Here's my spam filter, which is implemented in procmail and gnus, which I inherited from a person I used to work with:
-- First check that it is addressed to a mailing list you subscribe to
-- Then check to see if it explicitly lists your address on the To: line
You will see that a majority of spams are not directly addressed to the target on the To: line.
But the problem is not whether something can get 90% of all spam. A variety of rules can be composed as you suggest to block a high percentage. It is the false alarms of these rules -- the emails that are blocked that aren't spam. I put all the spam in separate folder, and yet I still sometimes get mail there that I wanted to go to one of my regular folders. What is the false alarm rate of your system ?
I think the validation of the sending email address is a great idea. You can almost never respond to unwanted email, but you can respond to legitimate email.
Still, ultimately the spammers will find a way through. They are human and I think they will always be smarter than a structured filter system. Just look at the newest tactic -- it's called "viral marketing" -- it basically amounts to rewarding your friends for spamming you.
First, an important detail: it was Amazon that initiated legal proceedings with B&N, not B&N that sued Amazon. You can see the cnet story as a reference. This fact figures strongly in rms's opinion on the matter.
That said, it is not true that "if they don't do everything _legal_ to up their earnings, he stockholders can and will sue." The management of a company is simply not under obligation to explore every possible legal source of revenue. Can I sue Ford because they don't write software, or GM for not owning gold mines ? If I buy Redhat stock and they don't drill for oil, I was not injured in any way because I knew they weren't in the oil business when I bought the stock.
The responsibility of the mangement to the stockholders is merely to not fail to do something a stockholder would naturally expect without telling them first. You can't start a company with a nice looking published business plan and have a real secrete plan of paying yourself a salary to do nothing until the company folds.
There would be no reason to uphold a suit against Amazon for failing to pursue the patent, especially if they had declared publicly that they were not going to pursue software patents. A good court would hold that the stockholders can sell their stock if they don't like it. A good court would see itself as a protector of the stockholder from deceit and neglience. The market will take care of incompetence; the court just has to assure the information is not kept from the market.
Think about it this way: the ability of stockholders to take their money and run makes the stockmarket a very efficient judge of business strategies. Do you really think that a judge and jury can do a better job ? Resort to the courts may be justified when extreme and rare plots and colusions attempt to hender the efficiency of the market.
There existed no reasonable expectation that Amazon should have a business plan based on the bureaucratic incompetence of the PTO. Amazon could have choosen to explicitly deny themselves that path.
But they didn't. Jeff Bezos and his company have choosen a business plan which has as a primary feature the denial of my freedom. Their plan is make sure no one else (including me) can make a one-click shopping site, and they plan to live off of this government enforced monopoly. If that doesn't make Amazon the Antichrist, it at least puts them on the same team.
I loose freedoms every time the PTO gives out a patent, but much of the time I at least get something back -- a business that would not have been worth trying without the monopoly guarantee is attempted, bettering the economy on the whole. However, Jeff Bezos and Amazon are just a tax. If we did not award business method and software patents, someone still would be selling books online with one click, and I would still have a tiny piece of freedom that Jeff Bezos wants from me.
I fear legal responses to annoying spam. I would much rather use filters, or even absorb some cost of downloading unwanted messages and a tax on my attention/time, than have the government start prosecuting people.
In previous slashdot spam discussions www.privatecitizen.com has been recommended as an excellent guard against unwanted phone and snail mail advertisements. I don't use it ( I see using the court system to go after these people as only slightly less objectionable than using the legislature), but perhaps someone who does can comment on it's effectiveness and whether it would guard against cell phone mail spam ? It seems that privatecitizen depends on being able to distribute a list to known advertisers, and I think that many spammers ( wireless and regular ) are much more fly-by-night types.
Several times in this thread Europeans have jumped in saying that they have to pay to download spam because of non-free local calls. But it is technically possible to make a good spam defense even without having to download the entire spam:
Get a shell account and read your mail on the server, downloading only what is displayed to you. Of course, you pay for the connection while you stare at messages and compose them, so you might look into something more sophisticated, such as . . .
Get a shell account and split the spam with procmail or the equivalent (on the server), and set up something like fetchmail to only get the important stuff, plus a log file of all the headers/subjects of the spam, just in case an important one slipped through the filter and you want to actually look at it. Or . . .
Have a program that talks directly to a POP server once you are connected, that downloads just the headers and subjects, starts downloading everything obviously not spam in the background while presenting you with a list of of everything so you can select things to download and read and things to delete unread.
Why haven't European (or American, for that matter) ISPs already provided this as a service to differentiate them from their competitors ? Why haven't any of the free software people provided the same, just as they provided junkbuster ?
I suspect it is because annoyance at spam is not as widespread as a vocal minority would have us believe. It is just not that big a problem in the larger scheme of things. Otherwise someone would have already written the program I listed last above and they would be making money going to ISPs and integrating it with the little custom windows dialers and email clients.
Web banner ads are more annoying and take up time right when you are trying to actually do something (look at a web page), so fairly effective filters came out quickly. But I suspect that most people also stay on longer than necessary just to download their mail, because they briefly check the slashdot headlines for example, and their mail can download in the unused bandwidth while they browse.
I am afraid that we will let government regulation do it's usual heavy-handed solution that will only stop 50% of the problem anyway, rather than picking a technical solution which involves less emotionally gratifying yelling (and slashdot posts) and would solve 80% or 90% of the problem. If non-download filters were common and the default on ISP services, response rates to spam would drop.
Rather than making a disk image, I think you can use the command dump ( see "man dump" ) to put a giant file on a cd which you will restore to the disk with the dump command.
To get a target machine to the point where you can mount a CD and run dump, I suggest checking out some of the linux floppy distributions.
