Your profession is to interfere in market transactions to persuade participants to make irrational decisions in your favor. That's defective and antisocial behavior.
Keep in mind that many folks, like you, say that exploiting flaws in human psychology or knowledge is a Bad Thing. A good number of those go on to support cracking ("but the system *allowed* it!), which seems to me hypocritical.
Keep in mind that a major reason the West is wealthy is not because of natural resources (which are scattered all over the world). It's not because of knowledge (there are smart people all over the world with decent education). It's not even all because of the willingness to use military power in nasty ways. No, the West got mass media early on, learned how to market very, very well, and proceeded to drive up demand for western goods.
As a result, you make many, many times what a worker in the Phillipines makes.
Catfish thinks Goldfish has been brought back to life from the tape you made. Catfish is a goddamn moron.
Why? Catfish has a simple brain, and is not capable of understanding video recording technology. In the world as perceivable by Catfish, Goldfish *is* alive. There is no difference to Catfish whether or not Goldfish is, from our point of view, alive or not.
You may be quite dubious about this, but suppose I have a sheet of paper. On it are a whole set of two-dimensional organisms. They have evolved to work in two dimensions alone -- they cannot percieve a third dimension. Furthermore, their brains have evolved to think about two dimensional systems. Every now and then, an organism is deleted, vanishing from the paper. If I occasionally lift an organism organism from the paper, and move it to a second sheet, it is not possible for the remaining organisms to distinguish between a movement and death. Furthermore, these creatures would not understand such a concept. They might be able to develop systems to deal with such a concept (as we can with mathematical systems where some of our rules of logic do not apply), but they have no way of understanding it.
We are a series of continually changing patterns. Because it's useful, we call a series of similar patterns over time an entity -- a person, a piece of grass, whatever. We have an idea of an "I" beyond the simple mechanics of being, also because it's useful. And so, a system where a pattern's next phase can be produced a thousand miles away or right here violates a set of models about how entities exist that we've produced over time. It doesn't work -- we say that the entity is "killed" or similar. The problem is that we then move beyond the realm of definition and try to derive meaning from our definition. We say that because something doesn't fit our model, it will affect something beyond our experience and understanding -- a soul, or the experience of the "teleported" creator, or what have you.
Arguments about what "me" is are pointless (at least as far as trying to derive meaning from the definition), because they're arguments about definitions, not about the nature of reality. You want to call "me" only "me" if I'm a mattern in approximately the same place "me" was half a second ago, or "me" "me" if I'm a pattern a thousand light years away? Fine, do either.
We commonly subscribe to a model where entities cannot "fork". Nothing wrong with that most of the time, much like there weren't any problems with Newtonian physics...most of the time.
Lots of people want a tiny phone and a separate PDA with large screen and built in keyboard.
Also, a lot of people want privacy. By the nature of the current (reasonably efficient) model, cell phones constantly send out a heartbeat signal. This allows tracking of a personal on a national scale easily. There is no native way to establish anonymous connections, so a history of all your calls is available. There is no support for end-to-end encryption (and significant federal opposition to such), so all your conversations can be monitored from the telco. Cell encryption to and from the telco's cell tower has had a long, ugly history of severe flaws (and in Europe, deliberate back doors in the form of key escrow). Finally, G3-class phones support a remote-turn-on feature without a visual "on" indicator (intended for law enforcement use) that allows remote eavesdropping.
Given these constraints (and the fact that I find it irritating to constantly be at the beck and call of others), I intensely dislike the idea of carrying a cell phone. Yes, you can turn them off, but then you're open to the question "why did you turn off your cell" and having people request that you keep yourself available.
Finally, there's the expense, the care you have to take not to damage them (I once was with a group of people that spontaneously decided to go swimming at the ocean and ended up destroying a number of cell phones), the hassle of understanding a compressed voice, the hassle of dealing with signal reception, the hassle of dealing with spam, the hassle of having an extra lump in your pocket, annoying people trying to sell you plans, the constant temptation to sign onto "services" rather than buying one-time products, etc, etc.
I realize that for some folks, there is really no option. If you're running a small consulting business, you may need to be available 24/7, and don't have an option to *not* be on-call and still be competitive.
VoIP may offer an eventual solution if someone develops a free decentralized VoIP network that catches on.
Or you will. IBM, Compaq, DEC, SGI? All got hammered by selling high-end machines and being undercut by competent competition that sold at much lower prices.
The computer world has not been gentle to people trying to maintain high prices.
but remember that the money comes out of the hands of idiots who buy stocks at grotesquely inflated prices. It's at least a slight improvement.
Well...sort of. Remember that not all (actually, I believe the majority of) investments are not direct -- a group, like a mutual fund or an investment bank is managing someone else's money. So you could say that it's the fault of the original fund contributors and they weren't paranoid enough, but I 'm not sure that such an expectation is reasonable.
It's well known that MS uses Unix stuff in it's operating systems - the BSD TCP/IP stack for instance. And these 21$ million might seem much to you but for a company with such huge revenues like MS these are just peanuts.
That's irrelevant. You're trying to mash a bunch of unrelated facts together.
SCO is claiming IP issues with Linux, not BSD. Microsoft used code (which may not even exist in the kernel any more, for all we know) from BSD in the TCP stack. Furthermore, I've yet to hear even any insinuations that the TCP stack contains any infringing code, and find it extremely unlikely.
And these 21$ million might seem much to you but for a company with such huge revenues like MS these are just peanuts. If you finish you education some day and get a job in big business, you'll see that such stuff are really minor contracts.
