If you read the article introducing the kittens concept, you'll see that the author intends it to be customized to each site, thus preventing spambots from simply memorizing the pictures. And randomly picking three out of 9 images only gives a possiblity of success of 1/84, better than many word captchas are achieving these days.
Anyone who wants to custom-program a bot for a single site would just be better off manually posting their spam.
It's not that we think all information should be free, it's that we believe ideas should be free. Scientific, cultural, and even economic progress is built on the free flow and exchange of ideas. It is hard to make a good case for the perpetual monopolization of ideas or art, as the WTO "intellectual property" regime seeks to do. You can't make the case on scientific, cultural, economic, or moral grounds.
Frankly, it's a matter of good public policy. Does it make good sense in the long run to deprive the public of its cultural and scientific heritage, to the point that an entire generation of people will die before it becomes public domain, in the name of profit? Similarly, does it make good sense to expose the public to thieves, scam artists, and stalkers, to the point that their livelihood and even lives could be in danger, in the name of profit?
Now, there are occasions where some types of privacy data - web surfing habits or credit purchases, for example - can be used for scientific progress, but it is appropriate to be careful with that information. Data of this type, when stripped of its personal identifiers and used in bulk, can be very useful in tracking trends and improving products and processes. When it is identifiable with a person, however, most of the time it is only used to annoy, harass, or steal.
For me, it comes down to this: with ideas, the burden of proof should be on the creator to show why it should not be shared freely with the world as soon as possible. With private information, the burden should be on the collector to show why it shouldn't be kept private. And "because I'll make more money" is not an acceptable answer in either case.
We were under restrictions from legal when using GPL'd code because there was no clear definition of 'distribution'. According to some of the legal review of the GPL, we would have been distributing our code when we made it available to solely owned affiliates of the parent company. They were part of a separate business line with their own IT and executive staff, more a maintenance organization than a retailer (they did not have stores).
As an overall enterprise we were consolidating on common platforms (bulk purchasing power) across the board, but were still separate entities. The legal advisors indicated that our sharing of code with GPL components to these affiliates consituted distribution which would have activated the viral nature.
Is it really such a problem to have to distribute the source code to your own solely owned affiliates? I would think they could be trusted to keep it all "in the family." Even if they have the "right" to redistribute it freely, they could still be under "guidance" from the parent company not to exercise that right. The only time it would get sticky is if you sold off that subsidiary to a competitor.
The GPL only requires that you give the source code to those who get binaries. Just because you "distribute" it to your subsidiaries doesn't mean you have to set up a public FTP or CVS server for the code.
Yes, but that week of work by the Firefox team fixing the bugs for 1.0.4 is what made it possible for Netscape to re-package the same fix in a day. And let's not forget that 1.0.4 had been out for a week already before Netscape launched 8.0.
Really, Netscape had no business launching 8.0, and should have waited a day to get the 1.0.4 fixes in. Known exploits with a known solution should be showstoppers - period.
The market being what it is, I was lucky to be gainfully employed for the first time in 18 months. Besides, in general, they're a pretty good company - I've been there 18 months now, and it's really my only complaint. That doesn't mean it's not a big beef for me still, but certainly not enough to risk another long unemployment streak.
My company has a policy that we're "always open." Last year, we experienced a blizzard so severe that the city issued tickets to any non-emergency vehicles on the road. The power at the building was also knocked out. Yet when we got back to work, we were expected to make up the time or use Paid Time Off. The excuse made by management was, "Sometimes you have to use you PTO benefit for the company's benefit." (!)
I'd been working there all of a month, or I probably would have contacted a lawyer. If it happens again, I'll definitely be exploring a class-action suit. In a state of emergency, when it's not safe to go to work, it's completely unreasonable to expect employees to come in. If it's not illegal, it ought to be. If I were you, I'd contact a lawyer.
We shouldn't be using "intellectual property" or any replacement at all. The term implies that such disparate ideas as copyright, trademark, and patent have an overarching parent idea enshrined in law, which they do not. It deliberately confuses them, with an intent to grant the most restrictive properties of any one to all three. That is to say, referring to "intellectual property" conjures the illusion that copyrights are perpetual as long as they are not abandoned, like trademarks; that patents can be used to prevent derivative works, like copyrights can; and that you can use trademarks to prevent a competitor from entering your line of business, like patents. Think about any of the cases in the last ten years that revolved around "intellectual property" rather than specific claims of copyright, trademark, or patent infringement, and you will see what I mean.