There is one other option. There used to be a web site that offered you all the choices offered in RedHat's install script, and then created a custom install floppy which would automatically install according to those specifications, no human intervention necessary. So you could just go to this site (or it's equivalent for slack or whatever other distribuiton you want, or download whatever tool the site is running and use it yourself) and then make a boot install disk and just shove it and the cdrom in every machine. I think the boot floppy/dump from cdrom method will work faster.
If I buy three wireless LAN equiped laptops (such as those ibooks or PCs with cards, whatever) and arrange them so that laptop A is at one end of a very long room, and laptop B is at the other end of the room, so that B and A *cannot* communicate, but I place C in the middle such that B and C *can* talk and A and C *can* talk -- does the obvious thing happen ?
That is, does C begin re-broadcasting packets going between A and B so that they get to right place ?
As I understand things, this is a software issue -- do the machines recognize when to configure themselves as routers ? If only some OS's or drivers do it, which ones ? If I install linux on an ibook or PC with the wireless LAN card, can I make linux do the right thing ?
If there were two wireless ethernet devices on the machine in the middle, I think I could figure out how to make linux do this. But can linux re-route packets through a single interface device ? I know it is possible to re-route things heard on one device to be broadcast on another, since that is what my IP-masqing linux box connected to cable modem on one card and the apartment-mates machines on the other does.
ZoomAir has something they call "AcessPoint Software" which does this on a machine with one wire card and one wireless card. Can this software be tricked to use a single wireless card as a bridge between two networks ? Furthermore, can you configure it to automatically recognize the situation in which it should start routing, and stop when it shouldn't ?
I've been doing google searches -- does anyone have any URLs of where to start ?
Why buy old technology that just works when you can spend a lot more for a wireless ethernet card in each machine and be totally cool ?
One machine can have one of those obsolete non-wireless cards in it, so you can plug into your clients hopelessly out of date non-wireless network, and route through that machine.
I'm actually saying this only half tongue-in-cheek. If you have the cash to be carting around multiple laptops and a palm, you can afford it. It depends on the business you are in, but the coolness factor of that might get you enough new business to pay for itself the first time out.
In my opinion and experience, CASE tools are an abomination that merely serves to produce some eye-candy along with the code to distract everybody from worrying about whether or not the stuff works.
I know that in some areas they are very well entrenched, such as in some areas of the aerospace industry. I think it must be related to the way projects that have a high cost of failure become highly rigidized and over-managed, instead of simply being over-engineered and receiving the best brains.
In my opinion, the decision to use a CASE tool or the standardization of a project on a particular IDE is the first sign of ultimate failure, or at the very least, a sign that you are going to spend a lot of money to get bloated half-working mediocrity (which passes for success in some places).
I will ignore the side issue of whether collaborative document development without extensive communication through a side channel (telephone, email, face-to-face converstations) is a good idea in the first place.
I have worked on several different projects in which we did or attempted to do collaborative document editing. Usually it starts out with one person maintaining an official web site version and merging in everyone's changes, and then it devolves into a token-ring-like situation with emailed attachments being mailed around to everyone clogging mail systems and confusing the issue who has the official version.
Just use LaTeX and cvs. The people who do word won't like it, but tell them tough. They get used to it surprisingly quickly, especially if you tell them that no-one should be worrying about margin widths or other formatting issues until it is to the point of one person editing it.
At that point you can convert it to Word. Converting LaTeX to word goes surprising fast and easy.
Where can I get a copy ? They seem to have been stolen from all of the libraries, and my old fallbacks for out-of-print books (MacIntyre and Moore in Davis Square, powellsbooks.com) don't seem to have it. I started wanting a copy after I read this one, and spoke with one of the authors:
S(h)elf Help Guide - The Smart Lifter's Handbook by Gabor Caime and Gabor Ghone
(It says, "If you paid for this book, you might learn something! If not, take it back -- you don't need it." Probably commercially smarter advice than "steal this.")
I think the point is to know about these things before the deal is closed and it's all over the newspapers.
Also, I believe the US intelligence community has many people who are assigned to a particular region or country and come to work every day to read all the newpapers from there, listen to TV and Radio news, educate themselves tecjhnically on the major economic industries, and otherwise be Johnny-know-it-all for every that has anything to do with that area.
I have never had data loss with linux unless it was me inadvertently deleting something. I loose my machine to power outages at least once a month, and once when netscape was open it came back with an empty bookmarks file, but the whole bookmarks file was in/lost+found. I also regularly start huge compile runs on my laptop, put it in my backpack, and let it compile until the battery runs out. When I plug it in it fsck's on the way up, but I just restart the compile and it is fine. So I don't have your experience of loosing data on a power outage.
I recently bought a cheap big IDE drive for my home machine. I plan to have a couple of partitions for experimenting with ext3 and other journaling filesystems. But I don't really ever stress the ext2 filesystem, so all I can test is that it is as good as ext2. How would you test a filesytem for reliability and crash resistence ?
Another question: I have noticed that access from linux to a FAT32 partition is much slower than from windows to a FAT32 partion. Linux FAT32 access is also much slower than Linux ext2 access. I would expect some difference -- who wants to spend a lot of time making the FAT32 fs code fast -- but the difference seems rediculous, at least 2 times slower. Why is this ?
"And make sure the cleaning people don't unplug your server to plug in a vacuum cleaner.(or is that an urban legend?)"
