Actually SCO said this was a syn flood, which means it IS as simple and download and run. However, I don't believe them because asking your ISP to filter your webserver's IP over a SYN flood is pointless and stupid. You either implement anti-syn-flood measures (syncookies or some firewall based option), or you wait it out. When the flood stops, your server works again. Asking your ISP to filter just prolongs the outage.
Naw. Its XP now, at least according to this assessment based on web stats.
That only measures machines that are on the internet, and Windows 95 stats are not available there. After 'fixing' their stats, 39.5% of all Windows machines run XP, and 23.2% run 98. That still leaves 37.3% of machines that run 'other' Windows OS's. I think we can assume that NT4 and 2000 are not well represented there, and are only really used within businesses. That leaves 95, and Me, to take a large chunk of that 37%. I'm also willing to bet that a higher percentage of machines running XP are connected to the internet than those running older microsoft OS's.
Now, I wish I had some hard data to back any of this up, but the closest I can come up with is part of a Microsoft announcement where they brag XP is the best selling version of Windows ever. From this article, they state, "NPDTechworld noted that XP sold 400,000 copies at retail in October 2001 and 250,000 copies in November 2001. These numbers compare with the 580,000 and 350,000 copies that Win98 sold in its first and second months, respectively." Now, Microsoft's claims in this were taken with a grain of salt because Microsoft used the fact these licenses were sold to OEMs and corporate clients, rather than that the OEMs had sold these licenses to customers. Personally, I feel that the retail numbers are a more accurate representation than the number of licenses commited by OEMs.
I'm willing to concede that Windows XP is more popular than Windows 98+95, however, but I bet the numbers are pretty close.
So when did you last check? In 1999? You are wrong, mistaken and misrepresent the facts.
There were plenty of things wrong with the poster's statements, but that wasn't one of them. The majority of PCs out there ARE running Windows 95 or Windows 98/98SE. This was the magical point where "just about everyone who wanted to own a PC, owned a PC". Microsoft may not sell either one anymore, bt this doesn't change the fact the installed base for these systems is HUGE, and much bigger than the installed base for Windows Me and Windows XP.
Well, funny thing, my version of linux 1.4 doesn't work that well with my new hardware--oh but wait, I'm supported with a new version. Of course windows 98 doesn't work on the new machines--its 20 year old technology.
There never was a Linux 1.4 and there never will be. However, Linux kernel 2.0, which was released in 1996, is still usable on some current systems, and Linux kernel 2.2, released Jan 1999, is usuable on most current systems. There are exceptions, of course, such as laptops, but laptops are generally hopelessly tied to whatever version of Windows was supplied with it.
There's a real simple reason why Win98 is being EOLed -- money. It costs money to support each and every operating system they make. The Sun/Java thing is a convenient excuse for something they would've done anyway in a year or two anyhow. The OS is over five years old now, after all. Lets just hope a remotely exploitable hole isn't found in Win95/98 that can be used for worm propagation. 'Luckily,' most of the time, such bugs only blue screen the poor user's machine.
Interesting choice of companies. The low-end networking equipment market is healthy exactly because there's a lot of competition. If I don't like Linksys, I can buy D-Link, Netgear, SMC, OvisLink, Microsoft, 3COM, or any one of a dozen other manufacturers. If I find one manufacturer's product isn't working up to my expectations, ripping out one and replacing it with something else from some other manufacturer means little downtime, and no costs to 'migrate'. These are commodity products.
The high-end CRM market is a different creature. There is competition, but not enough. If Oracle isn't treating me right, I have alternatives, but I likely already gave Oracle a lot of money, and moving to another vendor's product will be expensive. I need to move all my data from one to the other (and hope it works). I would need to retrain the entire staff on the new package. Unfortunately, this 'lock-in' means CRM packages don't work as commodities.
They've already entered several counterclaims. No sense having two trials when one will suffice, right? IBM has claimed Breach of Contract, Lanham Act Violation, Unfair Competition, Intentional Interference with Prospective Economic Relations, Unfair and Deceptive Trade Practices, Breach of GNU General Public License, Promissory Estoppel (in regards to the GPL) Copyright Infringement, four counts of Patent Infringement, and is seeking a Declaratory Judgment. Should be enough to bury SCO, eh?
