Sure OpenSSH will protect you when you log into your *nix box. But what happens when you go to get your POP mail from your ISP? You send out your password in plaintext and then your mail is completely vulnerable. Does anyone make a mail server that encrypts with common clients?
You are correct, but you can tunnel POP (or whatever) over SSH. The fetchmail documentation explains how to do this.
Of course, there is still the problem that good old SMTP still goes unencrypted, but TLS-aware MTAs (TLS is the new name for SSL, basically) will encrypt the traffic between them! Recent versions of Sendmail are TLS-aware, there's Postfix-TLS, and experimental versions of Exim. Not sure about qmail.
As for POP and IMAP, I don't think anybody is talking about making encryption a standard part of them, but I could well be wrong.
Defaulting to www..com (or.org or whatever) would simplify and speed browsing.
This is WRONG. Sure, it may make things easier for the most naive of users, but it breaks things for more advanced users. (And I believe that if you keep aiming products at the lowest common denominator, you end up with stupider and stupider users...)
An example: Suppose I am testing a page on our development server (named "yeti"), and I type in the address http://yeti/path/to/page.html. Unbeknownst to me, Apache is down on yeti, so the browser can't reach it...and decides to go to www.yeti.com?!?! Which almost certainly does not have path/to/page.html, so I get a 404 which is almost certainly wrong. I typed yeti; I meant yeti. For the browser to assume it can read my mind and know what I really want is not helpful.
You can reduce the frequency of syncing. Syslog also does its writes synchronously, which causes the drive to wake up, but you can turn that off too. Whether doing so is a good idea is arguable...
"From Joe User's perspective", Joe User has absolutely no effing idea what constitutes a decade of UI advancement. Joe User thinks that gradient title bars are an amazing feat. Joe User thinks that menus that expand are cool
Truth be told, though, no one really knows what constitutes "advancement" in the UI field. There are decade-old UIs that (in some areas) beat the crud out of the latest and greatest.
I agree with your point about the analogy, but for a different reason: Shirky says...
Why does an increase in demand produce opposite effects on supply -- more available umbrellas and fewer available taxis?
*bonk* This is extremely sloppy terminology and extremely sloppy thinking. Rain and the corresponding increase in demand does absolutely nothing to the supply of taxis. There are just as many taxis as there were half an hour ago when it wasn't raining. Yes, there are fewer empty taxis - which is probably the point he's trying to make apropos to inflexibility of supply - but it's not a clear way to make the point.
$75 is cheap. There's a catch though...the licence doesn't include upgrades. So, if an exploit or other defect is discovered and fixed, you need to...
*whack*
You are correct that it doesn't include upgrades. However, the Recommended Cluster Patches, which include include security and reliability patches, are free. Check http://sunsolve.sun.com and look for 'patches'.
Wasnt Motorola partly behind a
certain satellite network that went (will go?) down in
flames?
Yup. I was there at the Circle-M ranch (as a contractor)
at the time Iridium went live, and it was a big deal
there. I have the lapel pin, which is probably the most
valuable part of Iridium at this point. *g*
Isn't Motorola stuck at half the clock
speed of Intel?
Not sure, but I don't think it's quite that simple. There's some
weird agreement about Apple only being allowed to buy
chips up to a certain speed or something strange like that.
Isn't Motorola screwed if Apple decides to make OS X Intel
Compatable?
Motorola and Apple, despite their ancient connections,
have been busily pooping on each other for a while now.
For example, you might think that Motorola would want to
support one of its largest and most visible clients by using
their hardware and software. But no....the common
desktop platform at Mother Moto is Wintel. Supposedly
some of it has to do with bad blood between personalities
at the top, or at least that was the scuttlebutt going around.
The only thing they are still doing ok at is the consumer
devices, like cell phones. Perhaps if they keep getting
screwed, they'll retract to that core?