I think what Paul is refering to here is not the fact that the late night staff occasionally unplug machines (an environmentally conscious janitor were I work used to turn my linux PC off -- surely similar things happen everywhere) but a particular old 60s or 70s anecdote about an IBM mainframe that always mysteriously and unexplicably crashed on a particular day of the week, at a certain time at night; and a support team at IBM was so baffled that they actually went to the site one night and sat down to watch it happen; and the janitor entered, unplugged the machine and plugged in the vacuum. It turned out he always vacuumed that room on that day of the week.
I believe this story is in the New Hacker's Dictionary, the most recent edition. Unforntunately my copy is at home so I can't consult it.
It's a neat anecdote when well told. It would be interesting to know if it were true or urban legend.
I think that sounds a little overly discouraging. I would say that the asker is probably much better equiped to learn about and set up Linux than most new users.
Rather, I would advise that you just check out a linux installation that is as standard as possible. See if that is reliable enough, and don't experiment with any new filesystems unless plain vanilla won't cut it.
Also, you said it was an "NT Server". Could you give more details on what it is serving ? If it is just diskspace to windows client machines, it should be easy to configure.
It didn't seem like a "great paper" to me. It had all the ideas that an afternoon a brainstorming with some reasonably intelligent folk should produce. And when they go off on that biological metaphor of the hardware being the body and software the soul etc -- and then they start making analogies about a single body having a succession of souls because you reprogrammed it -- that's just hippie talk. It reminds me of why I must stop myself from ever going to graduate school: when I am in a very overworked and sleep-deprived state and people start distracting my precious few neurons with meaningless strings of attention-grabbing words I want to kill them, and thus loose what little productivity I still had.
But even if these guys are just a couple of flaky pot-smoking vegetarian hippie coffee house denizens, they do have some interesting references at the bottom. I'll be checking these out on my next trip to Barker library.
I like the emacs gnus newsgroup reader. I also use it for mail on some of my accounts, and on my main account I have procmail directing my mailing lists into spools that I read with gnus.
Gnus just rocks.
I would really like a gnus interface to slashdot. Maybe some elisp hacker could use the emacs web borwser (w3) and link them together. Among the benfits that would bring would be the ability to compose posts in emacs, so I could use a spell checker.
I would read so much more of slashdot in so much less time. In fact, if I sat down and did that task in week or whatever it would take me, I bet I could make up the time in three years.
Other sites, such as arstechnica (their html sucks watermelons through a garden hose) and wired (same class) and nytimes (slightly better) could benefit from this.
Could you have access to the cookies and deal with them in elisp ? Because then you could write little elisp functions which would cycle the nytimes account randomly through a list of fake accounts you built up, putting noise in their database of what articles you visit.
"For one, it appears that the community expects the work to be done for no money."
It was "the community" which did the bulk of the work here, not Dell.
"They scream because the machine isn't cheaper with Linux on it, even though the amount of work to be done to get Linux to work decently with laptops is considerable, far more than with desktops . .."
If they would publish the specifications to the parts of the laptops, they wouldn't have to do any work, because the linux people would have drivers out before their slow-moving corporate programmers could assign a project number to the job.
" . . . and support costs can be expected to be higher (simply because there are far fewer experts on Linux-laptop issues than for Windows-laptop issues)."
On the other hand, a given linux user is far more likely to also be an "expert" (i.e., not want or need support) than a given windows user.
"Second, the community screams because support for all platforms isn't instantly available, even though many laptop components don't have Linux drivers present."
While I shouldn't speak for the rest of the community, what burns me is that they use those laptop components which have secret specifications; this is what generates the lack of linux drivers, not the lack of effort on linux programmers or lack of investment on the part of the manufacturers.
"So, it would appear that neither profit nor good will is available by doing Linux support for laptops. The users don't want to pay for it, and they'll hate you anyway."
Actually, I hope that is part of the lesson any other OEMs take away from this. Because I don't what their "support". I'm not interested in buying a computer which reverts into a silicon-and-plastic doorstop every time I don't get "support". I am interested in buying a computer *and never talking to the seller again*. To do that I need the specifications to all hardware to be published, so I or others can write drivers to use them how we please.
I don't want Dell to "support" my laptop or my desktop (I'm typing on a Dell Dimension XPS T450, on which I have never managed to get the sound card working, because it uses some built-in-to-the-motherboard non-published specs piece of shit). I just want them to sell a piece of hardware fully described so I know what I am getting and I can use it.
The interface to the winmodems are most often not documented. Thus it would be a great deal of work to write a driver for it -- you would have to reverse engineer it, probably starting with looking at the binary-distribution-only windows driver.
Some winmodems (Lucent I know, maybe others) have driver written by the manufacturer of the winmodem. But Lucent ( please correct me if I am wrong ) doesn't distribute the source to the driver, just a binary module that you can use with a particular kernel.
This traps you. This is the reason not to buy secret, undocumented crap in the first place. If you get a lucent winmodem, then you are always dependent on lucent deciding it is worth their time to compile up the module for the version of linux you want to use. You are probably out of luck runing the Herd or *BSD or BeOS or whatever. *You don't really own it, because you can't do whatever you want with it.*
When I plunk my money down and buy something, I want to *own* it. I don't want a pair of apron strings tying me back.
The problem with a lot of the capitalists in the world today is that they don't have the balls to be real capitalists. These loosers can't just fscking *sell* something. They want to rent the right to use it for only one purpose. They want to construe a meaningless EULA to be a contract between you and them, in which you somehow decide to pay them to restrict your behavior. Heck, even in real estate, all the developers are itching to slap on their deed restrictions or get their subdivisions under some zoning or neighborhood association.
Almost the only people doing real business any more are gun manufacturers. You have to respect someone who still has the balls to sell a piece of hardware and declare that whatever you do with it afterwards is your responsibilty.