It's really annoying number of folks on Slashdot like to try to sound sophisticated by pish-poshing large sums of money. Microsoft is *not* in the practice of doing this. At both of the large tech companies I'm familiar with, it's standard procedure to fight almost *anything* WRT IP claims (Intel, for instance, will fight claims that want more than $100K), and as I pointed out above, any claims would be phenomenally tenuous and extremely unlikely to stick. If you're a tech company flush with money and you don't have a policy of fighting any lawsuits aimed at you, you'll be stripped of your money in no time.
A good example is the recent fight between that little company that patented web browser plugins and Microsoft. The patent is pretty straightforward and clearly covers IE -- it's a far more reasonable IP claim than the SCO business. However, Microsoft is fighting it tooth and nail.
Yes, Microsoft is funding SCO. Yes, they are doing it because it's a cost-effective way of fighting Linux.
The problem is that from a long-term standpoint, Verisign is in a lousy situation.
While it's easy for them to "innovate" and abuse their monopoly (and if the CEO can pull it off, he's doing the right thing from a business standpoint), it's also quite easy to invent and come up with a new name resolution system. This is a facinating research area, and there are a lot of possibilities.
The original DNS system was designed for a far slower network that was mostly trusted. It was designed to be fairly light weight (and originally wasn't particularly scalable).
A new naming system would be a good thesis project. Furthermore, the FSF controls the name resolution system (via glibc) on a rather large number of machines, and has a significant amount of clout in this area. (Microsoft controls most of the rest, but given that Sitefinder primarily competes with MSIE's lookup system, it's likely that they'd support such a move).
Remember Unisys, the GIF and the PNG. In the end, the silver lining was greater than the cloud caused by Unisys's abuse. There was a new, wonderful, well-designed image format available.
If Verisign's actions produce a modern, robust, featureful and secure naming system, I will have considered the whole furor well worth the cost.
First of all, I'd like to say that Mandrake themselves is not griping about it. You are. While your $120 contribution may be laudable and good, it does not entitle you to ask everyone else to do the same. Furthermore, few companies are capable of surviving on charity alone -- some non-profits. I think that you're somewhat annoyed at the loss of $120, and expect that your actions would somehow be universalized. If you contribute $n, then so will everyone else. And that isn't reasonable.
Second, I don't use Mandrake (I prefer Red Hat). However, Mandrake has their *own* debt to the community. Their product is mostly based on packaging -- a relatively cheap task. They are freely provided with huge amounts of software to make money with. I have contributed code and expertise to a number of software packages that are included in the Mandrake distribution. Why should I also be expected to pay Mandrake money, on top of all this? I'm sorry if it's difficult for Mandrake to find a good business model to work with open source -- but it's their responsibility to do so, not ours. I believe that one can be had without the need to beg OSS folks to do foo or not to do bar. RH is in no financial trouble, unlike RH. IBM has already made its Linux investment back and more.
Third, Mandrake has made a decision to make ISOs available. I do not believe that they have clauses preventing redistribution. Their Mandrake Club product is providing a service -- namely, a fast download site to obtain Mandrake. Their service has been made somewhat obsolete by recent technological developments -- the design of BitTorrent, which makes distribution of large files much more feasible and affordable. If they wanted to, they *could* limit reproduction of large chunks of their product. They do not need to have a GPLed installer, for instance. This would allow folks to redistribute source packages, but not copy full ISOs. However, for various reasons (including maintaining community goodwill -- as evidenced by Caldera, extremely important with open source software), they have decided to allow reproduction. I have no compuctions about taking them up on their offer. They were not required to make that offer.
Again, I would like to point out -- Mandrake is a Linux company. This does not equate to "eating children", however. Let us assume your worst case scenerio -- that Mandrake goes under as a result of this. Red Hat is doing well, and Debian is happily motoring along with no commercial backing (I don't know about other distributions, though I don't know of anything other than SuSE that have had financial problems.) Linux and Linux distributions will march on, no matter what, even if Mandrake falls. Linux is here to stay.
What proof do you have that Microsoft is behind the SCO situation?
Actually, Microsoft isn't behind the SCO situation. The primary player in the SCO situation is Linus.
As many have pointed out over the years, Linus is a very smart guy. He also has a ton of influence over the psyche of the Linux movement. So, about a year ago, Linus had a conversation with McBride that looked something like this:
Torvalds: Hey, McBride. I've got a great proposition for you.
McBride: I'm interested.
Torvalds: How about you claim that you've got a death grip on Linux, and start firing off nonsensical lawsuits? This will depress the stock prices of the various Linux businesses. Then, I buy back into some of them, we drop the whole business, and I ride the wave back up. If anyone just ignores you, I'll be sure to act really worried about the lawsuit. It'll be great PR!
McBride: Hmmm..ingenius. I like it. But what's in it for me?
Torvalds: That's the beauty of it. While you're shooting off all these lawsuits, SCO's stock price will be shooting up. You buy stock beforehand, and cash out while the thing is elevated, before it comes crashing down again.
McBride: Fantastic. I'm in, Linus!
You can be damned sure that Linus's dark hand is indeed in part of this if you see Linus purchasing *any* sort of portion of *any* Linux-related company over the next year.:-)
By "time until fixed in the source tree", I'm just pointing out that Microsoft may take months to roll out a patch to users in a hotfix or service pack.
Also, to be fair, I suspect that few users immediately apply kernel patches in the Linux world. They wait until RH's up2date or Debian's apt-get sucks down the latest and greatest. A fair comparison should might say "Microsoft does not attempt to supply a 'rapid-release' patch for technical users at all, unlike the Linux community. However, it's time-to-Joe-end-user-release is comparable to that of Red Hat." or something along those lines.
I certainly feel that, at least applying the immediately obvious and most useful criteria, Microsoft does *not* fix bugs (release patches) more quickly than the Linux community.
