts. They probably just thought "ok, gotta reserve space for a field, I think 16 characters is enough" and that was the end of the thinking process.

    11. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      You clearly have no idea what a hash actually is

    12. Re:Why have such short limits? by kbolino · · Score: 2

      Most sites (don't know about hotmail for sure) only let you enter alphanumeric characters. That's 26 (lower case) + 26 (upper case) + 10 (digits) = 62 characters, or about 6 bits of entropy per 8-bit byte.

      For a 160-bit hash like SHA-1, that's 26 characters.
      For a 256-bit hash like SHA-256 or GOST, that's 42 characters.
      For a 512-bit hash like SHA-512 or Whirlpool, that's 85 characters.

      No secure hash (by the current standards) would limit the password space to 16 characters (that would be a 96-bit hash, roughly).

    13. Re:Why have such short limits? by jrumney · · Score: 1

      As fun as it is to bash Microsoft, they're not the only ones who do this. Presumably there is some technical reason why this is done

      Probably something like the password is stored in a CHAR(16) column in their database. If they were to accept longer passwords, they'd have to do something radical like hash them before storing in their database.

      Personally, if I had a hotmail account, I'd be closing it and changing any shared password elsewhere - not because they're limiting the password length, but because they are technically able to log you in with only a part of your password, which proves they have been storing it without hashing all along.

    14. Re:Why have such short limits? by honkycat · · Score: 1

      The behavior, at least for me, was changed without warning or without any action on my part.

      I don't care nearly enough to bother looking for an option to correct this awful, awful user-hostile default.

    15. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      Just wait for 1024bit hashes! Assuming it wins the SHA3 contest.
      http://en.wikipedia.org/wiki/Skein_(hash_function)

    16. Re:Why have such short limits? by D'Sphitz · · Score: 1

      I assume he meant he intended to mod the ggparent up, but hit troll instead. There is no confirmation, the mod is applied immediately. If this is a setting somewhere, I don't remember ever seeing or setting it.

    17. Re:Why have such short limits? by D'Sphitz · · Score: 1

      which proves they have been storing it without hashing all along.

      Or they are truncating it before hashing, which is s still a WTF.

    18. Re:Why have such short limits? by ThatsMyNick · · Score: 1

      You mean people store passwords as plain text without hashing (not to mention salting the hashes)? Will people ever learn?

    19. Re:Why have such short limits? by xlsior · · Score: 2

      Under Windows NT4 Microsoft's NTLM password database would only store the first 14 characters of each password, silently ignoring everything else. Worse, they'd actually split it into two seven-digit chunks and store them separately, without applying a SALT. The result of this was that having a password that's a few characters longer than 7 digits would actually be worse than having a short 7-digit password, since those couple of extra characters would be trivial to brute-force and could potentially give clues about what the first part of the password could be.
      On top of that, LM backwards compatibility further dramatically reduced the key space .

    20. Re:Why have such short limits? by gman003 · · Score: 1

      Can we eject people for partial failure?

      The project I'm working on has a username/password system. I, and the rest of my company, are doing the back-end; another company is doing the front-end, all for a third company. Naturally, we have to work pretty closely together, but one of the guys on the front end is not the best programmer.

      I'm a stickler for password strength (I have a 20+ character root password, with enough symbols and numbers to keep anyone from ever breaking it), so I tested the system with insanely long and complex passwords (it breaks if you use characters outside the Unicode basic multilingual plane, but only because we're setting it as UTF-8). Hashed with some advanced algorithm, Twofsh I think (I wrote it months ago).

      I tested the front-end, and it actually worked OK with long passwords. But the front-end guy had thrown in one of those Javascript "how strong is this password?" widgets, and guess what? If you enter more than 63 characters, it immediately shows it as "very weak". Doesn't break the site, and anyone using a 60-character password is smart enough to not listen to that crap, but it is a little bug.

      I'd demand he fix it, but the poor guy's far enough behind as it is.

    21. Re:Why have such short limits? by SecurityGuy · · Score: 1

      I'm not afraid. I'm hopeful it would eject 80%.

    22. Re:Why have such short limits? by shutdown+-p+now · · Score: 1

      It's more likely that there was such a thing there somewhere years ago, and then it was converted to hashes - but, of course, since it only ever stored 16 chars, that's what was hashed, and the system had to keep trimming the incoming passwords to make sure that all the old ones keep working.

    23. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      There's not necessarily a technical reason. Some managers just have an idea of what password lengths are supposed to be like, and force it on the programmers for no good reason.

      I worked for a company where that happened. It was a requirement pushed down by management that passwords for a new application I was developing must have lengths within a small range, I think 6 to 12. I was directly responsible for implementing the password storage and there was no absolutely technical reason for limiting the length, because everything hashes to the same length no matter what. I pushed for a longer length and they settled on something like 16 or 18.

      But I designed the back-end code and password storage to accommodate arbitrary password lengths. The web guys enforced the limitation only in the UI, so if and when they ever decide to increase the length it won't require any back-end code changes.

    24. Re:Why have such short limits? by Billly+Gates · · Score: 1

      Yeah no kidding. Here is how I handle passwords from a real professional like myself

      $query_login="select * FROM user";
      $result_login = mysql_query($query_login) or die("Your passwrod is might be bad I think"); //$login_check = mysql_num_rows($result_login);
      while($row=mysql_fetch_array($result_login))
      {
      $username=$row["username"];
      if ($username==$username1)
      {
      echo "";
      echo "window.location.href='login_error.php?rec=qq';";
      echo "";
      exit;
      }
      }

    25. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      A guild of programmers would be excellent, it would also make programmers directly liable, however therein lies the problem, nobody wants to pay programmers massive fees so that they can pay their liability insurance.. Ala doctors, engineers and lawyers.

    26. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      95.26714096619001 bits, specifically.

    27. Re:Why have such short limits? by Rockoon · · Score: 1

      It seems with modern computer capability that absurdly long passwords would be trivial.

      We are not living in an age where servers are growing in power due to CPU enhancements. We are living in an age where many servers now share a single CPU due to CPU enhancements.

      Its not uncommon these days for a single rack to be powering dozens, or even hundreds of servers.

      --
      "His name was James Damore."
    28. Re:Why have such short limits? by Anonymous Coward · · Score: 0

      What fancy newfangled database are you using that has VARCHAR? CHAR and trimming whitespace when reading back should be enough!

    29. Re:Why have such short limits? by dstar · · Score: 1

      I am afraid that would eject a large number of programmers (aka "programmers"). Wouldn't surprise me if this would eject 80%.

      Hmm.

      There's probably a downside somewhere, but I'm not seeing it.

    30. Re:Why have such short limits? by Obfuscant · · Score: 1

      There is no confirmation, the mod is applied immediately.

      Only if you have your settings set to do it that way.

      If you don't like the way your account is set to do this, then change it. For the couple of people who modded my first comment down, telling people I don't understand why they don't change the setting they can to fix what they are admitting is a problem is not flamebait. It's common sense.

      If you don't care that you have to post to undo a moderation that you didn't want to make and didn't feel like changing the settings for your own account so it wouldn't be a problem anymore, and thus lose all your other moderations in that discussion and waste what you had, fine. This "instant setting" happened to me exactly once. Then I changed the setting. I didn't even consider leaving it set that way and then complaining in public every time I made a mistake in moderating later.

      As I recall, it is under "Discussions", "Classic Discussion System (D1)". Yeah, there's probably a lot of combined settings there, but you can fix the problem of having to post an inane comment just to undo an errant moderation if you want to.

  17. So What? by Anonymous Coward · · Score: 0

    Quit using Hotmail. Problem solved.

  18. could be worse by Anonymous Coward · · Score: 0

    Some very major sites are even more egregious -- take for example American Express, which limits passwords to 8 letters and numbers only, no special characters allowed. Even a decade ago that's like calling for a bodyguard and being sent an 8-year-old boy with a slingshot. Every month I think about closing that account for that reason alone.

  19. Shortens password? by nitehawk214 · · Score: 1

    Does this mean they were storing the passwords in cleartext? In a real system they would simply be storing the hashes, shortening the password would cause it to create a different hash and not match.

    --
    I'm a good cook. I'm a fantastic eater. - Steven Brust
    1. Re:Shortens password? by tepples · · Score: 1

      Does this mean they were storing the passwords in cleartext? In a real system they would simply be storing the hashes

      I'm under the impression that they hash it after chopping off everything after 16 characters. Perhaps it's easiest to express in PHP: $hash = sha1(substr($password, 0, 16));

    2. Re:Shortens password? by Rob+Kaper · · Score: 1

      Does this mean they were storing the passwords in cleartext? In a real system they would simply be storing the hashes, shortening the password would cause it to create a different hash and not match.

      Not necessarily. The UNIX crypt(3) algorithm uses only the first 8 characters of any password. Given Hotmail's age I'm sure something similar is going on here. Not every website was developed in an era of HMAC-SHA-512 with proper salt and pepper flavouring.

      It would however be possible for them to upgrade passwords upon login (in which case the unhashed match would be available from input), but for a system the size of Hotmail it would take forever before the legacy support could be deprecated.

  20. Rainbow tables by Anonymous Coward · · Score: 0

    This allows your password to be revealed with minimal computing time. Sounds more like it is to assist law enforcement, than end users. Anyone choosing a password over 16 characters, obviously didn't want the help in the first place.

    1. Re:Rainbow tables by Kaz+Kylheku · · Score: 1

      Rainbow tables are used for attacking hashed passwords. They are a reverse dictionary of hashes back to plaintext passwords.

      These hotmail passwords are obviously stored in clear text, otherwise how would they be able to just chop them at 16 characters?

      (The only way would be if they had anticipated such a change way back when the hashes were generated, and they generated a 16 character hash along side a full hash, and so now they are just switching which hash they use.)

      So if you gain access to Hotmail's password storage, you don't need any tables, you just read the passwords.

      If you don't have access to the password storage, then having rainbow tables is moot. You're reduced to making login attempts by brute force.

    2. Re:Rainbow tables by djmurdoch · · Score: 2

      (The only way would be if they had anticipated such a change way back when the hashes were generated, and they generated a 16 character hash along side a full hash, and so now they are just switching which hash they use.)

      That's not the only way. Another way would be that they silently dropped all characters after the 16th, then formed a hash from what was left.

    3. Re:Rainbow tables by Anonymous Coward · · Score: 0

      > That's not the only way. Another way would be that they silently dropped all characters after the 16th, then formed a hash from what was left.

      But in order to do that, they had to have the original password in cleartext.

    4. Re:Rainbow tables by Anonymous Coward · · Score: 0

      Yes but the argument is they would have needed to do that from the outset. There is no way to turn an already-hashed password into a hashed password of the shorter version. Take this simple example:
      password is set to "password". The hash (md5sum) is 286755fad04869ca523320acce0dc6a4, and the word "password" is forgotton.

      New policy, passwords are now only 4 characters. We need to figure out how to change 286755fad04869ca523320acce0dc6a4 into 4528e6a7bb9341c36c425faf40ef32c3, but the only way to come up with 4528e6a7bb9341c36c425faf40ef32c3 is to know that the first 4 characters were "pass" - which we don't know at this point in time (since we only saved 286755fad04869ca523320acce0dc6a4).

      The only way to actually do this, would be as each user logs in successfully, take a new hash and store it along side the old hash, because at login time we have the plaintext and could re-hash the first N characters.

    5. Re:Rainbow tables by farble1670 · · Score: 1

      These hotmail passwords are obviously stored in clear text, otherwise how would they be able to just chop them at 16 characters?

      they've always been chopping them. they are chopped before they go into the hash function. they haven't changed anything, they are just now letting you know what they are doing.

    6. Re:Rainbow tables by Anonymous Coward · · Score: 0

      These hotmail passwords are obviously stored in clear text, otherwise how would they be able to just chop them at 16 characters?

      By truncating the password before the hash is calculated. obviously. This in no way indicates the passwords have been stored as plain text, it's just now they are admitting they've been truncated all along.

    7. Re:Rainbow tables by Anonymous Coward · · Score: 0

      They used to accept longer than 16char passwords. If you had a password longer than 16chars, you now could only enter the first 16.