About the shrink wrap license -- do they attempt to get around the "I bought it second-hand at a garage sale" defense ? I.E., does the license deny you the right to re-sell the DVD player or mandate that you keep the license intact when you transfer it ?
Not that I think that these shrink-wrap licenses are ever going to stand up in a higher court. But they do give the big money a tool to harass people with until they can afford to appeal up to an educated judge.
Could one of the people who actually bought one of these devices post the text of the claimed license ?
"Legally, the keys and encryption are (i believe) the intelectual property of the MPAA(or someone related)."
This is meaningless. Everything he worked with was sold to him; those keys were somehow mixed in on the disk and player, which he bought. Figuring out the key on the disk is no different than processing any other legally acquired copyrighted material -- is it against the law for me to count the number of words in the paperpack I just purchased, or otherwise analyze it ?
If I apply some stylometry techniques to some of the junk paperbacks out there and discover that one of those prolific authors is actually four or five, can I be sued for revealing the secret ? No, because if the publisher didn't want me to look at the book, they shouldn't have taken my money and given me the book. If the DVD producer had a secret, then they shouldn't have put it on hardware and disks that they sold all over the world. Instead they wanted to both have a secret and share it, and their math wasn't clever enough. Not our problem.
You say:
"It's like breaking into someone's home (by whatever means, violent or nonviolent) to steal or copy something of theirs that you feel you should have."
It's not like that at all. Some moneyd interests might approve of you saying so, since they want people to feel guilty about cracking those keys, but he didn't go to anyone else's home; he was in his own home, with property he purchased legally. In what way did he steal anything ? Immitation is not stealing.
You say:
"Since they're still the creaters/owners of the encryption, it's their right to determine who has access to the keys."
Setting aside for the moment how you can possibly own an encryption, I'd like to point out they sold him those keys on the disk and hardware.
Now, they didn't count on him being able to read it. But that is simply a bad business break. You can't expect the courts to go around throwing people in jail everytime some little piece of information makes your business plan out of date.
You say:
"And while their not including Linux does suck (i agree!), how do you feel that what you did for DeCSS is justified?"
Why does he have to justify anything ? If he took that damn player out in the woods and blew it to pieces with a shotgun just for jollies, well, it's his player. Instead he looked at it and figured out a bit about how it worked, and told some other people. What's wrong with that ? If the MPAA wants an uncopiable medium, why don't they make one and sell it ? Ok, so they did try, but they missed. Is that reason to take your lumps and try again, or is that reason to run to the government that always takes your soft money campaign contributions and demand that they start throwing people in jail ?
You discribe the actions of the movie industry as "people protecting their right to protect what they made."
This has nothing to do with the issue at hand. If a company wants to make DVD's, and keep them, DeCSS does not help me break into their storehouse and take them. DeCSS, a DVD burner, or any other method of copying DVDs doesn't enable me to take away any other person's (or company's) property.
What it does do is make those DVD disks more useful than the producers had planned on. But that is their mis-step. Either they can suck it up, or stop selling DVDs and take the sales hit. But they can't go running to the courts to stop people from using the DVD disk however they please. It's not their DVD disk anymore, *because they sold it*.
Suppose someone sold printers that automatically broke after 2,000 pages. Suppose I looked inside and set some pin to ground or whatever and it now lasted until it physically wore out. Is that illegal ? No, because it was my printer after I bought it, and I can modify it as I please. That kid in Norway bought a computer and a DVD drive and a DVD, and whatever he did with them is completely OK, and he should be able to tell other people about what he did.
There is nothing wrong with cracking copy protection on software, writing code to access *your* hardware anyway you please, or looking under the hood of that Camaro to disable the 115 mph cut-out on the carburator.
To put it shortly, the movie industry wanted to sell a non-replicable disk, and it turned out they were accidently producing replicable disks. They decided to try to use the courts to make up for their technological miscalculation rather than go back to the drawing board.
In other more mature industries, companies usually expect and anticipate that people will try to find out how to get more use than out of products. Texaco isn't suing because because we have found out how to drive further on a gallon of their gas.
The DVD producers have two choices: 1) suck it up, 2) stop selling DVDs and invent another scheme.
I bought an HP LaserJet 5L at a swap meet a while back. It prints text just fine, which is actually what I want it for, but it messes up on postscript. After some investigation we decided either the memory or the postscript chip was bad. ( PS pages had stripes that were shifted and smeared to the right. My friend used gs to send the postscript commands for a diagonal line, and that also was messed up. )
I was told by someone that you can buy a replacement postscript chip for around $25 from hp's web site. I searched but couldn't find it. Where would I get a new postscript chip ? I haven't even taken the thing apart yet to check if you really can replace the PS chip. Has anyone else done this ?
I investigated WAP a month ago in relation to an idea I had.
It took me about ten minutes to dismiss them. If you attempt to download the technical specifications from www.wapforum.org, it presents you with a license agreement, just to look at the specifications. They call that open ? It irritates me when companies put these EULA which they know are unenforceable in court on packaging. But a supposedly open non-profit body ?
It is also not clear what part the EULA restrained me beyond what the law does anyway. It says that you agree not to violate their copyright. The whole thing gave me the creeps -- should I pick through the fine print to find where they hid the real restriction in all the clauses saying that I couldn't do things I can't do anyway ?
The whole thing had a juvenile feel to it, as if the people making the web site just did not have a cultural background in the industry. I don't know how to express this well, but an EULA for a technical publication is the type of thing a shiny-faced freshman or high school entrepreneur wannabee would come up with.
They may or may not have some good standards and technology, but I won't be associating any business efforts with them. I never read the technical documents. If they have something to publish, then why don't they just publish it, like the rest of the world ?