*I* think he's referring to the time until a bug gets fixed in the source tree. Furthermore, what measures as a "bug" may differ. Many "bugs" in open source software are feature requests, etc. The measurement may be from the time that Microsoft filter personnel classify something as a "bug" (i.e. a reported severe security hole may sit around for months before it reaches developers flagged as a MUSTFIX bug with MS's internal bug tracking system, for all I know). Finally, security-related bugs may differ in fix time from ordinary bugs ("this icon should be moved over by two pixels to the right"). This doesn't mean that Bill's wrong, but it certainly doesn't tally with my past experience, and given that this is currently a senstive Microsoft weak point, I'd be expecting at least a bit of coloring of the truth.
Actually, I'd say that this is more reasonably taken from a pessimistic view. *16%* of all CIOs were negatively influenced (seems unlikely that it's a positive influence) toward Linux because of this one SCO money grab.
(a) A couple places, like Jerusalem, are a big deal to both sides. Land is not fungible, in this case.
(b) (Assuming you're from the US) What if the Soviets came, took over, kicked you out of your house, and moved you up to cold Canada? Would you be thrilled at the idea of renaming Canada "The United States of America", and the exiles stop complaining and being resentful?
I've always wondered why we spend tons of tax money and incur a whole lot of international ill will by supplying Israel with military hardware and backing them in many unpopular moves.
I think it has to come down to one of two possibilities.
The first is the one that the Arabic countries around Israel like pushing -- that wealthy, influential Jews in the United States are able to put pressure and affect enough US leadership that Israel gets backing. While I'm sure this plays a role, I doubt that it's the primary cause.
What *does* seem a bit more plausible is that the only way the US can control the Middle East is if it's divided and fragmented, unable to use the oil supply as a weapon, and always providing us with at least one ally in the area for military staging purposes. US Middle Eastern foreign policy has seemed to focus on keeping the Middle East divided for *decades*. Israel is a wonderful divider, particularly because they're *dependent* upon US backing for their continued existance, so is certain to help out the US in the Middle East.
The first possibility is disturbing, as it means that US citizens are not properly being served by their government. The second shows the US being uncomfortably mercenary...but I suppose that somebody has to produce the economic imbalances that keep me living the good life, so I'm much less inclined to complain about politicians playing hardball on my behalf.
(a) Use kerberized FTP. Kerberos is a bitch to set up, but if your sysadmin is paranoid about security, he should be using it. Kerberos just deals with authentication, so it's possible to use non-encrypted systems that still use Kerberos authentication.
(b) Why is it "too slow"? A modern system running AES can saturate a 100Mbits/sec network.
Apparently, the article author, "Daniel Lyons", has had a wonderful time writing anti-OSS articles for some time. His articles generally at least touch on both sides, but his conclusion -- his closing words -- are reliabily anti-OSS, though certainly not fanatically so.
Take a look at his take on the the SCO lawsuit, his opinion on whether Linux will succeed, and an article criticizing IBM for not indemnifying Linux users.
The GPL did not come along and ambush these companies - they CHOSE to make use of GPL software. So tough-titty to them.
And this isn't just a matter of "choosing to use GPLed software", a la Linux on the desktop. They incorporated extremely large quantities of code from software that they did not own (and would cost hundreds of millions of dollars to develop in the commercial world) into their own product.
I'd like to see a company sign a MS shared-source agreement and try using Microsoft kernel source, and see how gently Microsoft treats them. Microsoft would *love* having a huge club like that. Yes, they probably wouldn't demand removal. They'd just use it as a lever to force Cisco into signing other useful agreements. Overall, I think Cisco would likely be hurt more in the long term by stealing Microsoft source than GPLed source.
I disagree. I really think that if Microsoft was doing so, they'd publically take a stand (a la "OSS is viral"), rather than paying off journalists. The thing that drove this article is the same thing that drove pro-Linux hype a couple years ago, before Linux was ready for desktop transitioning -- business journalists are terrified of being behind the curve, and are constantly trying to find something new that they can give as good tips to their readers.
I do think that I can comfortably claim that this article has little value from a useful content perspective. Linksys management almost certainly did not decide to use GPLed software -- some subcontractor put it in. Aside from the fact that it's easier to get ahold of GPLed source than source to most non-GPL software (and thus GPLed software and temptation might lead to more violations), there isn't really much causality with the GPL. Some engineer at Linksys (or some Indian subcontractor...there are good engineers in India, but in the current environment, there are also a *lot* of people that simply swipe code and let the company get screwed down the line) swiped some code.
I don't buy into the whole "a non-GPL company would just demand moderate royalties, but the FSF requires you to remove the software" claim. If you're Cisco (or IBM, or Oracle) and a company discovers that you're bundling its software in your products, you are going to get burned. Badly. There is simply too much potential money involved. No lawyer is going to let you off nicely. At that point, you are up shit creek, and the degree to which you are going to get screwed is fairly academic.
Slashdot has been bitterly complaining about SCO for some time -- a company that (well, in its own eyes) doesn't have GPLed source, and discovered (or at least wants to convince others) that it has its source in other, major products. Sorry, but $699/copy is not trivial.
Come to think of it, SCO licensing fees would also apply to the routers in the article. Funny how that works, eh?:-) They're large enough that they'd drive the product out of the market.
The article implied that using GPLed software immediatelye exposed one to such potential liability. I hope that no IT manager is going to hold off moving to Linux on the desktop because of random IP worries.
The fact that the FSF does not accept royalties could be a reasonable complaint for some companies that have deliberately or otherwise stolen FSF software. However, if you aren't maintaining enough control over your software to avoid having mass chunks of source that you haven't checked out the license on incorporated into your product, you have severe legal problems -- forget worrying about the GPL.