      1) This means they did store more than 16chars of data
      2) If it was stored as a hash, there is no way to truncating the original text without first knowing the original text.

    8. Re:Rainbow tables by djmurdoch · · Score: 1

      Yes, they probably did that from the outset. Other posters have said that's exactly what they did; I haven't checked, but it seems like a credible explanation, and doesn't require cleartext passwords to be stored.

  21. Bank website... by Anonymous Coward · · Score: 0

    A banking website I used silently dropped special characters, perhaps to prevent injection attacks on their form. Reduced you to letter and numbers only.

    1. Re:Bank website... by True+Vox · · Score: 1

      Wow. That's one way to prevent that. I mean, it would have worked here, but I'm not sure it's the BEST way...

      --
      "Gratuitous complexity is akin to chaos" - True Vox
    2. Re:Bank website... by Tastecicles · · Score: 1

      that is just feakin' hilarious!

      --
      Operation Guillotine is in effect.
  22. PASSWORDS NOT HASHED?!? by Anonymous Coward · · Score: 0

    So... it means Microsoft is not hashing passwords at all, because hashed passwords cannot be truncated (well, at least not without the user entering them the FULL password after the truncation system has been put in place)

    Wow...

    1. Re:PASSWORDS NOT HASHED?!? by farble1670 · · Score: 1

      it could mean that. but it could also mean they've always been truncating them from day one, and they are now just letting you know.

    2. Re:PASSWORDS NOT HASHED?!? by Anonymous Coward · · Score: 0

      They clearly stated that they were not truncating them and passwords greater than 16 chars worked as expected prior to this change.

  23. Ummm, no? by Anonymous Coward · · Score: 0

    I think this has been the case for many months, if not years. I don't think I'm mistaken, but I may be. I think if you entered a long password, only the first 16 characters were necessary to log into your account. Please correct me if I'm wrong.

  24. As opposed to... by tepples · · Score: 4, Interesting

    As opposed to the sign-up page at Phil's Hobby Shop, which pretty much advertises that it's 936-compliant.

    1. Re:As opposed to... by Anonymous Coward · · Score: 0

      Phil? Is that you?

    2. Re:As opposed to... by Anonymous Coward · · Score: 0

      Phil? Phil Connors?

    3. Re:As opposed to... by Anonymous Coward · · Score: 0

      As opposed to the sign-up page at Phil's Hobby Shop, which pretty much advertises that it's 936-compliant.

      What the fuck does it mean to be "compliant" with a comic strip? You just get more and more retarded with every post, don't you?

  25. Slashdot has a limit to by Robadob · · Score: 1

    Slashdot has a password length limit, iirc its 20. The input field for setting a password has a max length of 20 however the login field doesn't. So when i last changed my password i was confused for a short while till i realised that i hadn't read the password guidelines. To be honest i find that ~50% of websites that i try to use long passwords on are limited to around 20.

    1. Re:Slashdot has a limit to by Robadob · · Score: 2
      On the topic of weird password requirements, my university has the weirdest password requirements i have come across to date (i'm assuming its due to some software they must use);

      Note that passwords must follow these rules:
      * must be 6, 7 or 8 characters in length
      * must contain at least one numeric digit
      * must NOT start with a numeric digit
      * must contain only lower-case alphabetic letters and numeric digits (that is no punctuation characters).
      * the first three characters of your password must not be identical
      * the first three characters of your password must not equal sap or pass
      * the first three characters of your password and login name must not be the same

    2. Re:Slashdot has a limit to by jrumney · · Score: 2

      the first three characters of your password must not equal sap or pass

      Hmmm, I wonder what other opportunities for SQL injection there are that they don't have covered.

    3. Re:Slashdot has a limit to by Anonymous Coward · · Score: 0

      Yikes! Only up to 8 lower-case letters and numbers?! That's only (26 + 10) = 36 possibilities per character, for a total of pow(36, 8) = 2,821,109,907,456 passwords. With today's CPUs (let alone GPUs), finding the plain-text form of such a password would probably take less than an hour (assuming a 3 GHz CPU, and less than 3600 clock cycles to generate the hash for a given plain-text candidate password).

      But, if that university system uses a well-known hash algorithm for its password database, then it would be quite practical for people to crack the entire list of passwords -- past, present, and future -- instantly, by using a few 3 TiB hard drives ($100 each) to store the hash codes for all 2,821,109,907,456 passwords. For example, suppose the password hashed to 16 bytes. Then 16 x 3 TiB hard drives would be sufficient to store the 16-byte hash codes for all 2,821,109,907,456 passwords. These hash codes would be sorted by 16-byte values, and so the conversion of 16-byte hash code to plain-text password would only take O(log2(N)) = O(log2(2,821,109,907,456)) = 42 disk reads (scattered across the 16 drives, as the binary search converges on the relevant record). $1600 is less than the cost of many laptop and desktop computers! Even a person on a budget could do this!

    4. Re:Slashdot has a limit to by Anonymous Coward · · Score: 0

      If your password requirements take longer than a line to write ("You password must be at least X long."), you are doing it wrong.

    5. Re:Slashdot has a limit to by Tastecicles · · Score: 1

      OK, so my password must be the same as my AIM username?

      --
      Operation Guillotine is in effect.
    6. Re:Slashdot has a limit to by Anonymous Coward · · Score: 0

      6-8 characters makes them ridiculously insecure in the first place, the other requirements are all stupid, especially the restriction to lower-case letters.
      Maximum of probably about 40 bits of entropy is pathetic.

    7. Re:Slashdot has a limit to by AK+Marc · · Score: 3, Interesting

      Often it's due to single sign on, and it has to work on 100 different legacy systems. But yes, I've been told one place to set my password to 6 alpha characters and 2 numbers. I have no idea what would or wouldn't work, and the combinations I tried matching the published rules didn't work, but adopting the 6+2 scheme generated a valid password. Because of the 60 day pwd expiration, and no repeats in the last 12 passwords, everyone does 6 alpha in lower case, and numbers 1-12, rotated. It is roughly as secure as just 6 chars and no numbers, but no, we have to have the numbers and the short-ish expiration. I am of the opinion that expiring passwords leads to less password re-use, but less password entropy as well. For where that tradeoff gets us for security, I have no idea.

    8. Re:Slashdot has a limit to by colinrichardday · · Score: 1

      * the first three characters of your password must not equal sap or pass

      How can the first three characters be "pass"?

  26. Fun with passwords by WinstonWolfIT · · Score: 1

    Huh. Filter error: That's an awful long string of letters there. So spaces added.

    Things one might never tire of hearing:

    Ohmyitssolarge!
    Itwasthebestoftimes Itwastheworstoftimes
    Franklymydear Idontgiveadamn
    Youplayeditforher youcanplayitforme
    OnthewholeIwould ratherbeinPhiladelphia

    Also, some obligatory links for your benefit:
    http://xkcd.com/936/
    http://xkcd.com/792/

  27. Banks just as bad by Asmor · · Score: 3, Interesting

    TD Bank, my current bank, has the following password requirements:

    6-32 characters, no spaces, alphanumeric + the following symbols only: [list of characters removed because /. thought it was spam; it was a fairly short list, though. Didn't even include an asterisk]

    Additionally, back when I signed up for online banking with them, I filled in a bunch of garbage for the security questions because security questions are just an attack vector, and I don't forget my passwords (I highly recommend KeePass for managing passwords, it's amazing).

    Anyways, a few years ago I went to log in and was prompted to answer a security question. Wtf? I had to call customer service to get my security questions reset. Now, if they don't recognize the device, or every so often, in addition to password you need to answer a security question.

    This means that I'm forced to either give real answers that I'll remember (and that anyone else could figure out to hijack my account), bogus answers that I can try to memorize, or garbage that I write down and hang onto.

    I also recall, around 10 years ago, I was using Bank of America and they had a limit of either 12 or 16 characters on your passwords.

    Of course, my email, web hosting, and even my fucking World of Warcraft use actual two-factor authentication, with phone apps that generate codes that are only good for around 30 seconds, and outside of a man-in-the-middle attack they're practically bulletproof. Why the fuck can't my online banking be as secure as them?

    1. Re:Banks just as bad by Anonymous Coward · · Score: 1

      This means that I'm forced to either give real answers that I'll remember (and that anyone else could figure out to hijack my account), bogus answers that I can try to memorize, or garbage that I write down and hang onto.

      I just give the answer "43" to all security questions.

      I figure, everybody would guess 42, but 43? Never!

    2. Re:Banks just as bad by DamnStupidElf · · Score: 2

      The solution is fairly simple; keep extra random passwords in KeePass or whatever else you use, one for each security question.

    3. Re:Banks just as bad by Anonymous Coward · · Score: 0

      Even more simple,
      Use your password for security question answers

    4. Re:Banks just as bad by Anonymous Coward · · Score: 0

      TD BANK limits your passwords to 8 characters! I enter my 8 character password, then add some junk after that... still lets me in.
      Try it out. Enter only the first 8 characters of your password and you're in

    5. Re:Banks just as bad by AK+Marc · · Score: 1

      My bank assigned my username to be my SSN. They revently notified me that my username had to change. The *only* constraint was that it could not be 9 characters long. I'm assuming they did that so people wouldn't just set it back to their SSN. So not my username is my name, real and much more guessable.

    6. Re:Banks just as bad by Secret+Agent+Man · · Score: 1

      For your Secret Questions, I just put the questions and their respective garbage answers in KeePass as well! My bank also does the "unrecognized device? Answer security questions" thing. I just pretend I have four usernames and four passwords in that regard. This way you can keep the garbage answers (really just password strings now).

    7. Re:Banks just as bad by zmooc · · Score: 1

      Why the fuck can't my online banking be as secure as them?

      It can be; just switch to another bank that's not run by idiots. It's your choice; instead of complaining here, just fix the problem :)

      --
      0x or or snor perron?!
    8. Re:Banks just as bad by Anonymous Coward · · Score: 0

      BofA has two-factor now; I think they even distribute keyfobs now.

      I don't think Chase has it yet, which kind of sucks.

      Two-factor rocks; Google Authenticator rocks harder

    9. Re:Banks just as bad by Asmor · · Score: 1

      I'd love to. Which banks in the US offer this?

      I haven't found any.

    10. Re:Banks just as bad by PCM2 · · Score: 1

      Bank of America offers two-factor. It will either text you, or for a fee, it will send you a card that generates the one-time codes.

      I don't know if you can configure it to use two-factor for your primary login, though. Two-factor is only invoked for certain kinds of sensitive transactions, such as trying to reset your password, very large funds transfers, etc. You may be able to set different levels of security before two-factor kicks in, I forget.

      --
      Breakfast served all day!
    11. Re:Banks just as bad by TriezGamer · · Score: 1

      Battle.net accounts are NOT as secure as they ought to be. Yes, two-factor authentication is good, but not everyone has access to it, and actual passwords are not case sensitive.

    12. Re:Banks just as bad by Anonymous Coward · · Score: 0

      This! (I also use nonsense "questions" with nonsense "answers". They don't care as long as you can reproduce them when you call.)

  28. Re:Who cares? by Anonymous Coward · · Score: 0, Insightful

    Seriously who THE FUCK cares?

    Uh, the guy who wants to crack your password.

    Don't give a shit what color hat he has on, this dumbass move is making his life a lot easier regardless.

  29. Re:You need more than 16 char by ATMAvatar · · Score: 4, Informative

    Even if you as an attacker know that the user chose 2 arbitrary words out of the English language as their password (or that only two mattered), and you knew there was a space between them, and you knew the login was case-insensitive, you still have to deal with the (minimum) 29,403,847,100 possible password phrases (171,476 common-use words times 171,475 unique second words, if we ignore word duplication and obsolete words). This also assumes, of course, that the password used correct spelling and did not in any way try to obfuscate the words with replacement schemes like l33t speak.

    Tell me again why it is terrible advice to use phrases?