Slashdot Mirror
Here's my spam filter, which is implemented in procmail and gnus, which I inherited from a person I used to work with:
-- First check that it is addressed to a mailing list you subscribe to
-- Then check to see if it explicitly lists your address on the To: line
You will see that a majority of spams are not directly addressed to the target on the To: line.
But the problem is not whether something can get 90% of all spam. A variety of rules can be composed as you suggest to block a high percentage. It is the false alarms of these rules -- the emails that are blocked that aren't spam. I put all the spam in separate folder, and yet I still sometimes get mail there that I wanted to go to one of my regular folders. What is the false alarm rate of your system ?
I think the validation of the sending email address is a great idea. You can almost never respond to unwanted email, but you can respond to legitimate email.
Still, ultimately the spammers will find a way through. They are human and I think they will always be smarter than a structured filter system. Just look at the newest tactic -- it's called "viral marketing" -- it basically amounts to rewarding your friends for spamming you.
That said, it is not true that "if they don't do everything _legal_ to up their earnings, he stockholders can and will sue." The management of a company is simply not under obligation to explore every possible legal source of revenue. Can I sue Ford because they don't write software, or GM for not owning gold mines ? If I buy Redhat stock and they don't drill for oil, I was not injured in any way because I knew they weren't in the oil business when I bought the stock.
The responsibility of the mangement to the stockholders is merely to not fail to do something a stockholder would naturally expect without telling them first. You can't start a company with a nice looking published business plan and have a real secrete plan of paying yourself a salary to do nothing until the company folds.
There would be no reason to uphold a suit against Amazon for failing to pursue the patent, especially if they had declared publicly that they were not going to pursue software patents. A good court would hold that the stockholders can sell their stock if they don't like it. A good court would see itself as a protector of the stockholder from deceit and neglience. The market will take care of incompetence; the court just has to assure the information is not kept from the market.
Think about it this way: the ability of stockholders to take their money and run makes the stockmarket a very efficient judge of business strategies. Do you really think that a judge and jury can do a better job ? Resort to the courts may be justified when extreme and rare plots and colusions attempt to hender the efficiency of the market.
There existed no reasonable expectation that Amazon should have a business plan based on the bureaucratic incompetence of the PTO. Amazon could have choosen to explicitly deny themselves that path.
But they didn't. Jeff Bezos and his company have choosen a business plan which has as a primary feature the denial of my freedom. Their plan is make sure no one else (including me) can make a one-click shopping site, and they plan to live off of this government enforced monopoly. If that doesn't make Amazon the Antichrist, it at least puts them on the same team.
I loose freedoms every time the PTO gives out a patent, but much of the time I at least get something back -- a business that would not have been worth trying without the monopoly guarantee is attempted, bettering the economy on the whole. However, Jeff Bezos and Amazon are just a tax. If we did not award business method and software patents, someone still would be selling books online with one click, and I would still have a tiny piece of freedom that Jeff Bezos wants from me.
In previous slashdot spam discussions www.privatecitizen.com has been recommended as an excellent guard against unwanted phone and snail mail advertisements. I don't use it ( I see using the court system to go after these people as only slightly less objectionable than using the legislature), but perhaps someone who does can comment on it's effectiveness and whether it would guard against cell phone mail spam ? It seems that privatecitizen depends on being able to distribute a list to known advertisers, and I think that many spammers ( wireless and regular ) are much more fly-by-night types.
Several times in this thread Europeans have jumped in saying that they have to pay to download spam because of non-free local calls. But it is technically possible to make a good spam defense even without having to download the entire spam:
- Get a shell account and read your mail on the server, downloading only what is displayed to you. Of course, you pay for the connection while you stare at messages and compose them, so you might look into something more sophisticated, such as . . .
- Get a shell account and split the spam with procmail or the equivalent (on the server), and set up something like fetchmail to only get the important stuff, plus a log file of all the headers/subjects of the spam, just in case an important one slipped through the filter and you want to actually look at it. Or . . .
- Have a program that talks directly to a POP server once you are connected, that downloads just the headers and subjects, starts downloading everything obviously not spam in the background while presenting you with a list of of everything so you can select things to download and read and things to delete unread.
Why haven't European (or American, for that matter) ISPs already provided this as a service to differentiate them from their competitors ? Why haven't any of the free software people provided the same, just as they provided junkbuster ?I suspect it is because annoyance at spam is not as widespread as a vocal minority would have us believe. It is just not that big a problem in the larger scheme of things. Otherwise someone would have already written the program I listed last above and they would be making money going to ISPs and integrating it with the little custom windows dialers and email clients.
Web banner ads are more annoying and take up time right when you are trying to actually do something (look at a web page), so fairly effective filters came out quickly. But I suspect that most people also stay on longer than necessary just to download their mail, because they briefly check the slashdot headlines for example, and their mail can download in the unused bandwidth while they browse.
I am afraid that we will let government regulation do it's usual heavy-handed solution that will only stop 50% of the problem anyway, rather than picking a technical solution which involves less emotionally gratifying yelling (and slashdot posts) and would solve 80% or 90% of the problem. If non-download filters were common and the default on ISP services, response rates to spam would drop.
Rather than making a disk image, I think you can use the command dump ( see "man dump" ) to put a giant file on a cd which you will restore to the disk with the dump command.
To get a target machine to the point where you can mount a CD and run dump, I suggest checking out some of the linux floppy distributions.
There is one other option. There used to be a web site that offered you all the choices offered in RedHat's install script, and then created a custom install floppy which would automatically install according to those specifications, no human intervention necessary. So you could just go to this site (or it's equivalent for slack or whatever other distribuiton you want, or download whatever tool the site is running and use it yourself) and then make a boot install disk and just shove it and the cdrom in every machine. I think the boot floppy/dump from cdrom method will work faster.