I believe IBM has indicated that if SCO does not produce the documents IBM wants soon, IBM will ask for a dismissal of SCO's claims. After all, if SCO won't tell IBM what IBM has done to harm SCO, what case does SCO have against IBM?
Personally, I like what Linus had to say about this in an Infoworld interview.
"I'm a big believer in copyrights," Torvalds wrote in an e-mail interview. "Of all the intellectual property (laws), copyright... is the only one that is expressly designed so that individual people can (and do) get them without having scads of lawyers on their side."
"If Darl McBride was in charge, he'd probably make marriage unconstitutional too, since clearly it de-emphasizes the commercial nature of normal human interaction, and probably is a major impediment to the commercial growth of prostitution," he wrote.
Actually, Apache runs 67.41% of all domains, versus IIS running 23.46% of them. I remember one Netcraft Survey where they said more than 50% of machines are running IIS/Windows. What you can gather from this is far more Apache machines run one or more domains than IIS machines do.
Yes, and from what I understand, despite collecting money for three years, they still haven't PAID artists anything (they claim they started early this year). They did, however, fly all the CPCC members over to France to attend a conference on music piracy. The copyright board sets the tarif each year after only consulting with the CPCC, which ALWAYS overstates their position. CPCC believes that 60% of all CD-Rs are used for copying music, and bases all tarif calculations on this number. The CPCC also has adjusted their payment schedule to reduce the amount of money that is paid out to authors and publishers, and increased the amount that goes to record companies and performers.
Re:20 years and a little analogy to biology
on
20 Years of Virii
·
· Score: 1
Jumping the gun a little bit, aren't we? The parent didn't say anything about Windows [in]security. He merely pointed out that reactionary anti-virus updates are not working.
"...what they don't see is that we're trying to defend the rights of capitalism for the silent majority. Linux leaders pound the table about a community-driven model where everything is free, and that's the flip side of capitalism. This country was founded on capitalism, the right to make a profit for what you own, not give it away for free."
[November 18 2003]
I spent all that time looking for the words "communism", "unamerican", etc, that I forgot to look for the obvious, "captialism". In that comment, he didn't use the 'C'-word, but otherwise he did say that they were communists.
You are quite right. He's never explicitly said the word communist. In fact, there's only one article that ever links him to that word, and it's not even a direct quote of anything. He does, however, like to wrap his entire lawsuit in the American flag. He likes to claim that Free and Open Source Software will destroy the American economy (I'm getting real tired of that complaint from everyone who has a problem with new technology), and millions of jobs. Everyone will be poor, and all IT jobs will be exported to India, China, and other foreign countries (Darl needs a reality check on this -- it's happening NOW). His retoric is that FOSS developers are communists or socialists, but explicitly avoids saying it because it would be a little too hard to take it seriously. The entire "GPL preempted by the constitution" and "they're violating export control laws" in their legal claims exist for no other reason than to make these developers look like very bad people who are trying to subvert the USA into something bad.
If you're contracting work out to third parties, you should not be surprised when something like this happens. The third party contractor will try to reduce any cost he can, and if that means reusing code, he will.
The reviewer was comparing it to commericial anti-spam software that was either auto-updating or click-and-update. He also missed the fact that SpamAssassin can use 'signatures' via DCC or pyRazor. I'm not sure what "Proprietary Methods" are supposed to be, feature-wise, though. In short, he configured it to use the built-in keyword heuristics only, and unsurprisingly was disapointed in the results.
Award was bought by Phoenix a long time ago (and IMO, when they did, Award BIOS's took a turn for the worse). AMI BIOS based machines are today, few and far between. I'm not sure why Award BIOS's were so popular, back in the day, but when Phoenix bought Award, they became the biggest BIOS vendor instead. I get the feeling that Award BIOS 6.0 was really just Pheonix BIOS in disguise (instead of the old two column list menu, it tries to pretend it's a dropdown menu).