Furthermore, I severely doubt that Linksys made a measured decision to use GPLed software, or at least a decision that was properly passed through legal. I cannot understand how a corporate lawyer would sign off on incorporating GPLed code into non-GPLed code. I have a sneaking suspicion that the work was done by a subcontractor or an engineer that wasn't being reviewed by legal.
Finally, I think that, as per the article, if the FSF is simply giving the option of removing the code (with no damages, which I find extremely unlikely), that's more than generous. There have been many, many, many Linksys routers shipped. A company could sue for phenomenal back damages.
You worry that MS could change the interface and break compatibility with Mono. And yet, many, many vendors produce DBMSes with an SQL interface. SQL is horribly, horribly extended by all sorts of vendors, and has truly awful compatibility. However, it's a decent bit of work, and the abovementioned issues haven't stopped postgresql and mysql from rapidly consuming large chunks of the market.
I doubt Mono could ever be in a situation as bad as SQL is. The compatibility issues there are almost overwhelming, and yet I find it very fortunate that the MySQL and Postgresql developers did *not* decide to give up on SQL because someone else already had a closed SQL-using implementation.
I wish I knew whether you were the real Miguel. Slashdot is loaded with fake ones.:-(
"hasuseraclue()"...Note: by reverse engineering the code, we know that above system call return 0 when ran on the system of the author of the previous paper.
Actually, this is one of the funniest and most germane comments you could have made. Think about it for a moment. Traditionally, MS library calls return TRUE for success and FALSE for failure. Traditionally, Unix library calls return the opposite -- 0 for success and non-zero (well, negative) for failure. You don't have a problem with MS-produced APIs and so wrote a comment that the other person would consider a compliment -- because presumably he adheres only to Unix APIs.:-)
In a world where Mono is vastly successful, if Microsoft changes/introduce new APIs, do you think it will matter?
While I wish you luck and respect your work (if you are, indeed, the real Miguel), I think that you may have missed a point. Your argument that Mono will be a binding influence on Microsoft is predicated on the fact that Mono will achieve significant "market" share, before any sort of nasty compatibility-splitting moves are made by Microsoft. I'm not sure Mono will do that. Microsoft has a tremendous team of language designers that had a head start of *years* on you folks. It takes a while to catch up, and given that this is still a new, sexy initiative at Microsoft, it's going to be a moving target.
However, I really don't see the problem as all that severe anyway -- I'd support your argument for entirely different reasons. (a) Presumably Mono can be made to work on Windows, providing a common runtime everywhere. So if Microsoft changes things, there will still be a runtime that C#-using developers could ship with their applications. (b) Even if Microsoft changes course...so what? How would the situation be worse than before? We already don't have binary compatibility between the C Unix applications written to an entirely different set of APIs and the MFC-based Windows software. The only different thing would that there would be a fairly solid development tool available to to rapid application development on Linux -- and a tool that would make it easier for experienced C# developers to transition to Linux.
But let's assume that Mono *will* hurt Linux. I'd like to point out to the grandparent poster that Miguel is hardly under any onus to drop a project because it isn't in Linux's immediate favor, anyway. Open Source is all about writing good software and then sharing it with other interested people, folding improvements back in. Miguel's doing exactly that -- providing a good tool. He's done immense amounts of work in the past that have enormously benefited the OSS and Linux worlds -- likely more than the grandparent poster.
Finally, Java is, frankly, not a particularly well designed language. It was well marketed, and it managed to win over a large chunk of the C++ crowd, and is an improvement over earlier languages for folks who want a C-like safe language. However, many of the design decisions were poor. Furthermore, the JVM has some fundamental design limitations that prevent it from being useful for some non-Java languages like ocaml. I believe that this is not the case for the.NET VM. C# is supposed to (as may be obvious, I'm not a C# user) fix a number of issues with Java.
Miguel, I have my differences with some design in GNOME -- I think that, in retrospect, CORBA and XML were overly heavyweight. I'm sad that the decision was made not to include an barber pole-like "infinite progress bar" widget in GNOME to denote tasks with an unknown completion time. However, it's the fact that I can pick such minor nits that makes it really clear how much your work in the past has done for us now. I can sit down and work on an Excel spreadsheet. My Windows-using friends can comfortably use Linux.
Thanks again. And whatever happens, don't wind up like Eazel.
Now, what games for the GameCube use bump mapping? Nyo-ho!
The GameCube doesn't have support for hardware bumpmapping. There isn't any hardware that isn't being used -- it's just CPU cycles that are going to other things.
The surround sound...all right, I suppose that you could argue that it's not being "pushed to the limit".
I don't believe the GC *has* a higher resolution mode, though I can't swear to it.
As for the steel backplate in the thing: it's known that consumers generally associate some heft to objects with higher quality, to the point that some products are weighted to take advantage of this.
I have to disagree with you there. OEM-supplied components *are* quite flimsy, since they compete on essentially nothing but price (well, perhaps extra volume buttons on the keyboard might be a selling point).
There are a number of manufacturers that produce more highly-price, nicer input devices. Among them, Logitech, which sells a whole line of wireless keyboards, PCKeyboard, which has among its offerings modern versions of old-IBM-Model-M style buckling-spring mechanism keyboards (CLICK CLICK CLICK...), Fingerworks' unusual offerings, the expensive-but-ultimate-RSI-avoidance Kinesis keyboards, and Goldtouch's ergo products.
Slashdot Mirror
Your profession is to interfere in market transactions to persuade participants to make irrational decisions in your favor. That's defective and antisocial behavior.