If I buy three wireless LAN equiped laptops (such as those ibooks or PCs with cards, whatever) and arrange them so that laptop A is at one end of a very long room, and laptop B is at the other end of the room, so that B and A *cannot* communicate, but I place C in the middle such that B and C *can* talk and A and C *can* talk -- does the obvious thing happen ?
That is, does C begin re-broadcasting packets going between A and B so that they get to right place ?
As I understand things, this is a software issue -- do the machines recognize when to configure themselves as routers ? If only some OS's or drivers do it, which ones ? If I install linux on an ibook or PC with the wireless LAN card, can I make linux do the right thing ?
If there were two wireless ethernet devices on the machine in the middle, I think I could figure out how to make linux do this. But can linux re-route packets through a single interface device ? I know it is possible to re-route things heard on one device to be broadcast on another, since that is what my IP-masqing linux box connected to cable modem on one card and the apartment-mates machines on the other does.
ZoomAir has something they call "AcessPoint Software" which does this on a machine with one wire card and one wireless card. Can this software be tricked to use a single wireless card as a bridge between two networks ? Furthermore, can you configure it to automatically recognize the situation in which it should start routing, and stop when it shouldn't ?
I've been doing google searches -- does anyone have any URLs of where to start ?
Why buy old technology that just works when you can spend a lot more for a wireless ethernet card in each machine and be totally cool ?
One machine can have one of those obsolete non-wireless cards in it, so you can plug into your clients hopelessly out of date non-wireless network, and route through that machine.
I'm actually saying this only half tongue-in-cheek. If you have the cash to be carting around multiple laptops and a palm, you can afford it. It depends on the business you are in, but the coolness factor of that might get you enough new business to pay for itself the first time out.
I totally agree.
But sometimes they want a final version of the document in Word to deliver to people who have no idea how to deal with anything else.
It's not something that I can do anything about.
In my opinion and experience, CASE tools are an abomination that merely serves to produce some eye-candy along with the code to distract everybody from worrying about whether or not the stuff works.
I know that in some areas they are very well entrenched, such as in some areas of the aerospace industry. I think it must be related to the way projects that have a high cost of failure become highly rigidized and over-managed, instead of simply being over-engineered and receiving the best brains.
In my opinion, the decision to use a CASE tool or the standardization of a project on a particular IDE is the first sign of ultimate failure, or at the very least, a sign that you are going to spend a lot of money to get bloated half-working mediocrity (which passes for success in some places).
I will ignore the side issue of whether collaborative document development without extensive communication through a side channel (telephone, email, face-to-face converstations) is a good idea in the first place.
I have worked on several different projects in which we did or attempted to do collaborative document editing. Usually it starts out with one person maintaining an official web site version and merging in everyone's changes, and then it devolves into a token-ring-like situation with emailed attachments being mailed around to everyone clogging mail systems and confusing the issue who has the official version.
Just use LaTeX and cvs. The people who do word won't like it, but tell them tough. They get used to it surprisingly quickly, especially if you tell them that no-one should be worrying about margin widths or other formatting issues until it is to the point of one person editing it.
At that point you can convert it to Word. Converting LaTeX to word goes surprising fast and easy.
S(h)elf Help Guide - The Smart Lifter's Handbook by Gabor Caime and Gabor Ghone
(It says, "If you paid for this book, you might learn something! If not, take it back -- you don't need it." Probably commercially smarter advice than "steal this.")
Check it out here:
I think the point is to know about these things before the deal is closed and it's all over the newspapers.
Also, I believe the US intelligence community has many people who are assigned to a particular region or country and come to work every day to read all the newpapers from there, listen to TV and Radio news, educate themselves tecjhnically on the major economic industries, and otherwise be Johnny-know-it-all for every that has anything to do with that area.
If you have problems with ext2 you might want to look at some of the links posted here:
2 .shtml
http://slashdot.org/askslashdot/00/01/22/195821
Why is the root of this thread marked offtopic ? I'll have to start meta-moderating more regularly.
I have never had data loss with linux unless it was me inadvertently deleting something. I loose my machine to power outages at least once a month, and once when netscape was open it came back with an empty bookmarks file, but the whole bookmarks file was in /lost+found. I also regularly start huge compile runs on my laptop, put it in my backpack, and let it compile until the battery runs out. When I plug it in it fsck's on the way up, but I just restart the compile and it is fine. So I don't have your experience of loosing data on a power outage.
I recently bought a cheap big IDE drive for my home machine. I plan to have a couple of partitions for experimenting with ext3 and other journaling filesystems. But I don't really ever stress the ext2 filesystem, so all I can test is that it is as good as ext2. How would you test a filesytem for reliability and crash resistence ?
Another question: I have noticed that access from linux to a FAT32 partition is much slower than from windows to a FAT32 partion. Linux FAT32 access is also much slower than Linux ext2 access. I would expect some difference -- who wants to spend a lot of time making the FAT32 fs code fast -- but the difference seems rediculous, at least 2 times slower. Why is this ?
"And make sure the cleaning people don't unplug your server to plug in a vacuum cleaner.(or is that an urban legend?)"
I think what Paul is refering to here is not the fact that the late night staff occasionally unplug machines (an environmentally conscious janitor were I work used to turn my linux PC off -- surely similar things happen everywhere) but a particular old 60s or 70s anecdote about an IBM mainframe that always mysteriously and unexplicably crashed on a particular day of the week, at a certain time at night; and a support team at IBM was so baffled that they actually went to the site one night and sat down to watch it happen; and the janitor entered, unplugged the machine and plugged in the vacuum. It turned out he always vacuumed that room on that day of the week.