If for some reason you need to refer to copyrights, trademarks, and patents in one breath, the term you should use is "copyrights, trademarks, and patents!"
just checked the quark site, upgrade pricing is $199-499 depending on the version you've got, so one would presume 5.x users (myself included) can get it for $199.
Still, isn't it a little aggaravating to pay $200 for the same program you bought a year ago, just so you can run it on an operating system that's been out and stable for almost 2 years? They'd never get away with charging $200 for an upgrade whose sole discriminating factor is that it works on Windows XP when it used to only work on Windows 95 (especially if Windows XP had already been out two years)!
Go ahead and mark me a troll, but I do know whereof I speak when it comes to Quark attitudes and culture, having worked there for a year until my whole project was laid off to celebrate getting a release out the door.
If Quark keeps to its m.o., the team in the U.S. who actually built XPress 6 will now be pink-slipped and the product responsibility transferred to Chandragar, India. No knock against Indian developers in general, but Quark has not adopted a "best programmer rupees can buy" mentality there, and the continued maintenance will probably be a nightmare. Quark India is very Windows-centric, and even at that their programmers are writing C++ and Java like it's Visual Basic.
N.B. by Quark's own versioning rules, this should be XPress 5.5, and they should be charging the minor upgrade price to XPress 5 users. Mac users who bought XPress 5 are getting screwed royally. I'm sure in Fred Ebrahimi's (the owner of Quark) mind, it's justified since the porting effort was so extensive, but the only notable feature is Carbonization. There was a post above that noted Ebrahimi's assertions that the Mac is a "dying platform." Quark didn't even commit to Carbonizing XPress until Mac OS X (and InDesign 2.0) shipped, and Ebrahimi realized the publishing market would dump XPress before they'd dump the Mac. When I was laid off, every program the company had in R&D was Windows-only by design. Talk about a company that doesn't know what side its bread is buttered on - Quark deserves to be reduced to irrelevance just for sheer lack of vision. Go Adobe!
Yes, perhaps I/we are still confusing the IBM != "IBM-compatable" architecture issue. IBM is barely relevant in the PC market but the "IBM-compatable" PCs are still very dominant. I may very well be wrong about how willingly the architecture was opened to the clone makers but even IBM credits clone makers for helping launch the IBM-compatable PC industry.
Agreed. My only point was that IBM did nothing to push IBM PC as a standard until it was too late to fight it. As a result of the design choices they made, it became easy to create interoperable products without their consent or participation. Apple, on the other hand, was able to retain ruthless control because of their particular design choices. Those choices weren't necessarily made to inhibit competition. In fact, Apple's main goal was technical excellence, while IBM's main goal was beating the Apple ]['s price point. At any rate, the IBM PC "standard" is an historical accident, and it's really only because IBM wanted Apple's market.
What was the point I was making again? Ah, yes: IBM's design conquered the market, but left IBM largely irrelevant to that process, while Apple has stayed in control of their design, which does well for them, but is not the 500-lb. gorilla of the personal computing world. Which company got the better deal overall? Discuss.
IBM PC is not standard because of IBM's power but rather because they opened up the architecture for other companies to clone and produce software for. This pushed the architecture into widespread use and therefore became a standard. Had Apple opened their architecture, perhaps Apples would be sitting on the majority of the desktops today.
Actually, IBM didn't really open up the architecture - it was reverse engineered by competitors, and IBM then sued (unsuccessfully) to stop them. Say what you will, but at least Apple is the dominant force in one segment of the personal computing market, while IBM is barely relevant. Whether or not that is a bad thing is beyond the scope of this post. There are two reasons this happened to IBM and not Apple.
IBM used off-the-shelf components, while Apple custom-built much of the chipsets for both the Apple ][ and the original Macintosh.
Apple owned the operating system software as well as the hardware designs, while IBM licensed MS-DOS for their machines. If Bill Gates had allowed IBM to purchase MS-DOS outright, IBM could have refused to sell licenses to clone makers, or at least "taxed" them heavily, thus eliminating them from, or reducing their share of, the market, just as Apple did before and after the clone makers of the late 1990s.
Much as it always pains me to say it, Microsoft, not IBM, drove the PC revolution.
Re:About as viral as accidentally giving away secr
on
What if SCO is Right?