Ah, but here's where you're wrong. Motorola is
hugely in with the government, and I suspect this is
why they've always been half-hearted about a lot of other
projects. When I was at the Scottsdale plant, there were
military guys around every single day, and military
communications vehicles in a gated section of the parking
lot. I don't know the exact numbers, but I hear that it's the
government contracts that account for most of their money.
Our search started in 1991 when one of us (PvA)
saw a black and white slide of a midsagittal magnetic resonance image of the mouth and throat of a professional singer
who was singing "aaa." He remembered Leonardo's drawing and wondered whether it would be possible to take such
an image of human coitus. We decided to try, as an ad hoc "instrument-oriented" study, despite the unscientific and
other irrelevant reactions we expected and received: honi soit, qui mal y pense.
I can just imagine the reactions they got: "So, you want a grant to take dirty pictures..." *g*
OK, so I just downloaded Opera and played around with it. It's got a few bugs, but OK, it's labelled "beta". What really annoys me, though, is a feature, and not a bug.
Go to File/Preferences/Network/Name Completion. You'll see two options: "Look for local network machine" and "Try name completion using". Why is this even an option?! If I type "http://wr", I damn well mean the host "wr" on the local network! This is the way name resolution works, for Crom's sake! Why does every browser in the world seem to think it knows better than you, your network admin, and the designers of your resolver put together?
David Conrad: "...we chose to create a 'lightweight resolver daemon' (similar in concept to Sun's 'ncsd'[sic])"
My only request of Mr. Conrad is that they actually make it do something useful, unlike nscd. For those who don't know, nscd is the Name Server Cache Daemon in Solaris. In theory, it caches requests for passwd, group, and hosts requests to make repeated lookups faster. In practice, you can't tell a damn bit of difference whether it's running or not.
What bugs me about Jobs' mention is this comment: "Apple was the first company to bundle networking with its computers."
Eh? Looking at our "History of Unix" chart, I see that both SunOS and HP-UX predate MacOS. Are they seriously claiming that neither Sun nor HP included networking capability in either their hardware or their software before Apple?! They don't say "personal computers", they do say "a list of networking people", and the list is full of non-PC types...
That led ultimately to a
letter from the National Park Service saying
the agency lacks jurisdiction over the moon.
I would love to have a copy of that one... "Dear Sir, In response to your recent query, we regret to inform you that we exercise no jurisdiction over the Moon. Yours Truly, John Smith, Directory of the National Park Service."
Even more, I would have loved to have been a fly on the wall when that issue came up at the NPS. "Hmm...hey, Joe, we got a guy on the phone who wants to know if we can declare a national historic site on the moon..."
I've never seen an explaination for why if Eric Corely thinks what he is doing is ethical, why he hides behind a false name that sounds suspiciously real.
Gee, maybe it's for the same reason that the authors of the Federalist Papers wrote under assumed names, or Stephen King has sometimes written under an assumed name - because it gets people to look at what you write, not who you are. It's not like his real name is a huge secret. (And it's "Corley", not "Corely".) And it only "sounds suspiciously real" if you're illiterate and unlettered. Engage brain, then speak, OK?
Firstly let me say that since service got set up, it's been great - consistent speed and no more than something like 30 seconds of downtime in three months. But getting service in the first place...oy.
I'm lucky because my "ISP" is my workplace, so I got to see information that most customers wouldn't. Here's the quick version: DSL was ordered for me and a few weeks later BA techs showed up and said, "Oops! You don't have a free pair of wires! Bye!" A Covad tech came by about a week later and dropped off the router. The service log accurately reflected all this, but now I was "off the install track" and in limbo.
I discussed with Covad the possibility of having BA disconnect my second phone line so that pair could be used, and the Covad guy was dismissive of the whole idea (Crom only knows why...) Then an entry showed up in the service log saying, "For technical reasons, you can't get the speed you ordered. Maybe you'd like to switch to a lower speed." Then a couple days later, the order got cancelled by Covad, with no explanation or notification.