I believe this story is in the New Hacker's Dictionary, the most recent edition. Unforntunately my copy is at home so I can't consult it.
It's a neat anecdote when well told. It would be interesting to know if it were true or urban legend.
I think that sounds a little overly discouraging. I would say that the asker is probably much better equiped to learn about and set up Linux than most new users.
Rather, I would advise that you just check out a linux installation that is as standard as possible. See if that is reliable enough, and don't experiment with any new filesystems unless plain vanilla won't cut it.
Also, you said it was an "NT Server". Could you give more details on what it is serving ? If it is just diskspace to windows client machines, it should be easy to configure.
The Resurrecting Duckling referred to above is here (first link on the google search):
g .html
http://www.cl.cam.ac.uk/~fms27/duckling/ducklin
It didn't seem like a "great paper" to me. It had all the ideas that an afternoon a brainstorming with some reasonably intelligent folk should produce. And when they go off on that biological metaphor of the hardware being the body and software the soul etc -- and then they start making analogies about a single body having a succession of souls because you reprogrammed it -- that's just hippie talk. It reminds me of why I must stop myself from ever going to graduate school: when I am in a very overworked and sleep-deprived state and people start distracting my precious few neurons with meaningless strings of attention-grabbing words I want to kill them, and thus loose what little productivity I still had.
But even if these guys are just a couple of flaky pot-smoking vegetarian hippie coffee house denizens, they do have some interesting references at the bottom. I'll be checking these out on my next trip to Barker library.
I like the emacs gnus newsgroup reader. I also use it for mail on some of my accounts, and on my main account I have procmail directing my mailing lists into spools that I read with gnus.
Gnus just rocks.
I would really like a gnus interface to slashdot. Maybe some elisp hacker could use the emacs web borwser (w3) and link them together. Among the benfits that would bring would be the ability to compose posts in emacs, so I could use a spell checker.
I would read so much more of slashdot in so much less time. In fact, if I sat down and did that task in week or whatever it would take me, I bet I could make up the time in three years.
Other sites, such as arstechnica (their html sucks watermelons through a garden hose) and wired (same class) and nytimes (slightly better) could benefit from this.
Could you have access to the cookies and deal with them in elisp ? Because then you could write little elisp functions which would cycle the nytimes account randomly through a list of fake accounts you built up, putting noise in their database of what articles you visit.
"For one, it appears that the community expects the work to be done for no money."
."
It was "the community" which did the bulk of the work here, not Dell.
"They scream because the machine isn't cheaper with Linux on it, even though the amount of work to be done to get Linux to work decently with laptops is considerable, far more than with desktops . .
If they would publish the specifications to the parts of the laptops, they wouldn't have to do any work, because the linux people would have drivers out before their slow-moving corporate programmers could assign a project number to the job.
" . . . and support costs can be expected to be higher (simply because there are far fewer experts on Linux-laptop issues than for Windows-laptop issues)."
On the other hand, a given linux user is far more likely to also be an "expert" (i.e., not want or need support) than a given windows user.
"Second, the community screams because support for all platforms isn't instantly available, even though many laptop components don't have Linux drivers present."
While I shouldn't speak for the rest of the community, what burns me is that they use those laptop components which have secret specifications; this is what generates the lack of linux drivers, not the lack of effort on linux programmers or lack of investment on the part of the manufacturers.
"So, it would appear that neither profit nor good will is available by doing Linux support for laptops. The users don't want to pay for it, and they'll hate you anyway."
Actually, I hope that is part of the lesson any other OEMs take away from this. Because I don't what their "support". I'm not interested in buying a computer which reverts into a silicon-and-plastic doorstop every time I don't get "support". I am interested in buying a computer *and never talking to the seller again*. To do that I need the specifications to all hardware to be published, so I or others can write drivers to use them how we please.
I don't want Dell to "support" my laptop or my desktop (I'm typing on a Dell Dimension XPS T450, on which I have never managed to get the sound card working, because it uses some built-in-to-the-motherboard non-published specs piece of shit). I just want them to sell a piece of hardware fully described so I know what I am getting and I can use it.
The interface to the winmodems are most often not documented. Thus it would be a great deal of work to write a driver for it -- you would have to reverse engineer it, probably starting with looking at the binary-distribution-only windows driver.
Some winmodems (Lucent I know, maybe others) have driver written by the manufacturer of the winmodem. But Lucent ( please correct me if I am wrong ) doesn't distribute the source to the driver, just a binary module that you can use with a particular kernel.
This traps you. This is the reason not to buy secret, undocumented crap in the first place. If you get a lucent winmodem, then you are always dependent on lucent deciding it is worth their time to compile up the module for the version of linux you want to use. You are probably out of luck runing the Herd or *BSD or BeOS or whatever. *You don't really own it, because you can't do whatever you want with it.*
When I plunk my money down and buy something, I want to *own* it. I don't want a pair of apron strings tying me back.
The problem with a lot of the capitalists in the world today is that they don't have the balls to be real capitalists. These loosers can't just fscking *sell* something. They want to rent the right to use it for only one purpose. They want to construe a meaningless EULA to be a contract between you and them, in which you somehow decide to pay them to restrict your behavior. Heck, even in real estate, all the developers are itching to slap on their deed restrictions or get their subdivisions under some zoning or neighborhood association.
Almost the only people doing real business any more are gun manufacturers. You have to respect someone who still has the balls to sell a piece of hardware and declare that whatever you do with it afterwards is your responsibilty.
About the shrink wrap license -- do they attempt to get around the "I bought it second-hand at a garage sale" defense ? I.E., does the license deny you the right to re-sell the DVD player or mandate that you keep the license intact when you transfer it ?