    --
    "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety."
  30. 1000 guesses per second assumption by WinstonWolfIT · · Score: 1

    In cases where there's no physical access to the data, how does one get 1000 guesses per second? If my bank is going to lock my account after three incorrect guesses and if I keep a reasonable spread of account names and passwords, what's the actual risk of a 'weak' password actually being compromised? If I use a repeated account name and password on Facebook, what's the drama of someone guessing passwords at 50 per second?

    1. Re:1000 guesses per second assumption by Pinhedd · · Score: 1

      Pretty low. The overwhelming majority of compromised user accounts are a result of phishing and social engineering.

  31. At least... by mschaffer · · Score: 1

    At least you don't need to use your real name.

  32. Ummmmm. Clear text storage?! by Anonymous Coward · · Score: 0

    For a shorter 16-char version of your password to work, this means Microsoft has most likely not been hashing your passwords, but storing them in clear-text. So that any length of your password string is identifiable...

    Here's your chance, hack hotmail, and get a treasure chest of emails and passwords, and subsequent Bank password reset opportunities....

  33. I generate my passwords by Anonymous Coward · · Score: 5, Interesting

    My password is thus: SHA1 HMAC( PW, domain + salt ) -- Output as Base64 (where + is concatenation). I use this method because I can recreate the password at any time from anywhere. I don't rely on anyone else's password systems, I just use this simple algorithm which I can implement on any machine with the simple cryptographic primitives (hashed message authentication code, and a hash). I get a different password for each site, while using the same password everywhere. I change the salt and/or main password every so often, and only have to remember the current and last PW as I migrate to the new password as I run into sites I use.

    At first I created a table within the bookmarklette that would allow me to set additional rules for passwords, limit length, use a different set of characters for the base64 output -- The hash would be filtered on a per site basis to comply with all the bullshit. I could deal with such shortcomings five or ten years ago, but not today. Synchronizing the booklmarklette defeats the purpose of using a simple algorithm. If a site won't accept something like: NzE1YWViMGQwMjU3NWRlNmI3ZDQ0NTQ0NzI4MjE3MGU5YzRlMWY3NiAgLQo= as a password then I just don't use the service.

    I'll never use any Microsoft products, so I'll have to rely on others to discover: I imagine MS would simply ignore characters beyond the new limit? If not it would surely break password entry systems like my own or even saved password mechanic in all browsers... Including IE. It wouldn't surprise me if MS did break password entry for long saved passwords -- Smart folks who are security aware aren't their target audience.

    1. Re:I generate my passwords by anilg · · Score: 2

      > SHA1 HMAC( PW, domain + salt ) -- Output as Base64 (where + is concatenation).

      Interesting! So as long as the attacker does not have access to your local machine, you have more protection. (your password couls still be known for a single site, but could not be used elsewhere, does nto give away your 'formula' for passwords (like ending it with the site name, etc).

      Do you have this integrated with keepass or something? This should be a desktop tool, hosted on github.

      --
      http://dilemma.gulecha.org - My philospohical short film.
    2. Re:I generate my passwords by subreality · · Score: 2

      I used to do this, but it fails miserably on sites that require you to change your password, or when they redesign and redirect you to their "secure" domain to log in, or where they have multiple domains that share an account. I found I was spending too much time fighting edge cases, so I caved in and went for LastPass (which I consider adequate since it encrypts on my end). No regrets.

    3. Re:I generate my passwords by Anonymous Coward · · Score: 0

      Here is an early version of my above mentioned bookmarklette, in case anyone is interested. This one doesn't have HMAC, custom Base64 charsets, or other advanced options, but you should get the gist.

      http://paste.ubuntu.com/603942/

      Also I've noticed that Chrome and some versions of Firefox have "javascript:" URLs disabled. Screw that, about:config that moronosity.

    4. Re:I generate my passwords by Anonymous Coward · · Score: 0

      My hotmail account was recently brute forced from a 11-letter password I've had. Yes - 11 characters, and still brute forced. This was after they change their password policy. I am completely baffled that they failed to take simple anti brute-force measures that would disallow this. My password was NOT a simple word.

      I'll never use any Microsoft products, so I'll have to rely on others to discover: I imagine MS would simply ignore characters beyond the new limit?

      As it turns out, you *cannot* use more than 16 characters. I wondered how they expected me to create a secure password with such an idiotic small limit. They could at least have given me 32 characters. Come on.
      Also - they don't allow me to change my password to previous passwords, however they lowercase all your old passwords and have them in their history somehow. This basically means you can't change your password to "foobar42!" and then immidately change it to "FooBar42!" - that's the same password in Microsoft's eyes.

      If I could, I'd kill all accounts that enforce idiotic password rules, but reality is: I need some services that rely on them - I don't usually just sign up for fun.

    5. Re:I generate my passwords by frank_carmody · · Score: 1

      This is ingenious. Thank you so much.

    6. Re:I generate my passwords by Anonymous Coward · · Score: 0

      ... why not just make a base36 and a base62 version? Why waste time messing around with base64?

      Also, I do this, but I do it in my head.
      I have a 2D grid of letters and numbers.
      One axis has a word, the other has a number.
      I do sentence + number + service( name encoded from the grid in letters and numbers)
      Creates a nice thing to remember and is also service-sensitive, and chaotic enough to be secure for a while.

      Sadly, terrible websites made by 5 year olds often means I need to use the short version occasionally.
      Some people shouldn't be allowed to make logins.
      They should be prevented from having websites if they can't do simple security.

      As for subreality below. An easy solution to forced changes is just adding a date in to it too. Just add the month and year, that would do most things pretty well.

    7. Re:I generate my passwords by Anonymous Coward · · Score: 0

      This works well when you have something like LastPass managing it; sucks terribly for lots of sites with shitty password complexity requirements or if you're using it from a mobile phone or something like that since there's the risk that someone can just see exactly what you're doing and, more importantly, the plain-text seed that you're using for everything.
       
        I think using a checksum on that hash and adding some additional entropy is best. It's short enough to remember after entering it a few times, easy enough for you to recreate provided you remember the hashing vector, random enough to keep the botnets working for a LONG time and will satisfy just about every site's password requirements. Yes, some web devs are stupid and create length restrictions and other types of BS, but until there is some sort of a standard that everyone can agree to (or if everyone decides to implement two-factor auth, every security guy's pipe dream), I think this is a more reasonable approach.

    8. Re:I generate my passwords by Anonymous Coward · · Score: 0

      Additionally, I'm pretty sure that if you're hashing a hash, the salt for the original hash becomes irrelevant since the chances of recovering the first hash are incredibly, incredibly slight and will take an unreasonably long time anyway. Plus, what's the point of doing this if you have to remember, essentially, two passwords?

    9. Re:I generate my passwords by tuffy · · Score: 1

      I use something very similar but without the salt. The problem is that Slashdot and other sites limit passwords to fewer than the 28 characters of base64-encoded SHA1 hash. This is annoying because the password fields don't always have a limit, so I need to track that separately.

      --

      Ita erat quando hic adveni.

    10. Re:I generate my passwords by Anonymous Coward · · Score: 0

      This should be a desktop tool, hosted on github.

      Why do you care where it's hosted, unless you work for github or something?

  34. Just under the wire... by Anonymous Coward · · Score: 0

    So I won't have to change from "Suck it, Trebek!"

  35. Re:The disturbing thing is: they must be cleartext by GodfatherofSoul · · Score: 2

    They might have been only be passing the first 16 characters into the hash all along.

    --
    I swear to God...I swear to God! That is NOT how you treat your human!
  36. False value plus what the hash? by EmperorOfCanada · · Score: 1

    This is the sort of MBA spreadsheet thinking that kills companies. I suspect that someone did an audit that showed the passwords taking up all this "Valuable" space or some other bizarre analysis. The tiny savings from having the shorter passwords will instantly be nullified the first major hack that comes along.

    So MS is faced with one of three expensive situations:
    They weren't hashing but storing my pass in some open or reversible format which when hacked will create a mega PR / liability problem or,

    They are hashing and the truncated passwords won't work and they are going to blow off any customers who had long passwords which will tend to be the more technologically savvy who, as a group, are a bad idea to piss off as they are the types recommending technology to the masses. MS does not need to lose even more of the techno aware. or,

    They are hashing which means a truncated pass won't work and they will then have tech support hand out access willy nilly resulting in the easiest social phishing in the history of the net. "I would like you to set the password for the account bgates@hotmail.com to 12345678. Your name sir? Billy Gates you fool, now hurry. Thank you sir your password has been reset to 12345678. I would like to spend 5 minutes asking you about your purchase plans for our new $10,000 tablet..."

    1. Re:False value plus what the hash? by Anonymous Coward · · Score: 0

      This is the sort of MBA spreadsheet thinking that kills companies. I suspect that someone did an audit that showed the passwords taking up all this "Valuable" space or some other bizarre analysis. The tiny savings from having the shorter passwords will instantly be nullified the first major hack that comes along.
       

      If we assume they are truncating and storing the hash, well, we all know that doesn't save any space. What it does save, however, is CPU time & network traffic - fewer characters to send with the login / password change forms, less memory required to store the arbirtrary length password while it's being input and processed, and less CPU time spent processing arbitrarily long strings.

      Granted, the savings must be utterly, utterly trivial, but it's about the only plausable justification that I can think of!

  37. Well that's no big deal by kiriath · · Score: 1

    I'm sure most hotmail users are more worried about the minimum character limit rather than the maximum.

    "My password can't be 'poop' anymore. How will I ever remember it?"

    1. Re:Well that's no big deal by Anonymous Coward · · Score: 0

      Change it to 'pooppooppooppoop', obviously.

  38. Two-factor or stay home. by thesandtiger · · Score: 1

    In this age of ubiquitous spyware and key loggers passwords are pointless. Two factor security or don't trust anything important to a system.

    My passwords for things are simple, but I only trust important data to to factor. I just assume anything only password protected is compromised.

    --
    Since I can't tell them apart, I treat all ACs as the same person.
    1. Re:Two-factor or stay home. by darkfeline · · Score: 1

      No, passwords in fact work very well and are very secure given that everything is done properly, on both ends. "Everything is done properly." Yeah, I know, I laughed too. *bitter tears*

  39. My new password..... by Anonymous Coward · · Score: 0

    My new hotmail password must be "correct horse ba".

  40. WTF?! by shentino · · Score: 1

    Could someone explain how this would be even possible to pull off in the first place unless our passwords were stored in plaintext?

    Last time I checked, you couldn't truncate a password like this after it's already been hashed.

    Microsoft, shame on you.

    1. Re:WTF?! by Amouth · · Score: 1

      Well if you wanted to go to a new system that is limited to 16 char.. and given they have an account disable due to inactivity.

      then from the day of enforcement to the limit of disable due to inactivity you could:
      A) let someone login and check the hash -> if invalid truncate incoming password to 16 and check hash (depending on method in C)
      B) if it's valid check the length of the string used to create the hash
      C) if the length is >16 warn/force user to change password AND/OR silently truncate password used to 16chars then hash and store new hash.
      D) after you pass the limit of the disabled due to inactivity you remove all checks as either accounts are all =16 chars or they are disabled by normal policy.

      I've seen a similar system used to transition a site from plain hashed passwords to hash + salt passwords, it's not instant, but it works just fine in getting you to your goal.

      --
      '...if only "Jumping to a Conclusion" was an event in the Olympics.'
    2. Re:WTF?! by jrumney · · Score: 2

      Well if you wanted to go to a new system that is limited to 16 char.

      The question is why would you do that. Microsoft are a big enough customer to force the supplier to change their backward ways. Another more likely scenario is that they are moving from an old system that silently discarded everything above 16 characters to an unlimited system, but since old passwords only used 16 characters, they need users to only type the first 16 characters to be able to still log in.

    3. Re:WTF?! by Anonymous Coward · · Score: 0

      Last time I checked, you couldn't truncate a password like this after it's already been hashed.

      They've been truncating it from the very beginning. The hash was based on the truncated password. They're only now telling you about it.

    4. Re:WTF?! by Anonymous Coward · · Score: 0

      Could someone explain how this would be even possible to pull off in the first place unless our passwords were stored in plaintext?

      Last time I checked, you couldn't truncate a password like this after it's already been hashed.

      Microsoft, shame on you.