Keep in mind that many folks, like you, say that exploiting flaws in human psychology or knowledge is a Bad Thing. A good number of those go on to support cracking ("but the system *allowed* it!), which seems to me hypocritical.
Keep in mind that a major reason the West is wealthy is not because of natural resources (which are scattered all over the world). It's not because of knowledge (there are smart people all over the world with decent education). It's not even all because of the willingness to use military power in nasty ways. No, the West got mass media early on, learned how to market very, very well, and proceeded to drive up demand for western goods.
As a result, you make many, many times what a worker in the Phillipines makes.
Catfish thinks Goldfish has been brought back to life from the tape you made. Catfish is a goddamn moron.
Why? Catfish has a simple brain, and is not capable of understanding video recording technology. In the world as perceivable by Catfish, Goldfish *is* alive. There is no difference to Catfish whether or not Goldfish is, from our point of view, alive or not.
You may be quite dubious about this, but suppose I have a sheet of paper. On it are a whole set of two-dimensional organisms. They have evolved to work in two dimensions alone -- they cannot percieve a third dimension. Furthermore, their brains have evolved to think about two dimensional systems. Every now and then, an organism is deleted, vanishing from the paper. If I occasionally lift an organism organism from the paper, and move it to a second sheet, it is not possible for the remaining organisms to distinguish between a movement and death. Furthermore, these creatures would not understand such a concept. They might be able to develop systems to deal with such a concept (as we can with mathematical systems where some of our rules of logic do not apply), but they have no way of understanding it.
We are a series of continually changing patterns. Because it's useful, we call a series of similar patterns over time an entity -- a person, a piece of grass, whatever. We have an idea of an "I" beyond the simple mechanics of being, also because it's useful. And so, a system where a pattern's next phase can be produced a thousand miles away or right here violates a set of models about how entities exist that we've produced over time. It doesn't work -- we say that the entity is "killed" or similar. The problem is that we then move beyond the realm of definition and try to derive meaning from our definition. We say that because something doesn't fit our model, it will affect something beyond our experience and understanding -- a soul, or the experience of the "teleported" creator, or what have you.
Arguments about what "me" is are pointless (at least as far as trying to derive meaning from the definition), because they're arguments about definitions, not about the nature of reality. You want to call "me" only "me" if I'm a mattern in approximately the same place "me" was half a second ago, or "me" "me" if I'm a pattern a thousand light years away? Fine, do either.
We commonly subscribe to a model where entities cannot "fork". Nothing wrong with that most of the time, much like there weren't any problems with Newtonian physics...most of the time.
Lots of people want a tiny phone and a separate PDA with large screen and built in keyboard.
Also, a lot of people want privacy. By the nature of the current (reasonably efficient) model, cell phones constantly send out a heartbeat signal. This allows tracking of a personal on a national scale easily. There is no native way to establish anonymous connections, so a history of all your calls is available. There is no support for end-to-end encryption (and significant federal opposition to such), so all your conversations can be monitored from the telco. Cell encryption to and from the telco's cell tower has had a long, ugly history of severe flaws (and in Europe, deliberate back doors in the form of key escrow). Finally, G3-class phones support a remote-turn-on feature without a visual "on" indicator (intended for law enforcement use) that allows remote eavesdropping.
Given these constraints (and the fact that I find it irritating to constantly be at the beck and call of others), I intensely dislike the idea of carrying a cell phone. Yes, you can turn them off, but then you're open to the question "why did you turn off your cell" and having people request that you keep yourself available.
Finally, there's the expense, the care you have to take not to damage them (I once was with a group of people that spontaneously decided to go swimming at the ocean and ended up destroying a number of cell phones), the hassle of understanding a compressed voice, the hassle of dealing with signal reception, the hassle of dealing with spam, the hassle of having an extra lump in your pocket, annoying people trying to sell you plans, the constant temptation to sign onto "services" rather than buying one-time products, etc, etc.
I realize that for some folks, there is really no option. If you're running a small consulting business, you may need to be available 24/7, and don't have an option to *not* be on-call and still be competitive.
VoIP may offer an eventual solution if someone develops a free decentralized VoIP network that catches on.
With time, perhaps they'll learn.
Or you will. IBM, Compaq, DEC, SGI? All got hammered by selling high-end machines and being undercut by competent competition that sold at much lower prices.
The computer world has not been gentle to people trying to maintain high prices.
but remember that the money comes out of the hands of idiots who buy stocks at grotesquely inflated prices. It's at least a slight improvement.
Well...sort of. Remember that not all (actually, I believe the majority of) investments are not direct -- a group, like a mutual fund or an investment bank is managing someone else's money. So you could say that it's the fault of the original fund contributors and they weren't paranoid enough, but I 'm not sure that such an expectation is reasonable.
It's well known that MS uses Unix stuff in it's operating systems - the BSD TCP/IP stack for instance.
And these 21$ million might seem much to you but for a company with such huge revenues like MS these are just peanuts.
That's irrelevant. You're trying to mash a bunch of unrelated facts together.
SCO is claiming IP issues with Linux, not BSD. Microsoft used code (which may not even exist in the kernel any more, for all we know) from BSD in the TCP stack. Furthermore, I've yet to hear even any insinuations that the TCP stack contains any infringing code, and find it extremely unlikely.
And these 21$ million might seem much to you but for a company with such huge revenues like MS these are just peanuts. If you finish you education some day and get a job in big business, you'll see that such stuff are really minor contracts.