·
· Score: 1
A problem occurs to me, here. Would a court find that the warning that "Linux contains infringing code" is sufficient to anyone distributing Linux in any way? Would a court find that the appropriate response for all Linux distributors is to immediately cease distributing it?
That's an awful wide swath, of course; among other things, it means TiVo has to stop selling their flagship product. (Sole product?) But is that a possibility? That a court would find that Red Hat, SuSE, Sharp, Mandrake, et al. should have just stopped?
In the absence of a cease-and-desist order from SCO, or a motion for an injunction brought by SCO, I would think not. SCO has an obligation to notify all suspected infringers before those alleged infringers would need to cease distribution. There are three main reasons why SCO is not doing this:
It costs money to get an injunction slapped against product distribution, especially if the judge doesn't think you have a substantial chance of prevailing in your claim. You have to post a bond used to reimburse the distributor if the lawsuit fails. So far, SuSE hasn't even filed suit against a single Linux distributor, only a development partner, so there's no grounds for an injunction, anyway. They could send out cease-and-desists, but then they have to say what behavior constitutes infringement, thus saying what the code in dispute actually is.
You have to serve each distributor and developer individually, so you have to ask for injunctions against RedHat, SuSE, TurboLinux, Debian, LibraNet, Lindows.com, Yellow Dog, Knoppix, and on, and on, and on. You also have to serve Linus Torvalds, Alan Cox, Marcelo Tosatti, and every kernel hacker that's ever touched your disputed code. You also have to post a bond against each one.
Since this is nothing but a naked stick-up by SCO, it's obvious that they want the "infringing" behavior to continue to inflate their damages.
I think, deep down, SCO knows it can't win in a court of law. Now that IBM is refusing to be mugged quietly, SCO is applying a scorched-earth policy, where they hope to ruin Linux and thus drive business customers to UnixWare for their x86 server needs. The fact that the likely winner in that scenario would be the *BSDs, who based on the outcome of the Novell/UCB lawsuit are likely lawsuit-proof, apparently hasn't occurred to them.
Re:About as viral as accidentally giving away secr
on
What if SCO is Right?
·
· Score: 2, Insightful
As soon as they find them, they drop their Linux product, and launch a lawsuit.
But what they did not do is stop shipping their Linux distribution as soon as they claimed IBM had leaked their code into Linux. If you assume that their lawsuit was filed legitimately, in that they knew there was proprietary code of theirs included in Linux, it then follows that they, under the terms of the GPL, were giving an implicit license to that code by continuing to distribute it. They didn't cease shipping their distribution of the code for nearly two months after they first alleged infringment.
So no, they did not do everything they could to limit their damages. In fact, by not stipulating what the infringing code is (which, as many others have noted, would not prevent them from seeking damages for the past infringment), they are in fact trying to increase their alleged damages by not allowing the infringing developers to fix the problem. SCO is acting in bad faith on every front to try and get a court to give them what they have never been able to achieve on their own merits (either as Caldera or as SCO in its previous incarnations) - dominance in the UNIX for x86 space.
It's funny - I always thought Caldera was a bit of a joke with Ransom Love at its helm. I recall the cheering that went on when he was no longer in charge there. Now, though, I think we'd rather have the clownish Love, who just didn't get it, instead of Darl McBride and crew, who don't get it and are determined to mess it up for everyone else.
Spoken like a true Libertarian (or a demagoguing Republican). There are budget cuts happening all over California. They're coming in public schools, rural healthcare, state parks, higher education, and more. Those are services that most Americans, and especially Californians, think are important. In fact, in order to cover the necessary gap, Gov. Davis has proposed more cuts than new taxes.
Incidentally, I decided to reply instead of modding you down, even though zapping your "insightful" bonus was very tempting.
Actually, Quark has laid off most of its U.S. staff, and has shifted the porting effort to its staff in India. Given Microsoft's stranglehold on Indian Universities, it's not a real surprise that they're having trouble with the port.
While I worked at Quark (I was one who was laid off, but I didn't work on XPress), Fred Ebrahimi's (the sole owner now) disdain for Apple was very clear. Every new product the company tried was designed from the ground up to be Windows-only. Tim Gill, the real visionary behing Quark's original success, was the one who liked Macs, and he sold his half of the company back to Ebrahimi in 2000.