So I got BA to disconnect my second phone line and we opened another request for service with Covad. The BA techs showed up and did their thing. On the day that Covad was scheduled to come, the guy called twice and said he was running late, then called again and said, "I had the techs check your line, and the reflectivity test shows a break in your line - our records show you at 4500 feet from the CO and our reflectivity test shows the end of the line at 4000."
So, out of curiosity, I tested the line with the router the first tech had left (and never picked back up) - and it worked! The next eight weeks were odd...I had service and Covad thought I had nothing, and the tech kept dropping by without scheduling an appointment. Nothing like coming home to find an increasingly annoyed-sounding series of notes taped to your door...
Finally we set up a date, I stayed home (and took down my working setup), and the Covad guy showed up. He was sloppy and unprofessional (for example, he stapled CAT5 to the wall without asking if it were OK), and then he had the nerve to say, "Yes, we're known for our punctuality."
can Intel hardware currently run 200-prosessor SMP-systems?
Um...neither can Sun. Sun's boxen top out at 64-way in the E10000. You'll have to look to SGI for 200-processor systems (up to 512 in the Origin 2800. And ccNUMA. I am awed...I want one.)
When they do that, you can almost rest assured that they will NOT be in business in 6 or 12 months. Can't think of a better sign of something very wrong in any company than automatically assuming that youth implies quality.
It's not the age thing so much as the unrealistic requirements. My coworkers and I have often joked about the number of ads we see that say, "Must have X years experience with Solaris, HP-UX, Windows NT, Cisco routers, ATM, Veritas, Apache, IIS, C, Java, PeopleSoft, Rational Rose..." and, and, and. And they want someone to do it for 40K. Oh, they'll hire someone eventually...but they still get to whine about how there's no one qualified out there.
(Background: you have to answer three questions to get a shirt.)
The first "question" is "The patent expiration will allow more developers to create secure applications, making the electronic world a more secure place?"
The only answers you can give are "True" and "False". What I want is a "That's neither a question nor a true/false statement!" link, ala Slashdot polls...
FascDot pretty much makes the point, I think. Since when has language dominance ever been based on numbers? Right now, English is the language of educated and technical folks just about everywhere, and if you want, say, a scientific paper to get worldwide study, you publish it in English; 500 years ago, it was Latin - despite the fact that no one spoke Latin on a daily basis!
Language dominance seems all too often to have been based simply on conquest: French and much of Africa (and Vietnam), Spanish and South/Central America, British English and just about everywhere, American English and the Internet... We wuz here first, for better or worse, and the Internet is not likely to stop being English-centric any more than the US stopped speaking English because we stopped being ruled by Britain. No comments from the British peanut gallery about what Americans speak, please.:-)
Personally, I'd like to see Esperanto gain greater currency...now if I could only find a keyboard with the proper circumflex. But that's vivo por vi.
As far as I can tell, Ü sounds like the long "oo" in English.
No, no...Ü is pronounced further back in the throat. "oo" is all at the front of the mouth. Then again, I learned most of my pronunciation in Nord-West-Deutschland, and like you say, regional variations can be fairly pronounced. (*rimshot*)
Their trackballs use infrared transmitters and receivers to follow the motion of those little black dots on the ball, no rollers. It's a shame all the shapes are so wacky...
Hey, one man's wacky is another man's just right. I love the Trackman Marble - I have long thin hands, and the TM is about the only mouse/trackball I've found that actually supports my hand instead of leaving parts of it drooping off the edges. The no-mechanical-parts-dotball is a thing of beauty, too. The only thing that could improve the design, IMHO, is if they could come up with a way for the support points not to collect crud. Of course, in another case of OMWIAMJR, Logitech seems to have fallen in love with the scroll wheel and is sticking it everywhere possible, and I can't stand that stupid thing.
What would be cool is if the dialog box had a text box at the bottom which produced and updated the command line string as you checked and unchecked boxes, pulldowns, etc.
AFAIK, AIX's SMIT does exactly this. You go through the dialog boxes and the system displays what the corresponding command line would be...and you can then go and edit the command line by hand, so you can leverage strength in both directions.
