Unfortunately, EJB is a pretty terrible model for scalability. It provides no tools for real workload partitioning or control of object distribution, at least as part of the specification standard. Ugh. And EJB-centric apps seem to almost universally be slower than anything else. People rarely stop to think "why do I want these two different things to be remote objects that interact with each other" and "where should these remote objects actually be located relative to each other". An app server framework needs to encourage/force developers to consider these issues, not try to encourage them to ignore them and make each and every object a separate EJB that talks to other EJBs.
An app server should provide reliability and scalability to an enterprise application without costing so much in terms of performance overhead for base-case scenarios that the system is just a resource hog. Maybe the EJB spec is better these days, I gave up on it after 1.1.
Aaron Brown used the word "mensch" on CNN last night (to describe Sammy Sosa, I believe), which indicates that it's seeping into the common English language. In English, it originated as a loan word from Yiddish. It's commonly used in New York City, among Jews and non-Jews alike, and can be heard in areas where a lot of New Yorkers live or where a lot of Jews live, like South Florida. If you live in other areas, like the West Coast, or the Midwest, you may never have heard it before. Its use in English (and I think in Yiddish) generally implies that somebody is a nice, kind-hearted, likeable person, not just a "person" or "man" which is the literal meaning in German, I believe.
I think the argument to make to your managers or to the legal department is "this code we worked on provides a useful component of some project or projects but A) It doesn't represent intellectual property that's core to our business itself and B) If we don't contribute the changes back to the community, we incur substantial additional costs of merging and maintaining our own separate code base for our version of the product." Point B is most likely of interest to your manager, and Point A is most likely of interest to whoever is responsible for legal things in the company (this will depend greatly on the nature and size of the company).
Agreed. A lot of the Slashdot community foolishly writes of Joe Sixpack. And they are partially right - a substantial plurality of users are true Joe Sixpacks that just want to check email and use a word processor. But somebody has to install and fix things for these Joe Sixpacks - there is no corporate IT department in the suburbs of East Bumblefuck where Joe Sixpack and his family live.
Often, the person who installs and fixes simple stuff is Joe Sixpack's son (kids seem much more ept at figuring out how to follow basic installation instructions for programs, installing drivers and the like). If it's not his son, maybe it's his neighbor. The problem is that figuring out how to do these things on Linux is often still just TOO damned hard (unless Joe is lucky enough to have a serious geek as a son or neighbor).
If you want to know who to blame I'll tell you. For hardware, it's partially the vendors faults for not supporting Linux, and partially the kernel hackers faults for not making stable binary drivers possible (there should be ONE AND ONLY ONE binary driver interface per major kernel release). Open Source is fabulous, but forcing people to compile their own drivers (or forcing different pre-compiled drivers for each and every minor distro release) is sheer idiocy.
For software, I would point to the library hell that is most modern Linuces. An application installer should just fucking work - not have one RPM for Mandrake 8.1, one for Mandrake 9, one for RedHat 9, a DEB for Woody, a DEB for.... and so on. This is WAY too confusing. When I go to download.com I download one version of an application that works on Windows 98+. Or maybe there are two version - one for Win 95/98/ME and one for Win 2k/XP. If Microsoft can version their OS without breaking binary compatibility in all their apps, why can't Linux? Yes, it's rhetorical, I know that it's perfectly possible, it's just that GLIBC changes its interface with each version, and the UI widgets change between GNOME 1.0 and 1.2, and so on. Ugh. The lack of universal versioning combined with complete lack of concern for binary compatibility ("oh, just recompile it!") creates a packaging and distribution nightmare for Joe Sixpack's neighbor or son when he wants to install something that didn't come on the Linux CD. He concludes Linux is cool looking, but just too complicated ("if I can't figure out how to do X, how the hell is my dad/neighbor going to use this").
Actually, my final analysis is that some judge will toss all these things up in the air, and see which weighs the most. If it came to court, which I doubt.
That was my assessment too.:) And the other part of my assessment was that it's almost certainly cheaper and less effort to rewrite the damned thing than to litigate with AOL (just get Justin to rewrite it when he's outta there, or if he can't for contractual reasons, then via clean room re-engineering). From my past experience, once you've written something like this once, you can generally recreate the whole thing from scratch in about 1/8th the time it took you the first time around.
Well, I differ with your interpretation. I am aware of the concept of apparent authority (I have run a business before), but apparent authority has limitations and standards associated with it. I think it's quite possible that if AOL sued somebody who used the WASTE source code and it went to trial, AOL might lose on this. On the other hand there are standards like "reasonable reliance" - at this point everybody who has downloaded and checked out WASTE probably has become aware that AOL did not consider the release authorized. A third party that knows the agent lacks actual authority cannot rely on the appearance of authority to seal a contract - in other words, it would certainly be foolhardy to rely on the GPL to protect your redistribution rights when the company declared that they considered the release unauthorized so rapidly.