The point is not, and never has been, that Microsoft allowed Java programmers to access the Win32 APIs from Java. The problem is that Microsoft ignored the proper, documented, supported, and contractually obligatory method - the JNI API. Apple has proven, with their Cocoa-Java Bridge, that it is possible to create a set of classes that call out to native methods cleanly, with a compliant JVM. Before anyone objects, there's nothing particularly difficult about it that required the Java2 spec - the capability was there all along. Microsoft deliberately chose a method that polluted the bytecode level, thus forcing programmers who wanted to use the Windows Foundation Classes to use a compiler that was available only for Windows, and a runtime that was available only for Windows, even if the class interfaces themselves could have been duplicated on other platforms. They did this in violation of their contract with Sun. The anti-Microsoft attitude at Sun came for the most part from this episode, not before it.
As to your second point, adding features that are not yet standardized works the same as any other language - you write or buy a library, which in this case may or may not require JNI calls for hardware or OS-level access. Once the standard library for those calls in available, you make the decision to convert or use your established interface. That's what's happening in J2EE all the time. IBM, Oracle, BEA, et al, decide a feature is needed in a Java Application Server, and they all implement a version of it themselves. Early adopters can take the plunge with their favorite vendor and go. Those who want to wait for cross-platform compatibility wait for the Java Community Standards process, where all these vendors hammer out a common standard that they will all support. If Microsoft had behaved as a partner in Java instead of a usurper, they would have been able to do exactly that with, for your example, USB - release a set of com.ms.usb APIs that operated only on Windows without perverting the bytecode specification, and worked with Sun, Apple, HP, etc. on a cross-platform version to be released as javax.usb.
Why are DVD less epensive than CDs? Movies cost more to make than music. DVDs cost more to make than CDs. So why are movies almost always sold for less than soundtracks for movies?
Market uptake. One reason LaserDiscs never took off is that they were too much more expensive than VHS for the average consumer to choose them - better quality was not a compelling enough reason. The motion picture industry has learned a little better this time around. So DVDs will be (relatively) cheap until they have eliminated the consumer demand for VHS, at which point prices will start climbing just like they have for CDs ever since demand for compact cassettes nearly vanished.
Come on! If I buy a used CD, it's because I want a copy on CD, and find it wasteful and extravagant to buy a brand-new copy when older copies are going spare. By the time a CD has hit the used bin, I could have downloaded it a thousand times from any of the various music-sharing sources. I hardly need to pay $7 to buy a CD, rip it, and sell it back for $3. That's $4 I never needed to spend.
The issue, as always, is about price fixing. Used CDs, like their digitally-shared cousins, compete with their still-shrinkwrapped brethren to drive down the price from the ever-encroaching $20 mark. The RIAA is not an "industry trade group" - it's a trust by any reasonable interpretation of the Sherman Act. Record executives deciding anything together - especially legislative agendas and lobbying efforts - should be illegal!
I think Microsoft puts out some great products, at least as far as user-interface is concerned. They are on top for that and a variety of other reasons.
Actually, most UI critiques of Microsoft products read like a litany of cardinal sins. Intuitive: no, natural: no, adaptable: no, etc., etc. The one category where they really excel is in consistency. Unfortunately, it's consistency in bad paradigms. The reason that the Windows 95 interface has been copied over and over again (by both KDE and GNOME, notably), is because it has been grudgingly learned by a public with no other options. I mean, we're talking about an interface that needed a massive marketing blitz and a best-selling "how to use it" video starring TV celebrities to help people figure it out!
In a market with true competition, a competitor would have been able to make major inroads in 1994, just because Windows 95 was so frightening to users.
> yet Phillips (the company who invented the CD spec) has said that it intends to try and make its next CD burner able to circumvent the copy mechanisms.
I'm glad Phillips is doing this. It's great to finally see a company supporting people's fair use* and right to listen to music they have bought, regardless of the medium.
However, I've always wondered if Phillip's didn't have an exterior motivation? If [audio] copy protection worked, would Phillips have less sales of CD's? What do they have to gain by taking a stance against copy protection?
It's very simple - Philips/Magnavox is not primarily a content provider. Sony, RCA, and most of the other equipment manufacturers also have an interest in content production. Since Philips makes their money by enabling fair use, they have a vested interest in continuing to do so, against the interests of other manufacturers to inhibit it to protect their content providing divisions. If Philips is swallowed up by the AOL/TW, Sony, or Viacom behemoths, expect that to change instantaneously.