      Easy, they always ignored characters past the first 16 anyway, so the hash was already based on that limit. Says so at the end of the article.

    5. Re:WTF?! by Amouth · · Score: 1

      But honestly if that was a true statement, why on earth would they take the PR hit for it? IF you had an old system that was limited to 16 and silently discarded the rest, why not just use that to grow.

      A) Person logs in with password of length >16 char. Hash and test, if failed (B)
      B) Hash first 16 char and test, if passed (C)
      C) Ask user to confirm password
      D) If original and confirmed password match hash full password and Store. User will now pass on step A next time.

      Really isn't' that hard, they could have silently done it without any PR, and nothing more than a couple of people who might have to reset their password. And this just like my previous does not require them to store it in plain text, i have no idea why people keep thinking you would need to do that to make a change like this.

      --
      '...if only "Jumping to a Conclusion" was an event in the Olympics.'
  41. Re:The disturbing thing is: they must be cleartext by darkfeline · · Score: 1

    In which case typing the full password would still work.

  42. Watch out past self! by Anonymous Coward · · Score: 0

    Your email account in 1998 is now more vulnerable!

  43. Re:Ummmmm. Clear text storage?! by VTI9600 · · Score: 2

    Here's your chance, hack hotmail, and get a treasure chest of emails and passwords, and subsequent Bank password reset opportunities...

    Thanks, but the real opportunity was back in 1999 when they limited your password to two characters. Now those were some good times!

  44. Re:The disturbing thing is: they must be cleartext by Anonymous Coward · · Score: 0

    Such a low ID, yet such lack of understanding. Mighty impressive.
    Do yourself a favor, don't comment on anything security related anymore.

  45. Seriously? by holophrastic · · Score: 0

    A website chooses not to store an infinite length password of yours, and that makes headline news on slashdot? seriously, that's a problem? Guys, it's free third-party e-mail. It's not your safe-deposit box.

    Oh, and by the way, you can make the key to your safe-deposit box as long as you want, the lock will still only accept the first inch of it. Your girlfriend also won't accept more than 16 inches, by the way. Sometimes things are larger than capacity will allow.

    Not to mention, we all know exactly why they won't take more than 16 characters. Any bets your password's simply hashed into a 16 byte string anyway? Congrats, on your 17 character password being converted into 16 anyway.

    But hey, car doors and house doors with entry codes have 5 buttons each doubly-labelled. So 1 & 2 are on the same button. Making 11, 12, 21, and 22 the same double-press of the same button.

    Complain harder. Maybe then things like this might matter. Right now they make absolutely no difference whatsoever.

    1. Re:Seriously? by Kozz · · Score: 1

      I know you're trolling hard, but I'll bite anyhow. Some people reading here might (gasp!) actually agree with your drivel.

      A website chooses not to store an infinite length password of yours

      Here's the problem: they shouldn't be storing the password. They should be storing a HASH (that's a one-way function). Storing plaintext passwords is bad, m'kay?

      Guys, it's free third-party e-mail. It's not your safe-deposit box.

      Considering the provider, the infrastructure, the advertising dollars they make, etc. I think I'd expect more from Hotmail than I would the crappy POP3 account I might get from Charter (*blech*). Are you saying that because it's "free" (again, advertising dollars), it shouldn't be secure? Are you also saying that people don't have a right to ask for the simplest of security?

      Not to mention, we all know exactly why they won't take more than 16 characters. Any bets your password's simply hashed into a 16 byte string anyway? Congrats, on your 17 character password being converted into 16 anyway.

      Again demonstrating you don't understand decent hashing algorithms. Here's an exercise for you: pick any widely-used hashing algorithm (des, md5, sha-*, etc.) and then create a 100+ char string. Run that string through each of the hashing algorithms. Measure the length of the output (L) from each. Now feed the first L characters of your original 100+ char string into the same hashing algorithms again, and tell me if there's any difference in output.

      Go ahead, I'll wait.

      And finally, let's state the obvious: Limiting the password lengths also reduces the keyspace. See Brute-force attack.

      --
      I only post comments when someone on the internet is wrong.
    2. Re:Seriously? by kiriath · · Score: 1

      Not sure that anyone is necessarily complaining. It is just strange that in this day and age a company would opt to allow for something to be less secure than you would like for it to be. Gmail will accept gigantic passwords, being able to manage a password of any reasonable size is relatively trivial from a technical standpoint, and generally better security practice.

      Indeed locks invented hundreds of years ago are not quite as secure as a biometric lock, but you wouldn't go and replace a bunch of biometric locks with home grade locks... it seems like backwards progress is all.

      Someone might be a password nut, and wish to have a phrase or sentence as their password, why would anyone argue with that?

    3. Re:Seriously? by holophrastic · · Score: 1

      Wow, ok, now it's you who doesn't understand hashing algorithms. sure the first 16 of my 100 will produce a different hash than the 100. But some other 100 will produce the same hash as my 100. in the end, there are only 16 bytes worth of hash. more of your parameter fed into the hash doesn't change that. I thought I gave a pretty good example with the door lock, which effectively hashes 10 digits into 5 digits -- albeit by position which is quite apparent to the observer.

      And yes, as a free service, you get no *right* to anything. the advertizers are the client, not you. you get nothing more than what you're given.

      And no, 16 characters isn't insufficient. Again, these aren't your most prize possessions, these are shitty e-mails that you aren't even willing to pay for -- and it doesn't cost much to run your own e-mail server. Roughly $100 annually can cover you. $500 annually will cover you and half of your neighbourhood. So if you're willing to pay absolutely nothing, you don't get to complain about it. Do it yourself, it was never difficult.

      And once again, 16 characters is sufficient. There's no end to the amount of security that's possible, and you'll never stop Ethan Hunt. So once you stop 99% of the attackers, you're done -- physical safety and actual dollars aside.

    4. Re:Seriously? by holophrastic · · Score: 1

      First, you don't know how much google's passwords accept. You know that they don't tell you it's only 16, and it may be 17, but it probably isn't 10'000. So where's your line? Is 17 enough? What about 32? How about six megs?

      Second, I think that if you look into it, you'll find that biometric locks are actually inferior to good mechanic locks. Generally, no matter how much digital interface you put onto a door, the lock itself is still mechanical -- something physically stops the door from opening.

      So given that the lock itself is really good, what's more difficult to defraud: a) a piece of metal carved specifically for the lock; or b) an electrical voltage from a near-by circuit?

      Your fingerprint doesn't actually open the latch. It runs in a computer, which then has a simple circuit that triggers the latch. Hardware hacking is often VERY easy purely because biometric locks actually ADD a new way to open the latch -- with an electrical signal. Mechanical locks don't have any such entry path. The key actually does physically unlatch the door -- assuming it's a deadbolt.

      What you call "progress" is what I'm calling "paper progress". Why would anyone pay ten times the price for a mechanical lock that looks the same as every other mechanical lock? But wait, this new high tech lock has computers inside, and reads your fingerprint, that's why it's worth ten times the price. Yeah, in raw materials certainly. But not in actual safety.

      Not to mention, again, fingerprints are the all time dumbest way to lock anything. The reason fingerprints exist in technology is for forensics. And the reason forensics like fingerprints is because humans tend to leave them everywhere accidentally. Why oh why would you want to lock your home with a key that you leave copies of everywhere you go?

      Sounds incredibly retarted to me. And I hope it does to you too. It's like every time I touch anything, I leave a very delicate version of my house key -- instead of brass it's, I don't know, jello. It's something that everybody can see, and with basic equipment can copy and reproduce in brass all by themselves.

      But they don't need to because hardware hacking is cool. A pin inserted in the right place can open that biometric lock. How many slashdot stories have featured such things in the last month alone? I recall gun safes, and airport office doors off teh top of my head.

    5. Re:Seriously? by holophrastic · · Score: 1

      Oh, and hotel doors too.

    6. Re:Seriously? by Kozz · · Score: 1

      Apparently, then, it's far too difficult for the Hotmail folks to actually use hashing algorithms? Otherwise, why would they limit your password length? Are their hashing algorithms being executed with a stack of punchcards that some intern has to feed into the hopper?

      Please go further to defend this absurdity.

      --
      I only post comments when someone on the internet is wrong.
    7. Re:Seriously? by Rob+Kaper · · Score: 1

      First, you don't know how much google's passwords accept. You know that they don't tell you it's only 16, and it may be 17, but it probably isn't 10'000. So where's your line? Is 17 enough? What about 32? How about six megs?

      Six megs should be enough to break to most web server configurations for the maximum HTTP POST size. :-)

    8. Re:Seriously? by kiriath · · Score: 1

      Well, I did say 'reasonable size'.. I honestly doubt I'd be able to remember 6MB worth of password data. Let's not blow things out of proportion.

      What I'm saying is that we've worked on making things more secure - or making more in depth security possible. Let's keep moving forward, whether on paper or in real life any progress is far better than regress. I feel that limiting a password to 16 characters could be seen as regression.

      I'll concede biometric locks may not have been the best example, but the principal is the same, let's not move backwards.

    9. Re:Seriously? by holophrastic · · Score: 1

      There's just no point in your input being any bigger, since the hash output never will be. welcome to hashes. you seem to be saying that you'd prefer they not tell you.

    10. Re:Seriously? by holophrastic · · Score: 1

      But I don't think it is moving backwords. You're talking about improving the part of the system that is not the bottle-neck. Who cares? That's always a waste of time. It's like making one link in the chain way stronger than the others. It doesn't help the chain at all -- not one bit. (ha ha, bit)

      The goal, in security, is to figure out where the effort is best spent. And in this case, I don't think it's in password length. So then making it even 17 characters is a total waste of everything. Like getting a thicker steel door, when the windows are still glass. Or better yet, like strengthening your door, but not the hinges. Or strengthening your dead-bolt but not the door frame into which the bolt locks. You can have the strongest titanium dead-bolt in the world. I can still kick your door in if it's in a wall of straw.

      Password length is only one link in the chain. and I'm saying that 16 characters is already way past the strength of the other links of that same chain.

    11. Re:Seriously? by Kaz+Kylheku · · Score: 1

      You can reduce an arbitrary length password to a hash. (Say, 128 bits). You store that.

    12. Re:Seriously? by holophrastic · · Score: 1

      yeah, and you get 16 bytes. that's what they do. and as a result, there's no reason to ever give it any more than the 16 bytes of input. since "tnloehurlildoeucidtnoehudibwhmoeudbitnh" hashed produces "tughetucnoturecn" and "ilcroeuoeuitnsh" produces the exact same hash. so what benefit is there to having the longer input? It doesn't increase security at all. you're just picking a longer equivalent string. Because the hashing LOSES information.

    13. Re:Seriously? by lingon · · Score: 1

      No, you're wrong. Not in stating that a hash is a fixed length, but you're totally wrong about the consequences. If you were right, a 256-bit secure hash couldn't be used to protect your new shiny 4 GB ubuntu DVD image from malicious changes, for example. The whole point of secure hashes is that it is impossible (i.e. extremely improbable, as in you would never succeed before the universe dies a heat death) to generate a new input (of arbitrary length) which hashes to the same value. If you can, the algorithm is broken and should not be used (and this happens all the time: MD5 is broken, SHA1 is deprecated and everyone should be moving to SHA2).

      It does not matter if you are using a one kilobyte password input to your fixed length hash algorithm. It's still just as secure because it's designed to be.

    14. Re:Seriously? by lingon · · Score: 1

      2^128 is 340282366920938463463374607431768211456 unique combinations. Every 2^128 input string (at random) will hash to the same value, that called a Hash collision. Finding them is what is difficult: if you have a hash value and can find a string that hashes to that value, the algorithm is broken: You have succeeded with a collision attack. The algorithms that we use for secure hashing are designed to be resistant to those kinds of attacks. You should read up on some design criteria for Secure hash algorithms and you'll know why you're wrong.

    15. Re:Seriously? by holophrastic · · Score: 1

      The fact that you wouldn't succeed in *intentionally* producing two inputs with identical outputs changes nothing. Your big input is reduced to a smaller output. It doesn't matter why, and it doesn't matter how. Your input has been reduced. Information has been discarded.