Re:debian is a truly great distribution...
on
Debian 3.0r2 Released
·
· Score: 3, Informative
Well, part of the reason Debian has so many old packages has to do with the number of architectures it runs on. If the package can't be compiled and run on Alpha, ARM, IA-32, IA-64, HPPA, M68K, big endian MIPS, little endian MIPS, PPC, s390 and SPARC, it doesn't get in. There are exceptions for things that are really arch specific, but for most cases, if you want it in Debian, it needs to run on every platform Debian is available for.
Another factor is packaging. Debian packages are built in a specific fashion, and in some cases when the upstream developer releases a new version that is binary and source incompatible with the old version, the package developer goes to great pains to make sure both packages can co-exist on the same system.
Then there's version stability. When a security hole is found in a Debian package, Debian doesn't just package up the latest version and ship that like some vendors do. Instead, the security fixes are backported to the previous version, and an update to the old version is released instead. Why would they do this? A new version with new features can have new bugs, or change the behaviour of certain things in various (sometimes subtle) ways. I've had an entire PHP-based website stop working because of a PHP upgrade. Something that was legal and worked fine in a previous version (storing objects in an array stored in the SESSION variable) completely ceased functioning in the new version.
Re:debian is a truly great distribution...
on
Debian 3.0r2 Released
·
· Score: 2, Insightful
Unstable is usable, but sometimes dealing with the package system gets to be a problem. I've had several cases when I had to uninstall a package I had installed because an update to a related package, and a file from the first package had migrated to the second, and apt/dpkg couldn't handle it automagically.
Re:Howto's..
on
Linux in 2004?
·
· Score: 2, Informative
Howtos will likely remain forever... I'd be fairly happy to see the howtos for rather mundane things go away or at least lose some of their utility. Today, things like the CDROM HOWTO, Ethernet HOWTO and ISP-Connectivity HOWTO are unnessesary for most people. Other things like the ADSL HOWTO, CD Writing HOWTO and DVD HOWTO are still nessesary evils.
Isn't it obvious? IBM MUST be paying Ransom Love to say this! I mean, isn't EVERYONE who says something bad about SCO on their payroll? SCO just can't stop them! They just keep stealing all the precious SVR4 code, and putting it in everything from Linux to BSD! And the SCO executives now are so scared for their lives, that they need to hire bodyguards to protect them from the evil Linux hippies!!!!
Paranoia is a disease that just progressively gets worse.
Uhm, right. Just talk to Sarah Ward, who was accused of sharing 2,000 songs on KaZaA with a maximum penalty of $300,000,000. Oh, except there's a few problems. For one, she's a Macintosh user (Kazaa only runs on windows), and a 66 year old sculptor. Not willing to fully back down, attorneys for the RIAA members reserved the right to harass the woman in future.
Or there's Ross Plank who was accused of being a big trader of latin music. Except, he doesn't speak spanish, and doesn't particularily like latin music anyway.
The problem with all of this is, the RIAA is bringing civil lawsuits against these people, which means you can either hire a lawyer, which will cost you more than the settlement, or you can just pay the settlement. And guess what! You're not even "innocent until proven guilty" in a civil trial. All the RIAA needs to prove is "more likely than not".
We have other names for this kind of behaviour like extortion. Do you think most people can afford skip work to appear in court for four or more months, and pay a lawyer to defend them? It's easy to get people to settle when the cost of fighting it would break them. It's truely a sad system, when accepting a guilty verdict is cheaper than fighting for your innocence.
By "BSD snobs", I meant the people who constantly on any story that mentions that SCO is suing IBM over Linux, say, "Well, you should just use BSD instead, because there's no infringing code in our beloved project". I'm talking about the opportunists, who, just like Sun and Microsoft, try to take advantage of SCO's lies to benefit themselves. I know full well that most BSD users and developers would not, and do not stoop to such lows.
Perhaps I should've been clearer. There are plenty of Linux snobs who do the exact same thing when it comes to negative stories about Windows, and it's just as silly.
Well, at least the BSD snobs can stop saying, "Use BSD instead because SCO has no claim over that." I think most BSD folk believed this was going to happen someday, and were just hoping that it wouldn't. SCO believes everyone owes them a cut, because no OS could exist without precious code from SVR4. I wouldn't be surprised if SCO threatens all the embedded OS makers and Microsoft before this is over.