Always remember - "copy protection" isn't about preventing copying - it's about preventing a level playing field for content production. The thing the RIAA fears most is independent artists and labels not having to pay the cover charge to the party, not that John Q. Pirate is going to take a $.01 bite out of their $9 profit. If they can ruin the CD for consumers, and force an exodus to DVD-Audio or some similarly harder-to-enter market, then they are back in the driver's seat. The CD burner isn't a threat because you can copy their CD - it's because you can make your own.
I've always thought it would be most cool if there were a client equivalent to a servlet engine that allows all these desktop apps to share a JVM.
Actually, as long as there are no namespace conflicts, this would be relatively easy to achieve using the Reflection APIs and dynamic class loading, I believe. Instead of a "java..." invocation on the command line invoking an entirely new JVM process, it would be more of an access point to the class loader. Determining which version of a particular namespace a newly-loaded class wanted when there are conflicts is a tricky problem, though. I'm kind of surprised there hasn't been more work on this by the HotSpot team, although if I recall correctly, they were targeting some of this for the 1.5 release. Apple has made some progress on this in Mac OS X, as well.
...doesn't mean the Court will hear the case. They already refused to hear it ahead of the appellate court, so it is as likely as not that the Supreme Court will allow the penalty phase to be decided before hearing the case.
That's especially true when you figure the the SC probably doesn't want to have to craft a penalty, and would just send the case back down for a penalty phase if it upholds the lower courts' rulings.
The Findings of Fact and of Law are already pretty firmly entrenched in this case. I feel reasonably secure that when and if the case does make it to the Supreme Court that it will stand by the findings made by the lower courts.
when these kind of outages happen, of Peter Deutsch's 8 Fallacies of Distributed Computing:
The network is reliable
Latency is zero
Bandwidth is infinite
The network is secure
Topology doesn't change
There is one administrator
Transport cost is zero
The network is homogeneous
This is, of course, why the idea of remote authentication being necessary to use your word processor is a bad thing. Heck, even losing something as innocuous as an instant messaging program brought thousands of people to a screeching halt for a week. It seems to me that Microsoft (although they're certainly not the only ones) seem to believe these 8 fallacies blindly, espcially 1, 4, and (they're hoping) 6.
Slashdot Mirror
If you read the article introducing the kittens concept, you'll see that the author intends it to be customized to each site, thus preventing spambots from simply memorizing the pictures. And randomly picking three out of 9 images only gives a possiblity of success of 1/84, better than many word captchas are achieving these days.
Anyone who wants to custom-program a bot for a single site would just be better off manually posting their spam.
It's not that we think all information should be free, it's that we believe ideas should be free. Scientific, cultural, and even economic progress is built on the free flow and exchange of ideas. It is hard to make a good case for the perpetual monopolization of ideas or art, as the WTO "intellectual property" regime seeks to do. You can't make the case on scientific, cultural, economic, or moral grounds.
Frankly, it's a matter of good public policy. Does it make good sense in the long run to deprive the public of its cultural and scientific heritage, to the point that an entire generation of people will die before it becomes public domain, in the name of profit? Similarly, does it make good sense to expose the public to thieves, scam artists, and stalkers, to the point that their livelihood and even lives could be in danger, in the name of profit?
Now, there are occasions where some types of privacy data - web surfing habits or credit purchases, for example - can be used for scientific progress, but it is appropriate to be careful with that information. Data of this type, when stripped of its personal identifiers and used in bulk, can be very useful in tracking trends and improving products and processes. When it is identifiable with a person, however, most of the time it is only used to annoy, harass, or steal.
For me, it comes down to this: with ideas, the burden of proof should be on the creator to show why it should not be shared freely with the world as soon as possible. With private information, the burden should be on the collector to show why it shouldn't be kept private. And "because I'll make more money" is not an acceptable answer in either case.
Is it really such a problem to have to distribute the source code to your own solely owned affiliates? I would think they could be trusted to keep it all "in the family." Even if they have the "right" to redistribute it freely, they could still be under "guidance" from the parent company not to exercise that right. The only time it would get sticky is if you sold off that subsidiary to a competitor.
The GPL only requires that you give the source code to those who get binaries. Just because you "distribute" it to your subsidiaries doesn't mean you have to set up a public FTP or CVS server for the code.
Yes, but that week of work by the Firefox team fixing the bugs for 1.0.4 is what made it possible for Netscape to re-package the same fix in a day. And let's not forget that 1.0.4 had been out for a week already before Netscape launched 8.0.