The statistic I keep hearing is that gun owners are 40+ times more likely to shoot yourself or a family member by accident than shoot an intruder.
Anyone know if this is true?
I hate getting involved in these stupid gun control threads, but in the interests of factual information... The claim was that a gun is about 40 times more likely to be used to shoot the owner, a family member, or an acquaintance than an intruder. One problem is that suicides are thus lumped together with homicides.
Another problem is that "acquaintance" was defined as anyone the shooter knew. This means that if Joe the crack dealer and Marvin the crack dealer knew each other from prison, and later Joe shot Marvin in a turf battle, it got recorded as an "acquaintance shooting". In the popular press, it mutated into "friend" half the time, which calls up images of best buddies killing each other in a moment of anger.
Likewise with family shootings - who's to say you can't be in mortal danger from someone who's related to you? If Rob got drunk and attacked his cousin Bill with a baseball bat, and Bill shot Rob, it got recorded as a "family shooting" (no matter how justified it was) which again, calls up images of Ozzie blowing away Harriet.
Lastly, the figure ignores cases in which no shots were fired - if a stranger approached Mary with a knife, and Mary drew a gun and the stranger fled, it just didn't go into the analysis at all, ignoring the fact that a life might have been saved.
I'm surprised no one has mentioned the old IE Solaris port. Which, as I hear it, had suckage in the mega-Lovelace range, and actually had to have it's own "registry".
The fake title (not the "Revenge" stuff to fake out merch pirates) was "Blue Harvest". "Horror beyond your wildest dreams" or something like that was the fake tagline for it. No one really bought it, anyway.
Slashdot Mirror
Of course, there is still the problem that good old SMTP still goes unencrypted, but TLS-aware MTAs (TLS is the new name for SSL, basically) will encrypt the traffic between them! Recent versions of Sendmail are TLS-aware, there's Postfix-TLS, and experimental versions of Exim. Not sure about qmail.
As for POP and IMAP, I don't think anybody is talking about making encryption a standard part of them, but I could well be wrong.
An example: Suppose I am testing a page on our development server (named "yeti"), and I type in the address http://yeti/path/to/page.html. Unbeknownst to me, Apache is down on yeti, so the browser can't reach it...and decides to go to www.yeti.com?!?! Which almost certainly does not have path/to/page.html, so I get a 404 which is almost certainly wrong. I typed yeti; I meant yeti. For the browser to assume it can read my mind and know what I really want is not helpful.
You can reduce the frequency of syncing. Syslog also does its writes synchronously, which causes the drive to wake up, but you can turn that off too. Whether doing so is a good idea is arguable...
"From Joe User's perspective", Joe User has absolutely no effing idea what constitutes a decade of UI advancement. Joe User thinks that gradient title bars are an amazing feat. Joe User thinks that menus that expand are cool Truth be told, though, no one really knows what constitutes "advancement" in the UI field. There are decade-old UIs that (in some areas) beat the crud out of the latest and greatest.
*whack* You are correct that it doesn't include upgrades. However, the Recommended Cluster Patches, which include include security and reliability patches, are free. Check http://sunsolve.sun.com and look for 'patches'.
Go to File/Preferences/Network/Name Completion. You'll see two options: "Look for local network machine" and "Try name completion using". Why is this even an option?! If I type "http://wr", I damn well mean the host "wr" on the local network! This is the way name resolution works, for Crom's sake! Why does every browser in the world seem to think it knows better than you, your network admin, and the designers of your resolver put together?
My only request of Mr. Conrad is that they actually make it do something useful, unlike nscd. For those who don't know, nscd is the Name Server Cache Daemon in Solaris. In theory, it caches requests for passwd, group, and hosts requests to make repeated lookups faster. In practice, you can't tell a damn bit of difference whether it's running or not.