As a side benefit, the bigger and more grandious SCO's claims become, the less believable they become, and sooner or later the press will pick up on this.
He's already taken home almost $10 million of SCO's money, and a bunch of SCO stock. Even if he loses, he's still winning. And he did win the antitrust lawsuit against Microsoft, it's just the government decided that a promise from Microsoft to police itself from now on, and to promise not to be bad anymore was good enough.
Slashdot Mirror
Actually SCO said this was a syn flood, which means it IS as simple and download and run. However, I don't believe them because asking your ISP to filter your webserver's IP over a SYN flood is pointless and stupid. You either implement anti-syn-flood measures (syncookies or some firewall based option), or you wait it out. When the flood stops, your server works again. Asking your ISP to filter just prolongs the outage.
Now, I wish I had some hard data to back any of this up, but the closest I can come up with is part of a Microsoft announcement where they brag XP is the best selling version of Windows ever. From this article, they state, "NPDTechworld noted that XP sold 400,000 copies at retail in October 2001 and 250,000 copies in November 2001. These numbers compare with the 580,000 and 350,000 copies that Win98 sold in its first and second months, respectively." Now, Microsoft's claims in this were taken with a grain of salt because Microsoft used the fact these licenses were sold to OEMs and corporate clients, rather than that the OEMs had sold these licenses to customers. Personally, I feel that the retail numbers are a more accurate representation than the number of licenses commited by OEMs.
I'm willing to concede that Windows XP is more popular than Windows 98+95, however, but I bet the numbers are pretty close.
There's a real simple reason why Win98 is being EOLed -- money. It costs money to support each and every operating system they make. The Sun/Java thing is a convenient excuse for something they would've done anyway in a year or two anyhow. The OS is over five years old now, after all. Lets just hope a remotely exploitable hole isn't found in Win95/98 that can be used for worm propagation. 'Luckily,' most of the time, such bugs only blue screen the poor user's machine.
The high-end CRM market is a different creature. There is competition, but not enough. If Oracle isn't treating me right, I have alternatives, but I likely already gave Oracle a lot of money, and moving to another vendor's product will be expensive. I need to move all my data from one to the other (and hope it works). I would need to retrain the entire staff on the new package. Unfortunately, this 'lock-in' means CRM packages don't work as commodities.
They've already entered several counterclaims. No sense having two trials when one will suffice, right? IBM has claimed Breach of Contract, Lanham Act Violation, Unfair Competition, Intentional Interference with Prospective Economic Relations, Unfair and Deceptive Trade Practices, Breach of GNU General Public License, Promissory Estoppel (in regards to the GPL) Copyright Infringement, four counts of Patent Infringement, and is seeking a Declaratory Judgment. Should be enough to bury SCO, eh?
I believe IBM has indicated that if SCO does not produce the documents IBM wants soon, IBM will ask for a dismissal of SCO's claims. After all, if SCO won't tell IBM what IBM has done to harm SCO, what case does SCO have against IBM?
Actually, Apache runs 67.41% of all domains, versus IIS running 23.46% of them. I remember one Netcraft Survey where they said more than 50% of machines are running IIS/Windows. What you can gather from this is far more Apache machines run one or more domains than IIS machines do.
Yes, and from what I understand, despite collecting money for three years, they still haven't PAID artists anything (they claim they started early this year). They did, however, fly all the CPCC members over to France to attend a conference on music piracy. The copyright board sets the tarif each year after only consulting with the CPCC, which ALWAYS overstates their position. CPCC believes that 60% of all CD-Rs are used for copying music, and bases all tarif calculations on this number. The CPCC also has adjusted their payment schedule to reduce the amount of money that is paid out to authors and publishers, and increased the amount that goes to record companies and performers.
Jumping the gun a little bit, aren't we? The parent didn't say anything about Windows [in]security. He merely pointed out that reactionary anti-virus updates are not working.
But what do I know -- I own a PS2 instead, and it certainly doesn't win any beauty contests.