Really, Netscape had no business launching 8.0, and should have waited a day to get the 1.0.4 fixes in. Known exploits with a known solution should be showstoppers - period.
The market being what it is, I was lucky to be gainfully employed for the first time in 18 months. Besides, in general, they're a pretty good company - I've been there 18 months now, and it's really my only complaint. That doesn't mean it's not a big beef for me still, but certainly not enough to risk another long unemployment streak.
My company has a policy that we're "always open." Last year, we experienced a blizzard so severe that the city issued tickets to any non-emergency vehicles on the road. The power at the building was also knocked out. Yet when we got back to work, we were expected to make up the time or use Paid Time Off. The excuse made by management was, "Sometimes you have to use you PTO benefit for the company's benefit." (!)
I'd been working there all of a month, or I probably would have contacted a lawyer. If it happens again, I'll definitely be exploring a class-action suit. In a state of emergency, when it's not safe to go to work, it's completely unreasonable to expect employees to come in. If it's not illegal, it ought to be. If I were you, I'd contact a lawyer.
We shouldn't be using "intellectual property" or any replacement at all. The term implies that such disparate ideas as copyright, trademark, and patent have an overarching parent idea enshrined in law, which they do not. It deliberately confuses them, with an intent to grant the most restrictive properties of any one to all three. That is to say, referring to "intellectual property" conjures the illusion that copyrights are perpetual as long as they are not abandoned, like trademarks; that patents can be used to prevent derivative works, like copyrights can; and that you can use trademarks to prevent a competitor from entering your line of business, like patents. Think about any of the cases in the last ten years that revolved around "intellectual property" rather than specific claims of copyright, trademark, or patent infringement, and you will see what I mean.
If for some reason you need to refer to copyrights, trademarks, and patents in one breath, the term you should use is "copyrights, trademarks, and patents!"
Still, isn't it a little aggaravating to pay $200 for the same program you bought a year ago, just so you can run it on an operating system that's been out and stable for almost 2 years? They'd never get away with charging $200 for an upgrade whose sole discriminating factor is that it works on Windows XP when it used to only work on Windows 95 (especially if Windows XP had already been out two years)!
Caution: bitterness alert!
Go ahead and mark me a troll, but I do know whereof I speak when it comes to Quark attitudes and culture, having worked there for a year until my whole project was laid off to celebrate getting a release out the door.
If Quark keeps to its m.o., the team in the U.S. who actually built XPress 6 will now be pink-slipped and the product responsibility transferred to Chandragar, India. No knock against Indian developers in general, but Quark has not adopted a "best programmer rupees can buy" mentality there, and the continued maintenance will probably be a nightmare. Quark India is very Windows-centric, and even at that their programmers are writing C++ and Java like it's Visual Basic.
N.B. by Quark's own versioning rules, this should be XPress 5.5, and they should be charging the minor upgrade price to XPress 5 users. Mac users who bought XPress 5 are getting screwed royally. I'm sure in Fred Ebrahimi's (the owner of Quark) mind, it's justified since the porting effort was so extensive, but the only notable feature is Carbonization. There was a post above that noted Ebrahimi's assertions that the Mac is a "dying platform." Quark didn't even commit to Carbonizing XPress until Mac OS X (and InDesign 2.0) shipped, and Ebrahimi realized the publishing market would dump XPress before they'd dump the Mac. When I was laid off, every program the company had in R&D was Windows-only by design. Talk about a company that doesn't know what side its bread is buttered on - Quark deserves to be reduced to irrelevance just for sheer lack of vision. Go Adobe!
His name is Fred Ebrahimi.
Agreed. My only point was that IBM did nothing to push IBM PC as a standard until it was too late to fight it. As a result of the design choices they made, it became easy to create interoperable products without their consent or participation. Apple, on the other hand, was able to retain ruthless control because of their particular design choices. Those choices weren't necessarily made to inhibit competition. In fact, Apple's main goal was technical excellence, while IBM's main goal was beating the Apple ]['s price point. At any rate, the IBM PC "standard" is an historical accident, and it's really only because IBM wanted Apple's market.
What was the point I was making again? Ah, yes: IBM's design conquered the market, but left IBM largely irrelevant to that process, while Apple has stayed in control of their design, which does well for them, but is not the 500-lb. gorilla of the personal computing world. Which company got the better deal overall? Discuss.