      That's why you can't use the 256-bit hash to install ubuntu. It's missing nearly 4GB of data. That hash isn't a security measure. It's a transmission check. It checks to ensure that transmission hasn't dropped packets. It's alslo used as security today, purely because it's tough to succeed in making something different hash the same. But a) that need not always be so; and b) the entire point of your perspective is that for "small" hash lengths it's very easy (consider a 1-byte hash output today); c) much like with every security measure, it'll be broken by someone eventually anyway. So your point is moot.

      More importantly, welcome to security theatre. No I can't guess your 100-byte password. Excellent. I can still ssh into the server and gues it's 10-byte password. Congrats on barring the door, and not the window.

    16. Re:Seriously? by holophrastic · · Score: 1

      You might want to learn the definition of "resistant". Try it first with your wristwatch. Don't use mine.

    17. Re:Seriously? by Anonymous Coward · · Score: 0

      Again demonstrating you don't understand decent hashing algorithms. Here's an exercise for you: pick any widely-used hashing algorithm (des, md5, sha-*, etc.) and then create a 100+ char string. Run that string through each of the hashing algorithms. Measure the length of the output (L) from each. Now feed the first L characters of your original 100+ char string into the same hashing algorithms again, and tell me if there's any difference in output.

      Go ahead, I'll wait.

      And finally, let's state the obvious: Limiting the password lengths also reduces the keyspace. See Brute-force attack.

      Okay, but the keyspace is still limited when hashing is used. It is much bigger, but still limited. It's just that you have 256 different choices for each character as opposed to maybe 95 for all printable ascii chars. Oh and by the way, the last time I checked, with a fairly decent 8 character password, bruteforcing takes a lot of time or resources even on a "defeated" hashing scheme like crypt() des. (To search the whole 95^^8 keyspace you need like 2000 PS3-s to finish in 10 days)

    18. Re:Seriously? by lingon · · Score: 1

      Look, "Resistant" in this case means "Such a attack is not possible to execute". It is, by design, impossible to carry out the kind of attack you are proposing against today's secure hash functions (e.g. SHA2). That's the whole point of a secure hash function.

      It sounds like you're insinuating that it is trivial to generate hash collisions for a 128 bit secure hash, I don't know where you're getting that. There are 128 bit hash functions susceptible to hash collision attacks (MD5 comes to mind), but you don't attack them by trying the whole 128 bit keyspace. You do it by utilizing weaknesses in the algorithm, reducing the keyspace and making the attack feasible.

    19. Re:Seriously? by holophrastic · · Score: 1

      It makes no difference how. none at all. name a security algorithm that's remained secure for ten years. I'll point you to the numeral in sha2,

      And "resistant" is different than "proof" for a reason.

      And I never said that you can find two strings that hash to the same value. I said that two exist, and hence there is no additional hash space created by a longer input string -- because the output is still the same 16 bytes.

      Read harder.

    20. Re:Seriously? by lingon · · Score: 1

      It makes no difference how. none at all. name a security algorithm that's remained secure for ten years. I'll point you to the numeral in sha2,

      I won't bite for this straw man because it wasn't what we were discussing. The interested reader is refered to google and wikipedia.

      Your original point was that using a password longer than the hash keyspace would not increase security, quoting your post:

      yeah, and you get 16 bytes. that's what they do. and as a result, there's no reason to ever give it any more than the 16 bytes of input. since "tnloehurlildoeucidtnoehudibwhmoeudbitnh" hashed produces "tughetucnoturecn" and "ilcroeuoeuitnsh" produces the exact same hash. so what benefit is there to having the longer input? It doesn't increase security at all.

      And "resistant" is different than "proof" for a reason.

      Spare me the nitpicking. Unless you're a professional mathematician or cryptologer, let's just say your opinion about mathematical proofs in cryptology is invalid and leave it at that.

      And I never said that you can find two strings that hash to the same value. I said that two exist, and hence there is no additional hash space created by a longer input string -- because the output is still the same 16 bytes.

      Read harder.

      Of course, but that is a straw man and wasn't your whole post. The important part is that you added:

      so what benefit is there to having the longer input? It doesn't increase security at all.

      which implies that a) either a deliberate or random hash collision would easily occur or b) the keyspace is small enough to let it happen at random.

      For the first point: If you don't trust their one-way property, I recommend you to retreat into a non-electrified shed in the outback and never use the Internet, as all cryptographic security in the whole world are based on it. Every single last protocol we use are based on one way hashes in some manner.

      As to the second, the possibility of two 128 bit numbers, both chosen truly randomly, being the same is 2^128 (i.e. 3.4*10^38 or, in laymans terms, "never gonna happen during the lifetime of the universe, even if we get our best supercomputers cracking on it").

      I can put this to scale if you want as there are also thermodynamic proofs to this. An ideal computer using the least amount of energy possible to cycle through a 128 bit counter would require a about 1/100th of all the energy usage on the Earth combined for one year. A 256 bit counter in this ideal computer would require the entire energy output of a substantial number of supernovas to flip through all states. If this is your threat scenario, then go ahead and design for it, but please keep in mind that there are probably more credible threats out there.

      However, I won't have the time to reply to you any more. My point was that anyone reading this thread in the future and thinking along your lines would know why they're wrong. I think that has been accomplished, both by me and others. Feel free to read up when you feel you have the time.

  46. MICROSOFT BRILLIANTLY ....S too peed.... by Anonymous Coward · · Score: 0

    MOST older people or just forgettful people will substitute a real sound "èò_bfR43" type of password with much longer but equally sound "IndianEtherodyneForest33LakehurstManor" which would just be a switch from a MACHINE sound uppercase, symbol etc password to a password a actual human is capable of associating with events or places he's sure not to forget... remeber we ARE made out of flesh after all.

    Seems to me the usual pig headed faceless bureaucrat sort of decision made ba a techie with out ANY regads to human "haptics"...

  47. Re:You need more than 16 char by godel_56 · · Score: 2

    Even if you as an attacker know that the user chose 2 arbitrary words out of the English language as their password (or that only two mattered), and you knew there was a space between them, and you knew the login was case-insensitive, you still have to deal with the (minimum) 29,403,847,100 possible password phrases (171,476 common-use words times 171,475 unique second words, if we ignore word duplication and obsolete words). This also assumes, of course, that the password used correct spelling and did not in any way try to obfuscate the words with replacement schemes like l33t speak.

    Tell me again why it is terrible advice to use phrases?

    And at 100 billion guesses a second, using multiple GPU cards in a custom setup, you can test all those password in about 0.3 seconds.

  48. Re:The disturbing thing is: they must be cleartext by Anonymous Coward · · Score: 0

    Or, hotmail will delete your account if you dont log in for 360 days. Another option would be to re-hash the password as the user successfully logs in, and store two hashes. You would only need to do this for a maximum of a year, because after a year either everyone has logged in (and thus created the second hash) or their account has been deleted by the cleanup robot.

  49. But the first 16 characers of my password are by nbauman · · Score: 2

    0123456789ABCDEF

    1. Re:But the first 16 characers of my password are by Anonymous Coward · · Score: 0

      Hey! That's the last 16 characters of my password! I added those just to pad it out. An attacker doesn't know my short easy to remember high entropy password has this tacked on, so they still have to resort to brute forcing 24 chars worth of password, even though mine's only got 8 significant digits. When it comes to passwords, length matters.

  50. I don't even know my own passwords by 1000101 · · Score: 3, Insightful

    This is, well, stupid. I don't even know my own passwords. I have so many of them and they are so long with so many special characters that it would be impossible to keep up. I keep them in KeePass and just copy/paste them in the text box (it deletes the clipboard). Why place such a restriction on passwords when it is more important now then ever?

  51. Re:You need more than 16 char by Obfuscant · · Score: 2

    And at 100 billion guesses a second, using multiple GPU cards in a custom setup, you can test all those password in about 0.3 seconds.

    It's remarkable when anyone has 100Mbps network to the world, you've got a large multiple of 100 Gbps, and the server actually responds with "invalid password" pages that fast?

  52. Re:Who cares? by Anonymous Coward · · Score: 0

    I do. I have been complaining about this for a long time. I am one of the few people that chooses to use a password longer than 16 characters. The reason I do it is also Microsoft's fault. Back in the day, passwords 14 characters and shorter were victims of LM hash attacks. Does anyone remember that?

    Windows passwords 14 characters or shorter where hashed. But for some reason, MS decided to hash the first 7 characters, then the next 7 characters. So you effectively had 2 passwords of 7 characters. These LM hashes were still in use in Windows XP.

  53. Wrong by wisnoskij · · Score: 1

    Sites cannot just shorten passwords, as they do not know our passwords, only their hash.
    They may have started telling people about this, it it has always been this way (at least for the last decade). Like 7+ years ago I had long password on Hotmail (16+), I eventually learned that both the signup and login password field just allowed 16 characters so I was just typing extra characters for no reason.

    --
    Troll is not a replacement for I disagree.
  54. Re:You need more than 16 char by Anonymous Coward · · Score: 0, Informative

    29,403,847,100 possible words
    2 random words used for simple passphrase

    29,403,847,100^2 = 864,586,224,280,178,410,000 combinations

    You must live in a fun world where 8.64E20/1E11 equals 0.3

  55. Re:The disturbing thing is: they must be cleartext by hobarrera · · Score: 1

    if (password.length() < 16)
        rejectPassword();
    else
        hashAndStore(password);

  56. Old hash algorithm? by JDG1980 · · Score: 1

    Giving MS the benefit of the doubt and assuming that they are hashing the passwords in their database at all, I have to wonder if this might mean that there is some kind of weird legacy algorithm at work here. After all, Hotmail is an old Microsoft service, dating from the time period (roughly 1995-2005) when Microsoft was totally allergic to anything resembling industry standards. Perhaps they are using an old NTLM-style hash that only takes inputs up to 16 characters, because they never got around to updating the back end.

    1. Re:Old hash algorithm? by jrumney · · Score: 1

      After all, Hotmail is an old Microsoft service

      It was originally started by two Apple employees. Microsoft bought it after it was already well established (Yahoo had already bought rocketmail, which was the main competitor for free web based email in the mid 1990s).

  57. I don't understand... by 2fuf · · Score: 1

    ...what would be the benefit of restricting password length? It's not like you need to save storage space or anything. Why shouldn't Hotmail allow longer passwords, why is it important to them from a technical point of view to enforce a 16 char restriction?

    1. Re:I don't understand... by shutdown+-p+now · · Score: 1

      You can ask the same question for numerous other services that historically limited, or even continue to limit, password length.

      The answer is that 640Kb is enough for everybody.

  58. you should have read the thread by YesIAmAScript · · Score: 1

    This is an attack once you have the password hashes, not a front door attack. That's why rainbow tables were mooted.

    See parent (or grandparent or something).

    http://it.slashdot.org/comments.pl?sid=3135655&cid=41417323

    --
    http://lkml.org/lkml/2005/8/20/95
  59. bugzilla.kernel.org by grumbel · · Score: 1

    That problem sounds familiar.

  60. correct horse battery staple by Woogiemonger · · Score: 2

    So much for THAT strategy: http://xkcd.com/936/

    1. Re:correct horse battery staple by statusbar · · Score: 1

      well it still works! "'correct horse ba"

      who needs staples anyways? I hardly ever print anything out.

      --
      ipv6 is my vpn
  61. Re:if they used a hash...? by slashrio · · Score: 2

    The real question is how were they able to truncate your password if they used a hash?

    Somebody who understands the consequences of this, please mod it up!

    --
    "Trump!!", the new Godwin.
  62. I hate to say it by zhub · · Score: 1

    but create an account at redhat.com and you're limited to 18 characters.
    I dug through the recesses of my brain to remember a password that short and the JavaScript validation had the nerve to call it "Strong". (And the JS only worked on keyboard events.)
    Red Hat should know better.