It's really annoying number of folks on Slashdot like to try to sound sophisticated by pish-poshing large sums of money. Microsoft is *not* in the practice of doing this. At both of the large tech companies I'm familiar with, it's standard procedure to fight almost *anything* WRT IP claims (Intel, for instance, will fight claims that want more than $100K), and as I pointed out above, any claims would be phenomenally tenuous and extremely unlikely to stick. If you're a tech company flush with money and you don't have a policy of fighting any lawsuits aimed at you, you'll be stripped of your money in no time.
A good example is the recent fight between that little company that patented web browser plugins and Microsoft. The patent is pretty straightforward and clearly covers IE -- it's a far more reasonable IP claim than the SCO business. However, Microsoft is fighting it tooth and nail.
Yes, Microsoft is funding SCO. Yes, they are doing it because it's a cost-effective way of fighting Linux.
The problem is that from a long-term standpoint, Verisign is in a lousy situation.
While it's easy for them to "innovate" and abuse their monopoly (and if the CEO can pull it off, he's doing the right thing from a business standpoint), it's also quite easy to invent and come up with a new name resolution system. This is a facinating research area, and there are a lot of possibilities.
The original DNS system was designed for a far slower network that was mostly trusted. It was designed to be fairly light weight (and originally wasn't particularly scalable).
A new naming system would be a good thesis project. Furthermore, the FSF controls the name resolution system (via glibc) on a rather large number of machines, and has a significant amount of clout in this area. (Microsoft controls most of the rest, but given that Sitefinder primarily competes with MSIE's lookup system, it's likely that they'd support such a move).
Remember Unisys, the GIF and the PNG. In the end, the silver lining was greater than the cloud caused by Unisys's abuse. There was a new, wonderful, well-designed image format available.
If Verisign's actions produce a modern, robust, featureful and secure naming system, I will have considered the whole furor well worth the cost.
First of all, I'd like to say that Mandrake themselves is not griping about it. You are. While your $120 contribution may be laudable and good, it does not entitle you to ask everyone else to do the same. Furthermore, few companies are capable of surviving on charity alone -- some non-profits. I think that you're somewhat annoyed at the loss of $120, and expect that your actions would somehow be universalized. If you contribute $n, then so will everyone else. And that isn't reasonable.
Second, I don't use Mandrake (I prefer Red Hat). However, Mandrake has their *own* debt to the community. Their product is mostly based on packaging -- a relatively cheap task. They are freely provided with huge amounts of software to make money with. I have contributed code and expertise to a number of software packages that are included in the Mandrake distribution. Why should I also be expected to pay Mandrake money, on top of all this? I'm sorry if it's difficult for Mandrake to find a good business model to work with open source -- but it's their responsibility to do so, not ours. I believe that one can be had without the need to beg OSS folks to do foo or not to do bar. RH is in no financial trouble, unlike RH. IBM has already made its Linux investment back and more.
Third, Mandrake has made a decision to make ISOs available. I do not believe that they have clauses preventing redistribution. Their Mandrake Club product is providing a service -- namely, a fast download site to obtain Mandrake. Their service has been made somewhat obsolete by recent technological developments -- the design of BitTorrent, which makes distribution of large files much more feasible and affordable. If they wanted to, they *could* limit reproduction of large chunks of their product. They do not need to have a GPLed installer, for instance. This would allow folks to redistribute source packages, but not copy full ISOs. However, for various reasons (including maintaining community goodwill -- as evidenced by Caldera, extremely important with open source software), they have decided to allow reproduction. I have no compuctions about taking them up on their offer. They were not required to make that offer.
Again, I would like to point out -- Mandrake is a Linux company. This does not equate to "eating children", however. Let us assume your worst case scenerio -- that Mandrake goes under as a result of this. Red Hat is doing well, and Debian is happily motoring along with no commercial backing (I don't know about other distributions, though I don't know of anything other than SuSE that have had financial problems.) Linux and Linux distributions will march on, no matter what, even if Mandrake falls. Linux is here to stay.
What proof do you have that Microsoft is behind the SCO situation?
:-)
Actually, Microsoft isn't behind the SCO situation. The primary player in the SCO situation is Linus.
As many have pointed out over the years, Linus is a very smart guy. He also has a ton of influence over the psyche of the Linux movement. So, about a year ago, Linus had a conversation with McBride that looked something like this:
Torvalds: Hey, McBride. I've got a great proposition for you.
McBride: I'm interested.
Torvalds: How about you claim that you've got a death grip on Linux, and start firing off nonsensical lawsuits? This will depress the stock prices of the various Linux businesses. Then, I buy back into some of them, we drop the whole business, and I ride the wave back up. If anyone just ignores you, I'll be sure to act really worried about the lawsuit. It'll be great PR!
McBride: Hmmm..ingenius. I like it. But what's in it for me?
Torvalds: That's the beauty of it. While you're shooting off all these lawsuits, SCO's stock price will be shooting up. You buy stock beforehand, and cash out while the thing is elevated, before it comes crashing down again.
McBride: Fantastic. I'm in, Linus!
You can be damned sure that Linus's dark hand is indeed in part of this if you see Linus purchasing *any* sort of portion of *any* Linux-related company over the next year.
By "time until fixed in the source tree", I'm just pointing out that Microsoft may take months to roll out a patch to users in a hotfix or service pack.
Also, to be fair, I suspect that few users immediately apply kernel patches in the Linux world. They wait until RH's up2date or Debian's apt-get sucks down the latest and greatest. A fair comparison should might say "Microsoft does not attempt to supply a 'rapid-release' patch for technical users at all, unlike the Linux community. However, it's time-to-Joe-end-user-release is comparable to that of Red Hat." or something along those lines.
I certainly feel that, at least applying the immediately obvious and most useful criteria, Microsoft does *not* fix bugs (release patches) more quickly than the Linux community.