Actually, IBM didn't really open up the architecture - it was reverse engineered by competitors, and IBM then sued (unsuccessfully) to stop them. Say what you will, but at least Apple is the dominant force in one segment of the personal computing market, while IBM is barely relevant. Whether or not that is a bad thing is beyond the scope of this post. There are two reasons this happened to IBM and not Apple.
Much as it always pains me to say it, Microsoft, not IBM, drove the PC revolution.
In the absence of a cease-and-desist order from SCO, or a motion for an injunction brought by SCO, I would think not. SCO has an obligation to notify all suspected infringers before those alleged infringers would need to cease distribution. There are three main reasons why SCO is not doing this:
I think, deep down, SCO knows it can't win in a court of law. Now that IBM is refusing to be mugged quietly, SCO is applying a scorched-earth policy, where they hope to ruin Linux and thus drive business customers to UnixWare for their x86 server needs. The fact that the likely winner in that scenario would be the *BSDs, who based on the outcome of the Novell/UCB lawsuit are likely lawsuit-proof, apparently hasn't occurred to them.
But what they did not do is stop shipping their Linux distribution as soon as they claimed IBM had leaked their code into Linux. If you assume that their lawsuit was filed legitimately, in that they knew there was proprietary code of theirs included in Linux, it then follows that they, under the terms of the GPL, were giving an implicit license to that code by continuing to distribute it. They didn't cease shipping their distribution of the code for nearly two months after they first alleged infringment.
So no, they did not do everything they could to limit their damages. In fact, by not stipulating what the infringing code is (which, as many others have noted, would not prevent them from seeking damages for the past infringment), they are in fact trying to increase their alleged damages by not allowing the infringing developers to fix the problem. SCO is acting in bad faith on every front to try and get a court to give them what they have never been able to achieve on their own merits (either as Caldera or as SCO in its previous incarnations) - dominance in the UNIX for x86 space.
It's funny - I always thought Caldera was a bit of a joke with Ransom Love at its helm. I recall the cheering that went on when he was no longer in charge there. Now, though, I think we'd rather have the clownish Love, who just didn't get it, instead of Darl McBride and crew, who don't get it and are determined to mess it up for everyone else.
Spoken like a true Libertarian (or a demagoguing Republican). There are budget cuts happening all over California. They're coming in public schools, rural healthcare, state parks, higher education, and more. Those are services that most Americans, and especially Californians, think are important. In fact, in order to cover the necessary gap, Gov. Davis has proposed more cuts than new taxes.
Incidentally, I decided to reply instead of modding you down, even though zapping your "insightful" bonus was very tempting.
Actually, Quark has laid off most of its U.S. staff, and has shifted the porting effort to its staff in India. Given Microsoft's stranglehold on Indian Universities, it's not a real surprise that they're having trouble with the port.
While I worked at Quark (I was one who was laid off, but I didn't work on XPress), Fred Ebrahimi's (the sole owner now) disdain for Apple was very clear. Every new product the company tried was designed from the ground up to be Windows-only. Tim Gill, the real visionary behing Quark's original success, was the one who liked Macs, and he sold his half of the company back to Ebrahimi in 2000.
The point is not, and never has been, that Microsoft allowed Java programmers to access the Win32 APIs from Java. The problem is that Microsoft ignored the proper, documented, supported, and contractually obligatory method - the JNI API. Apple has proven, with their Cocoa-Java Bridge, that it is possible to create a set of classes that call out to native methods cleanly, with a compliant JVM. Before anyone objects, there's nothing particularly difficult about it that required the Java2 spec - the capability was there all along. Microsoft deliberately chose a method that polluted the bytecode level, thus forcing programmers who wanted to use the Windows Foundation Classes to use a compiler that was available only for Windows, and a runtime that was available only for Windows, even if the class interfaces themselves could have been duplicated on other platforms. They did this in violation of their contract with Sun. The anti-Microsoft attitude at Sun came for the most part from this episode, not before it.
As to your second point, adding features that are not yet standardized works the same as any other language - you write or buy a library, which in this case may or may not require JNI calls for hardware or OS-level access. Once the standard library for those calls in available, you make the decision to convert or use your established interface. That's what's happening in J2EE all the time. IBM, Oracle, BEA, et al, decide a feature is needed in a Java Application Server, and they all implement a version of it themselves. Early adopters can take the plunge with their favorite vendor and go. Those who want to wait for cross-platform compatibility wait for the Java Community Standards process, where all these vendors hammer out a common standard that they will all support. If Microsoft had behaved as a partner in Java instead of a usurper, they would have been able to do exactly that with, for your example, USB - release a set of com.ms.usb APIs that operated only on Windows without perverting the bytecode specification, and worked with Sun, Apple, HP, etc. on a cross-platform version to be released as javax.usb.