  63. Easy-stop on brute forcing by Immerman · · Score: 4, Interesting

    Actually it's not that hard to "outsmart" brute-forcing - two simplistic ways are to insert a verification delay (artificial or computational depending on the situation) so that brute force attempts will generally takes months or years to succeed, or just block any attacker that makes multiple attempt faster than a human could reasonably be expected to. Even a really lax limit like blocking an attacking IP for a day after five failed attempts in a minute will block upwards of 90% of brute-force attempts and probably won't effect legitimate users at all.

    Think of it as somewhat analogous to being the doorman at a speakeasy or illegal gamblng joint - you know, the guy in the movies that spends all night opening the tiny window and saying "Password?". It not exactly hard for him to tell when someone is just repeatedly knocking on the door and guessing wildly and politely ask him to leave while they still only have a few broken ribs.

    --
    --- Most topics have many sides worth arguing, allow me to take one opposite you.
    1. Re:Easy-stop on brute forcing by Anonymous Coward · · Score: 0

      Well, your analogy is not correct as far as that doorman doesn't account for concurrency. It's more, like, you have a hella lot of doormen ready to answer as many doors. You can delay the login process, it won't stop a potential attack from issuing hundred of simultaneous connections to your login page.

      As far as identifying the actual mischievous address the attack is coming from, problem still lies with botnet. Hell, I've seen spam posts on a blog which requires two-steps (preview -> post) use a different ip for each step.

    2. Re:Easy-stop on brute forcing by Anonymous Coward · · Score: 0

      Besides, delaying the auth is calling for an easy-to-trigger dos attack on your login system.

  64. Re:The disturbing thing is: they must be cleartext by afgam28 · · Score: 1

    I suppose that code could be somewhere in the Hotmail source code, but is it likely that they would put a completely arbitrary condition like that in there?

  65. Re:if they used a hash...? by amiller2571 · · Score: 5, Informative

    We understand what he means, but if you did not read the update here you go

    This doesn’t mean that your password has been shortened. Actually, Windows Live ID passwords were always limited to 16 characters—any additional password characters were ignored by the sign-in process. When we changed “Windows Live ID” to “Microsoft account,” we also updated the sign-in page to let you know that only the first 16 characters of your password are necessary. To avoid this error message in the future, you only need to enter the first 16 characters of your password.

  66. Don't be silly by bussdriver · · Score: 1

    Amusing idea but not how it works. You can migrate people to new hashes easily and that is what they did. Simply look up in a new table and if there is no hash you use the old hash and if it works then you hash the password into the new table. Eventually everybody who uses the service is migrated. Unless they used an idiotic system they couldn't possibly get user passwords but they always get them as input to generate the hash.

  67. Damn! Only 95-bits of password security. by Anonymous Coward · · Score: 0

    What ever will we do with only 95 bits of entropy between our chain letters of cats and the international spy agencies wanting access to our email?

  68. not on ip6 by cheekyboy · · Score: 1

    If the server or isp supports ip6, the attacker just needs a home that can use 100000000000 IP addresses, and on ip6 is easy.

    --
    Liberty freedom are no1, not dicks in suits.
    1. Re:not on ip6 by Vekseid · · Score: 2

      > If the server or isp supports ip6, the attacker just needs a home that can use 100000000000 IP addresses, and on ip6 is easy.

      All with the same /64 or, if you're lucky, a /60 or /56 prefix.

      For my own CMS, I do ratelimits on /56 and /24 subnets. I track the hostmask on ipv6 for things like logins, but that's largely just because it's there.It's about as useful and relevant to me as your connecting port. And don't expect a site owner to treat it as any more unique.

    2. Re:not on ip6 by Immerman · · Score: 1

      And it's only slightly more difficult to block an entire range of addresses if that's what's needed. Admittedly though, yeah, ipv6 does seem to tilt the "easy game" in favor of the attackers.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    3. Re:not on ip6 by Bert64 · · Score: 2

      Not really, you simply assume a /64 is the equivalent of a single ipv4...

      It also makes other forms of attacks much harder, for instance with ipv4 it is common to scan a whole ip range looking for vulnerable hosts, with ipv6 this becomes completely impractical.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    4. Re:not on ip6 by Havenwar · · Score: 0

      Not really - access should not be measured by IP since a personal account should only be accessed by ONE person, regardless of IP. So five attempts from five different IPs should count exactly the same as five attempts from the same IP.

    5. Re:not on ip6 by Anonymous Coward · · Score: 0

      Isn't a /64 the smallest block assignable by ISPs? If so, then it is actually equivalent to a /30.

      Oh, and lot's of things about IPv6 are impractical, especially the default 1514 byte MTU.

  69. Pretty smart. by sgt+scrub · · Score: 1

    My bet is their customer base are people that prefer shitty passwords. Eventually they will add two factor authentication. Of course it will mostly be so they can blame the other party when the method gets owned. But they still won't have to anger users that prefer shitty passwords.

    --
    Having to work for a living is the root of all evil.
  70. Also, matched against a swear filter by Quick+Reply · · Score: 1

    When I tried to set up a password "FuckingAce" (not used for anything anymore), it wouldn't let me, I worked out that it wouldn't allow the word fuck and a range of other swear words. Thanks Microsoft!

    1. Re:Also, matched against a swear filter by Tastecicles · · Score: 1

      so I can't use this as a passphrase on a microsoft service: "Thanks a lot you shit-brained, fuck-faced, ball breaking, duck fucking pain in the ass."?

      --
      Operation Guillotine is in effect.
  71. Only one good explanation by dutchwhizzman · · Score: 1

    There is only one good explanation for this and that is storing the passwords either plaintext, or in a reversible encryption.

    Why do I think that? Hashing a 64K bytes long password will give the same hash size as hashing a 4 byte one. Since they must store some representation of your password somewhere, they will run into space and performance trouble if they have to store long passwords. Imagine having to store 500 million passwords that can be 64K each. That's a whole lot more space to reserve (yes, even 4 byte passwords need that reserved, you never know in advance how big a password is going to be) than 16 bytes. 32 Terabytes of password storage vs. 8GByte that you need for 16 bytes and 500M passwords. For a typical hash using a 16 byte salt, you'd need less than 64 bytes per password. That would give you 32Gbyte of database for 500M users. I'd say that's a significant difference, especially since you want those passwords available all over your datacenters (think replication and synchronisation over WAN links, you easily have more than 10 of those databases on fast storage worldwide). That's a difference that is big enough to warrant a serious limit on password size if you choose to not hash it, but use plaintext or reversible encryption. The only other valid reasons I can think of for requiring a maximum size is the workload you give your browser, the internet in uploading the POST request and their servers in calculating the hash. Since I very much doubt that will be the reason and significance for 16 vs 256 bytes is negligible in terms of load, I can't see any other reason than plain text or reversible encryption.

    For those of you that didn't get what all this means:

    Hotmail and MSN are most likely storing your password in a way that hackers can trivially get to read it if they get hacked. You may want to use a unique password, or avoid their service completely.

    --
    I was promised a flying car. Where is my flying car?
    1. Re:Only one good explanation by Tastecicles · · Score: 1

      Wasn't this story eventually revealed to be the result of a Hotmail database intrusion rather than a phishing expedition? Which would confirm the suspicions of many; that Hotmail does or did store user passwords in plaintext or weakly encrypted form.

      --
      Operation Guillotine is in effect.
    2. Re:Only one good explanation by Anonymous Coward · · Score: 0

      (yes, even 4 byte passwords need that reserved, you never know in advance how big a password is going to be)

      Someone isn't very good at databases....

  72. Not a big surprise by juliohm · · Score: 1

    Like other people already said before me, Hotmail NEVER accepted passwords longer than 16 characters. It would simply truncate it in silence. This is really a crap security policy and is notoriously known to be one of the worst security practices ever for password storage. Giving everyone awareness of the max password length, sounds to me like they REALLY want people to stop thinking of Hotmail is the place to be... I mean, think about it... technically, all this does is it makes Hotmail sound outdated and insecure. This might just be one more step they are taking towards forcing users to migrate to their new Outlook.com mail service.

    --
    Julio Henrique Morimoto juliohm@gmail.com
  73. Re:You need more than 16 char by Cinder6 · · Score: 1

    I would hope that large systems are taking measures against theoretical systems that could theoretically guess 100 billion passwords a second, such as by using PBKDF2:

    http://blog.agilebits.com/2011/05/05/defending-against-crackers-peanut-butter-keeps-dogs-friendly-too/

    Now I await someone much more knowledgeable in this area to come in and tell me why I'm wrong. (I welcome it, too; I don't want to be wrong any longer than I have to be.)

    --
    If you can't convince them, convict them.
  74. DTSS allowed control characters in passwords by Anonymous Coward · · Score: 0

    Dartmouth Time Sharing System allowed control characters in passwords. Control characters were actually preferred because they did not print on the Teletype terminals, so the system did not have to take time to overprint them.

  75. Re:The disturbing thing is: they must be cleartext by shutdown+-p+now · · Score: 1

    It works, but it misleads you into thinking that your password is stronger than it really is.

  76. Buffy, is that you? by Anonymous Coward · · Score: 0

    It sounds like a case of Buffy coming by. You know. Buffy. Buffy? Buffy! Buffy Overflow! She is there to blow your stack! I suspect they are trimming to 16 so that they don't have to deal with Buffy blowing the stack. And hey! All you need to do is watch them uppercase your password too, and then 26^16 iterations can solve that bad boy (4.3x 10^22 combinations) in no time! A half decent computer can bust that bad boy in a few hours. Usually people wanting internet security want longer passwords. Microsoft has kits for law enforcement (to check your dental records and perform rectal probes). Is this policy related to that?

  77. Re:You need more than 16 char by artor3 · · Score: 1

    There aren't 171k common-use words in the English language. The number is closer to 50k, and for words that people would pick "randomly", I'd be surprised if it's even 5k. By the time you get anywhere near 171k you're including words like "adactylous" and "grugrus" that no one is going to think of using in a password. Source.

  78. You don't say? by Lord+Kano · · Score: 2

    At least they warn you; I've run into some sites over the years that silently drop characters after an arbitrary limit.

    I was here when Slashdot truncated our usernames after an arbitrary limit. I was originally "Lord Kano - The Gangster Of Love", then one day I log in and my username was shortened to "Lord Kano - The Gang" or some such. I had to get Taco to shorten it to its current form. That was a pain in the ass.

    LK

    --
    "Hi. This is my friend, Jack Shit, and you don't know him." - Lord Kano
  79. Re:if they used a hash...? by galaad2 · · Score: 1

    this has also been happening to Technet & MSDN logins for a while now

    trying to access https://msdn.microsoft.com/en-us/subscriptions/securedownloads/default.aspx (or the equivalent technet downloads page) you get redirected to a login page that starts with https://login.live.com/login.srf and that form only alows 16 chars

    i went bonkers when it started to happen, a few months ago, but then i got used to it... this is the regular crap that's pulled by MS these days. :(

    --
    root@127.0.0.1
  80. Re:if they used a hash...? by galaad2 · · Score: 3, Interesting

    PS: the password field itself allows more than 16 chars, but if you enter more characters, when you try to login you get back a message telling you that the password is wrong. I can only login if i enter ONLY the first 16 chars.

    --
    root@127.0.0.1
  81. Re:You need more than 16 char by Tastecicles · · Score: 1

    such speed would require a die-speed interconnect.

    You got one?

    --
    Operation Guillotine is in effect.
  82. That's nothing; think how they store the passwords by rtfa-troll · · Score: 5, Insightful

    That's enough for hotmail !!

    An AC makes a reasonable on topic first post with a more or less accurate entropy count (note that both sexconker and Immerman's posts are right; since most users will get a-z with first letter capitalized and a single numerical substiution you get about 26 variations per character + 2 bits for the substitution that gets you less than five bits per character; of course if you use a password safe then you can use A-Za-z1-9 + about 20 - 30 punctuation characters depending on your keyboard, for about 90 characters giving you just over six bits). The only possible explanation that it gets modded to zero immediately is that it's anti-Microsoft and the shills are out with their large number of mod points as ever.

    Now, for the next trick. If you store passwords as a hash, as you are supposed to, then there is no way to shorten them since without the end of the password you won't be able to make the hash match. This means that at least somewhere Hotmail is storing passwords in plaintext. That's actually a much worse breach than having limited passwords since there is no way for the user to overcome it.