Not that I think that these shrink-wrap licenses are ever going to stand up in a higher court. But they do give the big money a tool to harass people with until they can afford to appeal up to an educated judge.
Could one of the people who actually bought one of these devices post the text of the claimed license ?
You say:
"Legally, the keys and encryption are (i believe) the intelectual property of the MPAA(or someone related)."
This is meaningless. Everything he worked with was sold to him; those keys were somehow mixed in on the disk and player, which he bought. Figuring out the key on the disk is no different than processing any other legally acquired copyrighted material -- is it against the law for me to count the number of words in the paperpack I just purchased, or otherwise analyze it ?
If I apply some stylometry techniques to some of the junk paperbacks out there and discover that one of those prolific authors is actually four or five, can I be sued for revealing the secret ? No, because if the publisher didn't want me to look at the book, they shouldn't have taken my money and given me the book. If the DVD producer had a secret, then they shouldn't have put it on hardware and disks that they sold all over the world. Instead they wanted to both have a secret and share it, and their math wasn't clever enough. Not our problem.
You say:
"It's like breaking into someone's home (by whatever means, violent or nonviolent) to steal or copy something of theirs that you feel you should have."
It's not like that at all. Some moneyd interests might approve of you saying so, since they want people to feel guilty about cracking those keys, but he didn't go to anyone else's home; he was in his own home, with property he purchased legally. In what way did he steal anything ? Immitation is not stealing.
You say:
"Since they're still the creaters/owners of the encryption, it's their right to determine who has access to the keys."
Setting aside for the moment how you can possibly own an encryption, I'd like to point out they sold him those keys on the disk and hardware.
Now, they didn't count on him being able to read it. But that is simply a bad business break. You can't expect the courts to go around throwing people in jail everytime some little piece of information makes your business plan out of date.
You say:
"And while their not including Linux does suck (i agree!), how do you feel that what you did for DeCSS is justified?"
Why does he have to justify anything ? If he took that damn player out in the woods and blew it to pieces with a shotgun just for jollies, well, it's his player. Instead he looked at it and figured out a bit about how it worked, and told some other people. What's wrong with that ? If the MPAA wants an uncopiable medium, why don't they make one and sell it ? Ok, so they did try, but they missed. Is that reason to take your lumps and try again, or is that reason to run to the government that always takes your soft money campaign contributions and demand that they start throwing people in jail ?
You discribe the actions of the movie industry as "people protecting their right to protect what they made."
This has nothing to do with the issue at hand. If a company wants to make DVD's, and keep them, DeCSS does not help me break into their storehouse and take them. DeCSS, a DVD burner, or any other method of copying DVDs doesn't enable me to take away any other person's (or company's) property.
What it does do is make those DVD disks more useful than the producers had planned on. But that is their mis-step. Either they can suck it up, or stop selling DVDs and take the sales hit. But they can't go running to the courts to stop people from using the DVD disk however they please. It's not their DVD disk anymore, *because they sold it*.
Suppose someone sold printers that automatically broke after 2,000 pages. Suppose I looked inside and set some pin to ground or whatever and it now lasted until it physically wore out. Is that illegal ? No, because it was my printer after I bought it, and I can modify it as I please. That kid in Norway bought a computer and a DVD drive and a DVD, and whatever he did with them is completely OK, and he should be able to tell other people about what he did.
There is nothing wrong with cracking copy protection on software, writing code to access *your* hardware anyway you please, or looking under the hood of that Camaro to disable the 115 mph cut-out on the carburator.
To put it shortly, the movie industry wanted to sell a non-replicable disk, and it turned out they were accidently producing replicable disks. They decided to try to use the courts to make up for their technological miscalculation rather than go back to the drawing board.
In other more mature industries, companies usually expect and anticipate that people will try to find out how to get more use than out of products. Texaco isn't suing because because we have found out how to drive further on a gallon of their gas.
The DVD producers have two choices: 1) suck it up, 2) stop selling DVDs and invent another scheme.
You say "Hardly anyone uses ssh 2.x because of it's licencing issues."
What licensing issues ? How does the ssh1.2 license differ from the ssh2 license ?
I bought an HP LaserJet 5L at a swap meet a while back. It prints text just fine, which is actually what I want it for, but it messes up on postscript. After some investigation we decided either the memory or the postscript chip was bad. ( PS pages had stripes that were shifted and smeared to the right. My friend used gs to send the postscript commands for a diagonal line, and that also was messed up. )
I was told by someone that you can buy a replacement postscript chip for around $25 from hp's web site. I searched but couldn't find it. Where would I get a new postscript chip ? I haven't even taken the thing apart yet to check if you really can replace the PS chip. Has anyone else done this ?
I investigated WAP a month ago in relation to an idea I had.
It took me about ten minutes to dismiss them. If you attempt to download the technical specifications from www.wapforum.org, it presents you with a license agreement, just to look at the specifications. They call that open ? It irritates me when companies put these EULA which they know are unenforceable in court on packaging. But a supposedly open non-profit body ?
It is also not clear what part the EULA restrained me beyond what the law does anyway. It says that you agree not to violate their copyright. The whole thing gave me the creeps -- should I pick through the fine print to find where they hid the real restriction in all the clauses saying that I couldn't do things I can't do anyway ?
The whole thing had a juvenile feel to it, as if the people making the web site just did not have a cultural background in the industry. I don't know how to express this well, but an EULA for a technical publication is the type of thing a shiny-faced freshman or high school entrepreneur wannabee would come up with.
They may or may not have some good standards and technology, but I won't be associating any business efforts with them. I never read the technical documents. If they have something to publish, then why don't they just publish it, like the rest of the world ?