Better yet, how about one that let normal folks talk to politicians and be understood?
Market uptake. One reason LaserDiscs never took off is that they were too much more expensive than VHS for the average consumer to choose them - better quality was not a compelling enough reason. The motion picture industry has learned a little better this time around. So DVDs will be (relatively) cheap until they have eliminated the consumer demand for VHS, at which point prices will start climbing just like they have for CDs ever since demand for compact cassettes nearly vanished.
Come on! If I buy a used CD, it's because I want a copy on CD, and find it wasteful and extravagant to buy a brand-new copy when older copies are going spare. By the time a CD has hit the used bin, I could have downloaded it a thousand times from any of the various music-sharing sources. I hardly need to pay $7 to buy a CD, rip it, and sell it back for $3. That's $4 I never needed to spend.
The issue, as always, is about price fixing. Used CDs, like their digitally-shared cousins, compete with their still-shrinkwrapped brethren to drive down the price from the ever-encroaching $20 mark. The RIAA is not an "industry trade group" - it's a trust by any reasonable interpretation of the Sherman Act. Record executives deciding anything together - especially legislative agendas and lobbying efforts - should be illegal!
Actually, most UI critiques of Microsoft products read like a litany of cardinal sins. Intuitive: no, natural: no, adaptable: no, etc., etc. The one category where they really excel is in consistency. Unfortunately, it's consistency in bad paradigms. The reason that the Windows 95 interface has been copied over and over again (by both KDE and GNOME, notably), is because it has been grudgingly learned by a public with no other options. I mean, we're talking about an interface that needed a massive marketing blitz and a best-selling "how to use it" video starring TV celebrities to help people figure it out!
In a market with true competition, a competitor would have been able to make major inroads in 1994, just because Windows 95 was so frightening to users.
It's very simple - Philips/Magnavox is not primarily a content provider. Sony, RCA, and most of the other equipment manufacturers also have an interest in content production. Since Philips makes their money by enabling fair use, they have a vested interest in continuing to do so, against the interests of other manufacturers to inhibit it to protect their content providing divisions. If Philips is swallowed up by the AOL/TW, Sony, or Viacom behemoths, expect that to change instantaneously.
Always remember - "copy protection" isn't about preventing copying - it's about preventing a level playing field for content production. The thing the RIAA fears most is independent artists and labels not having to pay the cover charge to the party, not that John Q. Pirate is going to take a $.01 bite out of their $9 profit. If they can ruin the CD for consumers, and force an exodus to DVD-Audio or some similarly harder-to-enter market, then they are back in the driver's seat. The CD burner isn't a threat because you can copy their CD - it's because you can make your own.
Actually, as long as there are no namespace conflicts, this would be relatively easy to achieve using the Reflection APIs and dynamic class loading, I believe. Instead of a "java ..." invocation on the command line invoking an entirely new JVM process, it would be more of an access point to the class loader. Determining which version of a particular namespace a newly-loaded class wanted when there are conflicts is a tricky problem, though. I'm kind of surprised there hasn't been more work on this by the HotSpot team, although if I recall correctly, they were targeting some of this for the 1.5 release. Apple has made some progress on this in Mac OS X, as well.
...doesn't mean the Court will hear the case. They already refused to hear it ahead of the appellate court, so it is as likely as not that the Supreme Court will allow the penalty phase to be decided before hearing the case.
That's especially true when you figure the the SC probably doesn't want to have to craft a penalty, and would just send the case back down for a penalty phase if it upholds the lower courts' rulings.
The Findings of Fact and of Law are already pretty firmly entrenched in this case. I feel reasonably secure that when and if the case does make it to the Supreme Court that it will stand by the findings made by the lower courts.
when these kind of outages happen, of Peter Deutsch's 8 Fallacies of Distributed Computing:
This is, of course, why the idea of remote authentication being necessary to use your word processor is a bad thing. Heck, even losing something as innocuous as an instant messaging program brought thousands of people to a screeching halt for a week. It seems to me that Microsoft (although they're certainly not the only ones) seem to believe these 8 fallacies blindly, espcially 1, 4, and (they're hoping) 6.