You are quite right. He's never explicitly said the word communist. In fact, there's only one article that ever links him to that word, and it's not even a direct quote of anything. He does, however, like to wrap his entire lawsuit in the American flag. He likes to claim that Free and Open Source Software will destroy the American economy (I'm getting real tired of that complaint from everyone who has a problem with new technology), and millions of jobs. Everyone will be poor, and all IT jobs will be exported to India, China, and other foreign countries (Darl needs a reality check on this -- it's happening NOW). His retoric is that FOSS developers are communists or socialists, but explicitly avoids saying it because it would be a little too hard to take it seriously. The entire "GPL preempted by the constitution" and "they're violating export control laws" in their legal claims exist for no other reason than to make these developers look like very bad people who are trying to subvert the USA into something bad.
If you're contracting work out to third parties, you should not be surprised when something like this happens. The third party contractor will try to reduce any cost he can, and if that means reusing code, he will.
The reviewer was comparing it to commericial anti-spam software that was either auto-updating or click-and-update. He also missed the fact that SpamAssassin can use 'signatures' via DCC or pyRazor. I'm not sure what "Proprietary Methods" are supposed to be, feature-wise, though. In short, he configured it to use the built-in keyword heuristics only, and unsurprisingly was disapointed in the results.
Award was bought by Phoenix a long time ago (and IMO, when they did, Award BIOS's took a turn for the worse). AMI BIOS based machines are today, few and far between. I'm not sure why Award BIOS's were so popular, back in the day, but when Phoenix bought Award, they became the biggest BIOS vendor instead. I get the feeling that Award BIOS 6.0 was really just Pheonix BIOS in disguise (instead of the old two column list menu, it tries to pretend it's a dropdown menu).
Another factor is packaging. Debian packages are built in a specific fashion, and in some cases when the upstream developer releases a new version that is binary and source incompatible with the old version, the package developer goes to great pains to make sure both packages can co-exist on the same system.
Then there's version stability. When a security hole is found in a Debian package, Debian doesn't just package up the latest version and ship that like some vendors do. Instead, the security fixes are backported to the previous version, and an update to the old version is released instead. Why would they do this? A new version with new features can have new bugs, or change the behaviour of certain things in various (sometimes subtle) ways. I've had an entire PHP-based website stop working because of a PHP upgrade. Something that was legal and worked fine in a previous version (storing objects in an array stored in the SESSION variable) completely ceased functioning in the new version.
Unstable is usable, but sometimes dealing with the package system gets to be a problem. I've had several cases when I had to uninstall a package I had installed because an update to a related package, and a file from the first package had migrated to the second, and apt/dpkg couldn't handle it automagically.
Howtos will likely remain forever... I'd be fairly happy to see the howtos for rather mundane things go away or at least lose some of their utility. Today, things like the CDROM HOWTO, Ethernet HOWTO and ISP-Connectivity HOWTO are unnessesary for most people. Other things like the ADSL HOWTO, CD Writing HOWTO and DVD HOWTO are still nessesary evils.
Yes, and many of them end with them finally asking someone else, and the other person pointing it out within 10 seconds.
Paranoia is a disease that just progressively gets worse.
Or there's Ross Plank who was accused of being a big trader of latin music. Except, he doesn't speak spanish, and doesn't particularily like latin music anyway.
The problem with all of this is, the RIAA is bringing civil lawsuits against these people, which means you can either hire a lawyer, which will cost you more than the settlement, or you can just pay the settlement. And guess what! You're not even "innocent until proven guilty" in a civil trial. All the RIAA needs to prove is "more likely than not".
We have other names for this kind of behaviour like extortion. Do you think most people can afford skip work to appear in court for four or more months, and pay a lawyer to defend them? It's easy to get people to settle when the cost of fighting it would break them. It's truely a sad system, when accepting a guilty verdict is cheaper than fighting for your innocence.
Perhaps I should've been clearer. There are plenty of Linux snobs who do the exact same thing when it comes to negative stories about Windows, and it's just as silly.
As a side benefit, the bigger and more grandious SCO's claims become, the less believable they become, and sooner or later the press will pick up on this.
He's already taken home almost $10 million of SCO's money, and a bunch of SCO stock. Even if he loses, he's still winning. And he did win the antitrust lawsuit against Microsoft, it's just the government decided that a promise from Microsoft to police itself from now on, and to promise not to be bad anymore was good enough.