*I* think he's referring to the time until a bug gets fixed in the source tree. Furthermore, what measures as a "bug" may differ. Many "bugs" in open source software are feature requests, etc. The measurement may be from the time that Microsoft filter personnel classify something as a "bug" (i.e. a reported severe security hole may sit around for months before it reaches developers flagged as a MUSTFIX bug with MS's internal bug tracking system, for all I know). Finally, security-related bugs may differ in fix time from ordinary bugs ("this icon should be moved over by two pixels to the right").
This doesn't mean that Bill's wrong, but it certainly doesn't tally with my past experience, and given that this is currently a senstive Microsoft weak point, I'd be expecting at least a bit of coloring of the truth.
At full price, not for free or other incentives. SCO doing so would be wonderful PR.
Actually, I'd say that this is more reasonably taken from a pessimistic view. *16%* of all CIOs were negatively influenced (seems unlikely that it's a positive influence) toward Linux because of this one SCO money grab.
(a) A couple places, like Jerusalem, are a big deal to both sides. Land is not fungible, in this case.
(b) (Assuming you're from the US) What if the Soviets came, took over, kicked you out of your house, and moved you up to cold Canada? Would you be thrilled at the idea of renaming Canada "The United States of America", and the exiles stop complaining and being resentful?
I was unaware that Apple had repurchased their stock. However, the shares were always nonvoting, FWIW.
I've always wondered why we spend tons of tax money and incur a whole lot of international ill will by supplying Israel with military hardware and backing them in many unpopular moves.
I think it has to come down to one of two possibilities.
The first is the one that the Arabic countries around Israel like pushing -- that wealthy, influential Jews in the United States are able to put pressure and affect enough US leadership that Israel gets backing. While I'm sure this plays a role, I doubt that it's the primary cause.
What *does* seem a bit more plausible is that the only way the US can control the Middle East is if it's divided and fragmented, unable to use the oil supply as a weapon, and always providing us with at least one ally in the area for military staging purposes. US Middle Eastern foreign policy has seemed to focus on keeping the Middle East divided for *decades*. Israel is a wonderful divider, particularly because they're *dependent* upon US backing for their continued existance, so is certain to help out the US in the Middle East.
The first possibility is disturbing, as it means that US citizens are not properly being served by their government. The second shows the US being uncomfortably mercenary...but I suppose that somebody has to produce the economic imbalances that keep me living the good life, so I'm much less inclined to complain about politicians playing hardball on my behalf.
(a) Use kerberized FTP. Kerberos is a bitch to set up, but if your sysadmin is paranoid about security, he should be using it. Kerberos just deals with authentication, so it's possible to use non-encrypted systems that still use Kerberos authentication.
(b) Why is it "too slow"? A modern system running AES can saturate a 100Mbits/sec network.
Apparently, the article author, "Daniel Lyons", has had a wonderful time writing anti-OSS articles for some time. His articles generally at least touch on both sides, but his conclusion -- his closing words -- are reliabily anti-OSS, though certainly not fanatically so.
Take a look at his take on the the SCO lawsuit, his opinion on whether Linux will succeed, and an article criticizing IBM for not indemnifying Linux users.
The GPL did not come along and ambush these companies - they CHOSE to make use of GPL software. So tough-titty to them.
And this isn't just a matter of "choosing to use GPLed software", a la Linux on the desktop. They incorporated extremely large quantities of code from software that they did not own (and would cost hundreds of millions of dollars to develop in the commercial world) into their own product.
I'd like to see a company sign a MS shared-source agreement and try using Microsoft kernel source, and see how gently Microsoft treats them. Microsoft would *love* having a huge club like that. Yes, they probably wouldn't demand removal. They'd just use it as a lever to force Cisco into signing other useful agreements. Overall, I think Cisco would likely be hurt more in the long term by stealing Microsoft source than GPLed source.
Sounds like a Microsoft ... organized FUD campaign
:-) They're large enough that they'd drive the product out of the market.
I disagree. I really think that if Microsoft was doing so, they'd publically take a stand (a la "OSS is viral"), rather than paying off journalists. The thing that drove this article is the same thing that drove pro-Linux hype a couple years ago, before Linux was ready for desktop transitioning -- business journalists are terrified of being behind the curve, and are constantly trying to find something new that they can give as good tips to their readers.
I do think that I can comfortably claim that this article has little value from a useful content perspective. Linksys management almost certainly did not decide to use GPLed software -- some subcontractor put it in. Aside from the fact that it's easier to get ahold of GPLed source than source to most non-GPL software (and thus GPLed software and temptation might lead to more violations), there isn't really much causality with the GPL. Some engineer at Linksys (or some Indian subcontractor...there are good engineers in India, but in the current environment, there are also a *lot* of people that simply swipe code and let the company get screwed down the line) swiped some code.
I don't buy into the whole "a non-GPL company would just demand moderate royalties, but the FSF requires you to remove the software" claim. If you're Cisco (or IBM, or Oracle) and a company discovers that you're bundling its software in your products, you are going to get burned. Badly. There is simply too much potential money involved. No lawyer is going to let you off nicely. At that point, you are up shit creek, and the degree to which you are going to get screwed is fairly academic.
Slashdot has been bitterly complaining about SCO for some time -- a company that (well, in its own eyes) doesn't have GPLed source, and discovered (or at least wants to convince others) that it has its source in other, major products. Sorry, but $699/copy is not trivial.
Come to think of it, SCO licensing fees would also apply to the routers in the article. Funny how that works, eh?
The article implied that using GPLed software immediatelye exposed one to such potential liability. I hope that no IT manager is going to hold off moving to Linux on the desktop because of random IP worries.