Eh? Looking at our "History of Unix" chart, I see that both SunOS and HP-UX predate MacOS. Are they seriously claiming that neither Sun nor HP included networking capability in either their hardware or their software before Apple?! They don't say "personal computers", they do say "a list of networking people", and the list is full of non-PC types...
Even more, I would have loved to have been a fly on the wall when that issue came up at the NPS. "Hmm...hey, Joe, we got a guy on the phone who wants to know if we can declare a national historic site on the moon..."
I'm lucky because my "ISP" is my workplace, so I got to see information that most customers wouldn't. Here's the quick version: DSL was ordered for me and a few weeks later BA techs showed up and said, "Oops! You don't have a free pair of wires! Bye!" A Covad tech came by about a week later and dropped off the router. The service log accurately reflected all this, but now I was "off the install track" and in limbo.
I discussed with Covad the possibility of having BA disconnect my second phone line so that pair could be used, and the Covad guy was dismissive of the whole idea (Crom only knows why...) Then an entry showed up in the service log saying, "For technical reasons, you can't get the speed you ordered. Maybe you'd like to switch to a lower speed." Then a couple days later, the order got cancelled by Covad, with no explanation or notification.
So I got BA to disconnect my second phone line and we opened another request for service with Covad. The BA techs showed up and did their thing. On the day that Covad was scheduled to come, the guy called twice and said he was running late, then called again and said, "I had the techs check your line, and the reflectivity test shows a break in your line - our records show you at 4500 feet from the CO and our reflectivity test shows the end of the line at 4000."
So, out of curiosity, I tested the line with the router the first tech had left (and never picked back up) - and it worked! The next eight weeks were odd...I had service and Covad thought I had nothing, and the tech kept dropping by without scheduling an appointment. Nothing like coming home to find an increasingly annoyed-sounding series of notes taped to your door...
Finally we set up a date, I stayed home (and took down my working setup), and the Covad guy showed up. He was sloppy and unprofessional (for example, he stapled CAT5 to the wall without asking if it were OK), and then he had the nerve to say, "Yes, we're known for our punctuality."
Don't get me wrong; I really like Sun hardware for the most part. But it ain't perfect.
The first "question" is "The patent expiration will allow more developers to create secure applications, making the electronic world a more secure place?"
The only answers you can give are "True" and "False". What I want is a "That's neither a question nor a true/false statement!" link, ala Slashdot polls...
Language dominance seems all too often to have been based simply on conquest: French and much of Africa (and Vietnam), Spanish and South/Central America, British English and just about everywhere, American English and the Internet... We wuz here first, for better or worse, and the Internet is not likely to stop being English-centric any more than the US stopped speaking English because we stopped being ruled by Britain. No comments from the British peanut gallery about what Americans speak, please. :-)
Personally, I'd like to see Esperanto gain greater currency...now if I could only find a keyboard with the proper circumflex. But that's vivo por vi.
Another problem is that "acquaintance" was defined as anyone the shooter knew. This means that if Joe the crack dealer and Marvin the crack dealer knew each other from prison, and later Joe shot Marvin in a turf battle, it got recorded as an "acquaintance shooting". In the popular press, it mutated into "friend" half the time, which calls up images of best buddies killing each other in a moment of anger.
Likewise with family shootings - who's to say you can't be in mortal danger from someone who's related to you? If Rob got drunk and attacked his cousin Bill with a baseball bat, and Bill shot Rob, it got recorded as a "family shooting" (no matter how justified it was) which again, calls up images of Ozzie blowing away Harriet.
Lastly, the figure ignores cases in which no shots were fired - if a stranger approached Mary with a knife, and Mary drew a gun and the stranger fled, it just didn't go into the analysis at all, ignoring the fact that a life might have been saved.
I'm surprised no one has mentioned the old IE Solaris port. Which, as I hear it, had suckage in the mega-Lovelace range, and actually had to have it's own "registry".
The fake title (not the "Revenge" stuff to fake out merch pirates) was "Blue Harvest". "Horror beyond your wildest dreams" or something like that was the fake tagline for it. No one really bought it, anyway.