    AC's post was excellently insightful. It should be modded back up to infinity.

    --
    =~ s,(.*),<sarcasm>$1</sarcasm>,g if any_point_you_wish();
  83. This isn't new... by Anonymous Coward · · Score: 0

    My hotmail password has been limited to 16 characters for at least 3 years....

  84. Re:You need more than 16 char by Tastecicles · · Score: 1

    I hate to interrupt your bullshit with fully citable facts, but the OED says 171,476 in current use and 47,156 obsolete or deprecated.

    --
    Operation Guillotine is in effect.
  85. Re:if they used a hash...? by Sir_Sri · · Score: 1

    Um... Not many really.

    The hashing algorithm they use might have collisions past 16 characters anyway, so you'd get no added security out of extra characters, and you only hash and handle the hash from the first 16 characters.

    A limit of 16 characters is theoretically a problem, but not really, since the vast majority of users aren't making passwords that long anyway.

  86. Re:You need more than 16 char by Just+Some+Guy · · Score: 1

    Tell me again why it is terrible advice to use phrases?

    Because you still have to remember which phrase goes with which site. If you're going to go through that trouble, why not just use a strong random password generator and store the result in a password safe?

    --
    Dewey, what part of this looks like authorities should be involved?
  87. M$ stupidity strikes again... by Damouze · · Score: 1

    "Microsoft is making your life easier! You no longer have to input your whole password! Just put in the first 16 characters!"

    Unfortunately, so it is for the hacker trying to hack your account as well..

    --
    And on the Eighth Day, Man created God.
  88. Re:if they used a hash...? by Anonymous Coward · · Score: 0

    Just wait until you try logging into Hotmail if they migrated your account to "Outlook". I don't like the new look.

  89. Doesn't that imply clear text storing? by SpaghettiPattern · · Score: 1

    Doesn't that imply clear text storing? If a irreversible hash mechanism is used to store passwords then the clear text length wouldn't matter.

    An yes, IMNSHO clear text password storing should be made a criminal offence!

    --

    I hadn't the slightest objection to his spending his time planning massacres for the bourgeoisie... (P.G. Wodehouse)
  90. Re:You need more than 16 char by Havenwar · · Score: 3

    If you can make 100 billion incorrect guesses in a second to a remote webpage, there's really only two things to say:

    1. I want your internet connection.
    2. Password strength is not the critical flaw of that particular site - rather they should look into some way of not letting people try 100 billion passwords in a second without getting delayed/locked out.

  91. Re:You need more than 16 char by Tastecicles · · Score: 1
    --
    Operation Guillotine is in effect.
  92. Re:You need more than 16 char by Anonymous Coward · · Score: 0

    29,403,847,100 was the combinations, the number of words are 171,476.

    But thanks for playing, even if you failed really hard.

  93. Re:That's nothing; think how they store the passwo by JImbob0i0 · · Score: 5, Insightful

    It's feasible that the first time you log in since this was introduced that if the password validates then it gets truncated and the has based on the first 16 characters is stored.

    Once that's done any future password could be truncated to 16 and compared with the new hash based on the first 16...

    That way you can safely transition from one for to another without passwords stored in plain text.

  94. Re:The disturbing thing is: they must be cleartext by Anonymous Coward · · Score: 0

    We're talking about pre-existing passwords here. Truncating a pre-existing password to 16 characters requires that you obtain the password in plain text. If this is doable on the server-side as the summary implies, they have utterly failed at password hashing.

    If you're going to post smartass pseudocode, at least make it mean something. You clearly have no idea what the OP is talking about, unless you're implying that the site should update users' passwords to the next thing they type into the password field.

  95. Re:if they used a hash...? by Anonymous Coward · · Score: 0

    Yeah this is comptely true and it caused me a complete nightmare trying to get my Android phone with 3rd party mail app to connect to my Hotmail.

    Took me a good few password changes to realise that the website was only using the first 16 and the app sending all characters which didn't match.

    Lame having to choose a less secure password when using services from Ms. Even lamer that they didn't even provide any interface feedback on what had happened.

  96. Re:That's nothing; think how they store the passwo by ikkonoishi · · Score: 1

    Alternately they they match your long password against the hash, trim the password to the new length, and overwrite your old hash with a hash of the trimmed password.

  97. Re:if they used a hash...? by slashrio · · Score: 1
    I think the AC who posted the question originally,

    The real question is how were they able to truncate your password if they used a hash?

    was hinting to the fact that it implies that MS stores your passwords somewhere in plain text.
    In other words, vulnerable for hack attacks.

    --
    "Trump!!", the new Godwin.
  98. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 1

    ...If you store passwords as a hash, as you are supposed to, then there is no way to shorten them since without the end of the password you won't be able to make the hash match. This means that at least somewhere Hotmail is storing passwords in plaintext....

    Why? Couldn't hotmail be first trimming the password to the right length and then creating the hash?

  99. They only truncate new passwords by skovnymfe · · Score: 1

    Not existing ones, so quit pissin' yer pants over how Microsoft dun did it with storing passwords in clear text and whatnot.

    It's just like when they switched to a more reasonable password scheme some time ten years ago, they didn't force existing accounts to get longer passwords. My live id is still only 4 characters, all lower case.

  100. Re:That's nothing; think how they store the passwo by Ja'Achan · · Score: 2

    It's been trimming password to the first 16 chars for a while now. I only found out because Messenger only allows 16 chars in the password field, and when I would paste from KeePass my (longer) password I had set in the website, I'd get a beep.

  101. Re:if they used a hash...? by gnasher719 · · Score: 3

    The hashing algorithm they use might have collisions past 16 characters anyway, so you'd get no added security out of extra characters, and you only hash and handle the hash from the first 16 characters.

    A 128 bit hash doesn't have collision. In theory it can, if the hash function is cracked then collisions can be created, but in practice there are just no collisions. And there are plenty of devices (iPhone for example), where using lots of digits, upper/lowercase etc. makes the password impractical to enter, so I'd rather use a long (>16 chars) of lowercase characters and rely on length to produce bits.

  102. Stupid... by JimboFBX · · Score: 1

    My hotmail password was truncated to 16 chars over a decade ago. This isn't news. Amazing that there is 350 posts on this article and nobody pointed this out

  103. Typical by KrimZon · · Score: 1

    Here are some excerpts from the notes I took when creating accounts for various places. None of them are for Slashdot:

    "Registration truncated the password and then emailed it to me."

    "The password form said to use 8 to 6 chars, but seems to accept 20."

    "Not a valid password (must be 6-12 characters, contain at least one letter, at least one number and no punctuation, symbols or spaces)"

    "Passwords can't exceed 16 characters and causes a 'system error' when it contains non-alphanumeric characters. All this shit just for *****. Is it worth it?" UPDATE: It wasn't.

    "Kept throwing a strange error. It turns out it wanted to be alphanumeric."

    "Please use letters and numbers only."

    "Can't exceed 10 characters in password."

    "Truncated the password to 13 characters and failed to accept the full password."

  104. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    They might have never used more then 16 characters, there is no reason for the server side implementation to use all the input given to it, even if it doesn't prompt.

  105. anonymous_coward = 16 characters. Coincidence? by Anonymous Coward · · Score: 0

    just sayin'

  106. They always truncated to 16 and hashed that by oldlurker · · Score: 1

    Exactly. The fact that they can do this practically screams "We haven't bothered to implement even the most basic security precautions on our password database!" I mean come on - wasn't it established that storing recoverable passwords was a bad idea back in the text-only mainframe days? I could kind of understand it if it was some backwater site created by a high-school computer wiz, but Microsoft? Sigh. Yeah *sure* I'll trust your security software to keep my home PC safe - after all you're the company that did such a great job on the OS itself that running separate security software is practically mandatory.

    The explanation is in an update at the end of the article. They have always just used the first 16 characters, and hashed that, and ignored any additional characters you enter. The only change here is that they don't anymore let you enter the additional characters that they ignored anyway.

  107. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    Now, for the next trick. If you store passwords as a hash, as you are supposed to, then there is no way to shorten them since without the end of the password you won't be able to make the hash match. This means that at least somewhere Hotmail is storing passwords in plaintext. That's actually a much worse breach than having limited passwords since there is no way for the user to overcome it.

    The article says they have always just used 16 characters and ignored additional entered, so the hash was always based on the "shortened" password.

  108. Re:The disturbing thing is: they must be cleartext by AK+Marc · · Score: 1

    They've always hashed only the first 16 characters. They've never stored passwords in cleartext (though I think they sent them in cleartext previously).

  109. Re:That's nothing; think how they store the passwo by Qzukk · · Score: 1

    Or possibly they've been trimming the password all along, ala unix-style 8-character crypt.

    --
    If I have been able to see further than others, it is because I bought a pair of binoculars.
  110. I've seen a worse case... by BrokenHalo · · Score: 2

    Deliberately naming names to indict the guilty... :-|

    Bendigo Bank here in Australia truncates your password to just 8 characters, and like several other banks that I have come across, it also disallows punctuation or whitespace characters. So just enough characters to spell "FuckMeUp".

    At least they do have the grace to offer one-time-key widgets, (FWIW), at a price...

    1. Re:I've seen a worse case... by green1 · · Score: 1

      A bank I have had dealings with uses a 4 digit purely numerical password for online banking. They've recently allowed you to use a 6 digit one instead. And in a twist of irony, their mail-out newsletter recently had an article about online security telling you to use strong passwords for all online accounts...

    2. Re:I've seen a worse case... by davester666 · · Score: 1

      BMO here in Canada limits you to 4-digit passcodes for your bank card, and requires you to use that same passcode to login to the banking web site as well. And yes, digits means ONLY the numbers 0 through 9.

      --
      Sleep your way to a whiter smile...date a dentist!
    3. Re:I've seen a worse case... by green1 · · Score: 1

      Interesting... BMO mastercard doesn't have those rules, my BMO mastercard has a reasonably strong password (though the PIN is not as much, but at least I know they need my card to do much with that)

  111. moved almost all passwords by Anonymous Coward · · Score: 0

    Had to change passwords because of these various changes!

    www.pousadasemmonteverde.com.br

  112. Re:Dulls-ville night on /. by SternisheFan · · Score: 1

    With all the interesting stories slashdot users vote for waiting to be chosen, this and the last 4 or 5 are the lamest. I mean, the kindle isn't being sold at Walmart? Sheesh.

    I was wrong, and a man should admit it when he's wrong. If I could retract my above post, I would. I kind of lost it, and shouldn't have posted, was over-tired and not feeling 'nerdy', I guess. And now that I've read the 350+ comments that have accumulated, I once again have learned much. 10 lashes with a wet noodle for me.

    I apologize to the /. community, and to Timothy, for my blatant lapse of judgement. I won't allow myself to let this happen again. SF

  113. Thanks, Microsoft... by Anonymous Coward · · Score: 0

    I didn't need yet another reason NOT to use Hotmail, which I've never done, and most likely never will. However, you've gone ahead and provided me one. So, thanks. Why not make people's lives even easier, reducing the length and number of possible passwords even further, by just having your e-mail account holders type in their user names, and then giving them a multiple choice question, such as "what is your favorite color" and if they click the right one, they are logged in.

    If I want my password to be "fuck the following companies, Apple, Microsoft, and SCO!!!" then I should be able to have that as my password. No one should be able to tell me what my password MUST be, and I particularly hate it when someone insists I have two "special characters", two numbers, two capital letters, etc. as part of my password. If I am allowed to have an only eight character long password, the insistence that two characters each must be from some set only means each character has another 26 or maybe 36 or 40 possible states, which only marginally increases security.

    A password in all lowers, that is 8 characters long, has 26^8 possible words, or 208,827,064,576 combinations, starting with aaaaaaaa and ending with zzzzzzzz.
    A password in mixed case and the same length, has 52^8 possible words, or 53,459,728,531,456 combinations, that's only 256 times as many.