The fact that the FSF does not accept royalties could be a reasonable complaint for some companies that have deliberately or otherwise stolen FSF software. However, if you aren't maintaining enough control over your software to avoid having mass chunks of source that you haven't checked out the license on incorporated into your product, you have severe legal problems -- forget worrying about the GPL.
Furthermore, I severely doubt that Linksys made a measured decision to use GPLed software, or at least a decision that was properly passed through legal. I cannot understand how a corporate lawyer would sign off on incorporating GPLed code into non-GPLed code. I have a sneaking suspicion that the work was done by a subcontractor or an engineer that wasn't being reviewed by legal.
Finally, I think that, as per the article, if the FSF is simply giving the option of removing the code (with no damages, which I find extremely unlikely), that's more than generous. There have been many, many, many Linksys routers shipped. A company could sue for phenomenal back damages.
You worry that MS could change the interface and break compatibility with Mono. And yet, many, many vendors produce DBMSes with an SQL interface. SQL is horribly, horribly extended by all sorts of vendors, and has truly awful compatibility. However, it's a decent bit of work, and the abovementioned issues haven't stopped postgresql and mysql from rapidly consuming large chunks of the market.
I doubt Mono could ever be in a situation as bad as SQL is. The compatibility issues there are almost overwhelming, and yet I find it very fortunate that the MySQL and Postgresql developers did *not* decide to give up on SQL because someone else already had a closed SQL-using implementation.
I wish I knew whether you were the real Miguel. Slashdot is loaded with fake ones. :-(
:-)
.NET VM. C# is supposed to (as may be obvious, I'm not a C# user) fix a number of issues with Java.
"hasuseraclue()"...Note: by reverse engineering the code, we know that above system call return 0 when ran on the system of the author of the previous paper.
Actually, this is one of the funniest and most germane comments you could have made. Think about it for a moment. Traditionally, MS library calls return TRUE for success and FALSE for failure. Traditionally, Unix library calls return the opposite -- 0 for success and non-zero (well, negative) for failure. You don't have a problem with MS-produced APIs and so wrote a comment that the other person would consider a compliment -- because presumably he adheres only to Unix APIs.
In a world where Mono is vastly successful, if Microsoft changes/introduce new APIs, do you think it will matter?
While I wish you luck and respect your work (if you are, indeed, the real Miguel), I think that you may have missed a point. Your argument that Mono will be a binding influence on Microsoft is predicated on the fact that Mono will achieve significant "market" share, before any sort of nasty compatibility-splitting moves are made by Microsoft. I'm not sure Mono will do that. Microsoft has a tremendous team of language designers that had a head start of *years* on you folks. It takes a while to catch up, and given that this is still a new, sexy initiative at Microsoft, it's going to be a moving target.
However, I really don't see the problem as all that severe anyway -- I'd support your argument for entirely different reasons. (a) Presumably Mono can be made to work on Windows, providing a common runtime everywhere. So if Microsoft changes things, there will still be a runtime that C#-using developers could ship with their applications. (b) Even if Microsoft changes course...so what? How would the situation be worse than before? We already don't have binary compatibility between the C Unix applications written to an entirely different set of APIs and the MFC-based Windows software. The only different thing would that there would be a fairly solid development tool available to to rapid application development on Linux -- and a tool that would make it easier for experienced C# developers to transition to Linux.
But let's assume that Mono *will* hurt Linux. I'd like to point out to the grandparent poster that Miguel is hardly under any onus to drop a project because it isn't in Linux's immediate favor, anyway. Open Source is all about writing good software and then sharing it with other interested people, folding improvements back in. Miguel's doing exactly that -- providing a good tool. He's done immense amounts of work in the past that have enormously benefited the OSS and Linux worlds -- likely more than the grandparent poster.
Finally, Java is, frankly, not a particularly well designed language. It was well marketed, and it managed to win over a large chunk of the C++ crowd, and is an improvement over earlier languages for folks who want a C-like safe language. However, many of the design decisions were poor. Furthermore, the JVM has some fundamental design limitations that prevent it from being useful for some non-Java languages like ocaml. I believe that this is not the case for the
Miguel, I have my differences with some design in GNOME -- I think that, in retrospect, CORBA and XML were overly heavyweight. I'm sad that the decision was made not to include an barber pole-like "infinite progress bar" widget in GNOME to denote tasks with an unknown completion time. However, it's the fact that I can pick such minor nits that makes it really clear how much your work in the past has done for us now. I can sit down and work on an Excel spreadsheet. My Windows-using friends can comfortably use Linux.
Thanks again. And whatever happens, don't wind up like Eazel.
Now, what games for the GameCube use bump mapping? Nyo-ho!
The GameCube doesn't have support for hardware bumpmapping. There isn't any hardware that isn't being used -- it's just CPU cycles that are going to other things.
The surround sound...all right, I suppose that you could argue that it's not being "pushed to the limit".
I don't believe the GC *has* a higher resolution mode, though I can't swear to it.
As for the steel backplate in the thing: it's known that consumers generally associate some heft to objects with higher quality, to the point that some products are weighted to take advantage of this.
Dunno if that was the reason in this case.
I have to disagree with you there. OEM-supplied components *are* quite flimsy, since they compete on essentially nothing but price (well, perhaps extra volume buttons on the keyboard might be a selling point).
There are a number of manufacturers that produce more highly-price, nicer input devices. Among them, Logitech,
which sells a whole line of wireless keyboards, PCKeyboard, which has among its offerings modern versions of old-IBM-Model-M style buckling-spring mechanism keyboards (CLICK CLICK CLICK...), Fingerworks' unusual offerings, the expensive-but-ultimate-RSI-avoidance Kinesis keyboards, and Goldtouch's ergo products.