    A password in all lowers again, that is 16 characters long, has 26^16 possible words, or 43,608,742,899,428,874,059,776 combinations, you fucking Microsoft dumbasses. LONGER PASSWORDS ARE BETTER, ones that have a bunch of bullshit requirements for special characters are a pain in the ass. Far from contributing to security, they're hard to remember prompting some people (I once surveyed) the more complicated the password is required to be, the greater the odds that the average person will write it down somewhere, or several places, and that REDUCES security, Probably a lot more than the benefit recognized by using a greater number of letters/characters to begin with.

    If all you know about my password is that it is at least 8 characters, you would have a tough time getting, "fuck the following companies, Apple, Microsoft, and SCO!!!" by brute force or guessing, since you'd have to get the capitalization right, you'd have to get the punctuation right, and my hypothetical password, (or more appropriately pass-PHRASE) is about 60 characters long, including spaces, which count too. Now since I am (or should be) at liberty to use capital letters and punctuation, pretty much anything I can enter into my computer, and have no real arbitrary limit on length, let's see... 52 + 20+ 12 for the letters in both cases, numbers and punctuation above them, and a dozen for other marks the keyboard has, and I have 84^60 which gives me about 2.86 * 10 ^ 115 possible combinations, that's a lot higher than the Hotmail limit, and if I seem annoyed about someone with no business imposing arbitrary conditions and restraints on me and what I want to do, usually for idiotic reasons. I just tried to buy a gas can, and now they come in many assorted kinds but the thing they all have in common is that they all feature a CPSC mandated do-hickey in the spout that is supposed to reduce leaks, but actually it causes the container to leak, which is really fucking retarded. Think about it.

  114. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    Yep, but what are they "saving" by this? Storage, when they could change the algo for less storage without changing the actual password? Bandwidth, by having users log in with less characters?
    This is as fishy as Apple not allowing spaces in iTunes passwords ...

  115. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    What if the hashed value that was previous stored was already the truncated form of the password? It doesn't follow that Hotmail must be using plaintext passwords.

  116. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    Or they just were already only storing 16 and are now telling you about it.

  117. HA... by sidthegeek · · Score: 1

    16 characters ought to be enough for anybody.

    ;-)

  118. Re:You need more than 16 char by Anonymous Coward · · Score: 0

    methinks he mean cracking locally...

  119. Re:You need more than 16 char by Havenwar · · Score: 1

    Which isn't ever likely to be an issue for your hotmail password. While there are some leaks of password hashes now and then out there on the net, they are pretty rare compared to how many sites that require you to authorize. Saying that your hotmail password has to protect against offline cracking is a bit like saying your car should protect you from meteor strikes.

  120. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    Shill shill shill. Get aids and die you fucking bozo.

  121. Re:You need more than 16 char by MrL0G1C · · Score: 1

    Too right and that's why my password for a lot of sites is 'mypassword'.

    --
    Waterfox - a Firefox fork with legacy extension support, security updates and better privacy by default.
  122. Re:That's nothing; think how they store the passwo by davidorourke · · Score: 1

    I agree....password length should be infinite. Why? Reason is to keep hackers from figuring out your password. and when the special characters are allowed.....it throws a complete loop on a hackers time to access these types of passwords....ergo "SAFER THAN ALL OTHERS" Longer and more different characters that's created same way Roboform creates using an algorithm using capitals, lower case letters , numbers and special characters. The long the harder for a hacker. Maybe Microsoft is tired of trying to hack into these accounts............rotfl Shortening the passwords makes it easier for a hacker......DUH. Shouldn't take a rocket scientist to figure out that one. My 2 cents worth: a7%6j2#45%9kut%7ki8*9lifc#n6rc#,0plo See how hard that one would be to figure out. It would take the hacker over a million years to figure it out,,,,he would long be dead before his machine blew a circuit too.......lol

  123. Re:That's nothing; think how they store the passwo by Cow+Jones · · Score: 1

    If you store passwords as a hash, as you are supposed to, then there is no way to shorten them since without the end of the password you won't be able to make the hash match. This means that at least somewhere Hotmail is storing passwords in plaintext.

    Or that they've always silently thrown away anything past the 16th character.
    This used to be the case with old Unix passwords too, except that the limit was 8 characters: if your password was hunter123, you could log in with hunter12, hunter124, etc.

    CJ

    --

    Ah, arrogance and stupidity, all in the same package. How efficient of you. -- Londo Mollari
  124. Does Hotmail store clear text passwords? by bwbadger · · Score: 1

    How do Hotmail manage to shorten an existing password? I mean, unless Hotmail store the clear text version of the password all they have is a hash or something, and they can't work out a shorter version of the cleartext password from the hash ... can they?

  125. How about Canadian Imperial Bank of Commerce? by Anonymous Coward · · Score: 0

    This hotmail crap makes me laugh. CIBC's own iPhone app limits the password to 8 characters, simply ignoring the rest.
    I emailed them about it and the reply was "security is our main focus".
    And you worry about 16 char password for an email account...

  126. show me yours by Anonymous Coward · · Score: 0

    and i'll show you mine

  127. Re:if they used a hash...? by zippthorne · · Score: 2

    Assuming a perfectly distributed hash for the purpose, and that your passwords have 64 valid letters, and your password hashes are the equivalent of a 16 byte string, you are guaranteed to have more than one password string that hashes to the same hash string - 256^16 < 64^22. There may be passwords with no collisions, you cannot say that all passwords would have not collisions.

    Parent may have been assuming that the password box allowed any ascii character, in which case 256^16 < 256^(16+n) (where n > 0), which would also guarantee that at least some hashes would have colliding passwords.

    If the hash is hex encoded and only takes up 16 characters in that state, then it gets even worse.

    --
    Can you be Even More Awesome?!
  128. Simple Solution by swillden · · Score: 1

    Just turn on their two-factor authentication protocol, then you can use a somewhat shorter password and still be confident that your account is safe.

    What's that? You say they don't offer two-factor auth? Umm... what year is this? My gaming accounts offer two-factor, why in the world would anyone use an e-mail service -- given that your e-mail account tends to be the master key to all of your other on-line accounts via the "reset password" process -- which doesn't?

    --
    Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
  129. And Ivory Tower doesn't always tell tech support by Eric+Stratton · · Score: 1

    Ditto "At least they warn you"

    Suddenly, "Invalid Password"!

    Work with tech support for hours.

    Reset to original long password. Fail.

    Reset to short password. Success.

    Reset to original long password. Fail.

    Tech supports elevates problem to Ivory Tower.

    Ivory Tower: "What's your password?"

    16 character password.

    "Well, there's your problem. We just limited passwords to 12 characters."

    "Did you tell anyone?"

    "No. Why should we?"

    Must. Control. Fist. Of. Death.

    *

    Wait a minute! What happened to auto-truncate? We used auto-truncate XX years ago!

  130. My bank hates long passwords too by Anonymous Coward · · Score: 0

    And in fact i entered an 11 letter long one when I signed up, but they shortened it to 6 letters.. I don't know if it continues to accept the rest of the letters invisibly, but it only shows asterisks for the first 6 letters.. And this is after (this worldwide famous bank) implemented their recent "improvements" in security...

  131. MS notorius for POOR security - CRM is worse! by gabrieltss · · Score: 1

    Everyone here knows Microsoft is notorious for POOR security. Only allowing 16 character passwords now is not a surprise to me. But I couldn't believe how insecure they really could get until I started working on a project with my company to integrate Microsoft's Dynamic CRM (in the cloud) to a new application we are building. I was APPALLED to learn the ONLY way Microsoft will let you do "secure" web service calls from CRM is using HTTP Basic Auth (over HTTPS if you wish - oh gee how nice of them). They are even unable to do WSSecurity - in fact they don't support it. Gee how many people here couldn't rip apart Basic Auth and get the username/password? I'm sure very few couldn't. Now if you host Microsoft's CRM "in house" you can do more to secure your web service calls - but host it in the cloud and your screwed!

    --
    The Truth is a Virus!!!
  132. How does hotmail store passwords? by Anonymous Coward · · Score: 0

    If hotmail tells you to only use the 16 first characters of your password, doesn't that mean that they store passwords as passwords and not as hashes? There are no way they can tell the hash value of a shortened variant of a password if the password is not stored somehow.

  133. Re:leaked MSFT report on porting Hotmail from Unix by ei4anb · · Score: 1
  134. Re:if they used a hash...? by Sir_Sri · · Score: 1

    I figured that, but thats a hard assertion to make, because the password has to be passed in final form from the inputted text box through the hashing algorithm no matter what - ideally that operation is atomic and not interceptable, but you can't construct a hash character by character (at least not that I know of), so you have to take the whole input somewhere and then hash it. Somewhere along the line the whole input has to exist in memory.

  135. has been like this for a while by Anonymous Coward · · Score: 0

    This has been the case for at least a year, probably longer:
    http://www.blackberry.com/btsc/KB19709

    Dan.

  136. Home Again by T-Bone-T · · Score: 1

    I just signed up for a service for my dog and my password could only be 6-8 character. I couldn't believe it!

  137. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    This is the correct answer. Hotmail passwords have always been truncated after the 16th character.

    http://windows.microsoft.com/en-us/windows-live/microsoft-account-password-16-characters

  138. It's figurative language by tepples · · Score: 1

    What the fuck does it mean to be "compliant" with a comic strip?

    The same thing it means to be compliant with an HTTP error code: it's figurative language. But because I understand that people with certain mental conditions have trouble understanding figurative language, here's a literal version: The developer of the web site's user account system appears to have recognized the way of making passphrases expressed in the comic as a valid way of making passphrases with sufficient entropy.

  139. TD Bank Canada... by Anonymous Coward · · Score: 0

    limits to 8 characters..!

  140. Amazon did (or still does) by Anonymous Coward · · Score: 0

    I made a 32 character long randomly generated password, and despite Amazon saying it was accepted, it never worked. They told me I would have to call and tell one of their employees the password I wanted to use.

  141. Hotmail by betterprimate · · Score: 1

    Can't believe Mi

  142. Re:That's nothing; think how they store the passwo by rtfa-troll · · Score: 1

    yes; apparently so; it still shows total stupidity (if someone puts most of their entropy at the end of the password you really lose) but means that their security is not getting worse.

    --
    =~ s,(.*),<sarcasm>$1</sarcasm>,g if any_point_you_wish();
  143. Slow anyway by jago25_98 · · Score: 1

    Without spoofing the browserID to something mobile the high bandwidth needed for preloading all those emails in a presumably inefficient fashion means that you won't be checking spammail from a connection with packet loss or low bandwidth anymore (abroad on WiFi or Internet cafes)

    So it's not the first time we've seen functionality reduced. Also a shame for people using IMAP or pop3 where longer passwords can make more sense.

  144. nothing new by allo · · Score: 1

    my bank only allows 5 chars (alphanum)

  145. Vanguard only allows 10 characters by aggressivepedestrian · · Score: 1

    For years I used a long password at Vanguard. Then I discovered they only use the first 10 characters. So not only do that use weak passwords, they don't even tell you. You could have a long password, but if the first 10 were easily guessed or socially engineered, you wouldn't even know it.

  146. Re:The disturbing thing is: they must be cleartext by hobarrera · · Score: 1

    Uhm. Did you actually read the GPs comment? My reply makes perfect sense in that context.

  147. Artifact of BSD or Win2012? by Anonymous Coward · · Score: 0

    Supposedly hotmail is now running on Windows 2012 as a dogfooding measure. But before hotmail was acquired by MS and for a fair bit after, most if not all infrastructure was (?net)BSD based. So, is the password issue a leftover artifact from the BSD days that MS is just being more forthright about now, or is this a limitation of Windows 2012?

    Also, imagine Theo laughing in a corner while reading this...

  148. Re:That's nothing; think how they store the passwo by Anonymous Coward · · Score: 0

    Actually, hashes would match if characters after the first 16 were dropped in the original password capture process.

    What would not match in that scenario would the full longer password. In this scenario, it actually makes sense to automatically drop all characters after the first 16.

    And look mom, no plaintext passwords stored anywhere.

  149. Re:That's nothing; think how they store the passwo by ZygnuX · · Score: 1

    Am I the only one that gets the hunter2 reference? http://www.bash.org/?244321