I think this case is also less clear, because there's the question of whether the source code's appearance for a few hours on the website is sufficient to create contractually binding apparent authority. As other people have mentioned, clearly if a rogue employee sticks some source code up on a website with an Open Source license (the Microsoft Windows source code example people keep throwing around), I doubt any court would rule that its temporary presence on a website created apparent authority.
If anybody has more info on limitations on apparent authority and any lawyers have any thoughts on whether it is likely to apply here, I'm curious to hear it.
Copyright law distinguishes between a "user" and a "redistributor". You can feel free to use WASTE legally, I don't think they can revoke your right to use it, but _if_ the person who posted the source wasn't acting as a legal agent of the copyright holder, then the fact that they stuck the GPL on there means nothing. Only the copyright holder (or authorized agents) can do that. Whether Justin/Nullsoft was authorized at the time to make releases or whether they have contracts with AOL that say "all your code are belong to us" is another issue. My guess is AOL has enough lawyers to sufficiently CYA on their $86M acquisition of Nullsoft. But that's just a guess - without knowing the contracts in place, and really unless and until a judge decided on a case in a court relevant to the situation, there would probably be no clarity on whether WASTE was legally under the GPL or not.
In short, it's IP situation is muddied enough now that the code is tainted for good. Easier and cheaper to reverse engineer and rewrite than to hassle with AOL about it.
Re:Judge to do code review!
on
SCO SCO SCO!
·
· Score: 1
It's standard CYA talk. They have to disclaim that loudly and clearly, because eventually somebody will bite on their NDA offer and review the material under NDA, and they may very well say "we see no merit to the SCO case". In which case SCOX crashes and burns. I think SCO believes that by preemptively discrediting any experts who disagree with them that they can just invoke this shit again to calm investor panic if somebody gives their "evidence" the thumbs down (I'm sure they'll find some SCO/MS shills to give their evidence the thumbs up in any case).
I don't really see how those comments are damning. First of all, AFAIK, IBM created and owns the copyright to their own journaled file system and print driver - those were certainly cleaned of anybody else's IP before they were released as Linux kernel patches. There is some small chance that something got through that looks like parts of standard Unix header files or something, and this may be where SCO thinks they see a breach. But I sincerely doubt IBM would have taken chunks of SCO code without noticing it. Secondly, and equally importantly, the breach of contract that SCO seems to be alleging (once you dig through the layers upon layers of bullshit) is apparently with regards to Project Monterey IP jointly developed by IBM and SCO - not old, old SysV files. Unless those chunks just described were developed as part of Project Monterey and later folded into AIX, there's nothing terribly concerning. I don't know the history of this work, so I'm not certain by any means, but like Bruce said, IBM is usually VERY anal with approving things they open source - they don't usually "Nullsoft" shit out there and then change their mind a day later.
I'm somewhat surprised nobody has mentioned The Prisoner, the interesting (if sometimes very trippy) 60s series in which Patrick McGoohan, a former spy, is imprisoned in an isolated community designed to keep control of him. [spoiler alert] In the last episode of The Prisoner, McGoohan "escapes" finally and returns to London. And just as the closing credits roll, he is described (by a sign? - sorry, I saw some of these episodes when they did a rerun of the series about 10 years ago) as The Prisoner, implying that he was still imprisoned in a system designed to control and oppress him.
Anyway, The Prisoner is worth seeing, but don't see it out of order, and don't expect it all to make sense.
Yes, it will, somebody will take the illegally GPLed source and clean-room reverse it and you'll have OpenWaste on Sourceforge in a matter of months (weeks? days?). It just seems like a somewhat sad waste of code. If Justin or whatever his name is at Nullsoft wants to work on cool shit that he knows will get axed by AOLTW, he should set up a separate shop or organization to do such things outside of the context of Nullsoft so they won't get axed when they get released and the coding effort won't have been wasted. But then again maybe AOLTW wouldn't like that very much and would sue him for violating the takeover contract with Nullsoft.
Most likely he has some golden handcuff clause with clawback provisions (i.e. ownership of AOLTW shares reverts to AOLTW) if he leaves of his own free will, but that don't kick in if AOLTW terminates him without cause (maybe even with cause). Maybe he wants them to terminate him so he gets to walk _and_ keep his candy (err... shares).
This post shouldn't be modded offtopic. Legally, I believe it's a key issue (of course, this is just my opinion and IANAL). Ownership of the copyright of this code is a complex issue, undoubtedly, determined by a series of contracts between employees and their employers, and the companies involved. Just because we might say in the common parlance that "AOL owns Nullsoft" doesn't necessarily mean much. AOL might own every share of Nullsoft stock, but if they delegate management of Nullsoft to Nullsoft executives, those managers have the right and obligation to manage the company as they see fit (within the bounds of standard fiduciary obligations - which are complex). The managers do have the obligation to represent shareholder interests with respect to company assets - but that is a contractual and fiduciary responsibility issue, and does not retroactively impact who had the right to license that code.
As far as I know, in the absence of overriding contracts regarding copyright holdings of Nullsoft, Inc. that automatically assign such copyrights to AOLTW and prohibit sale or trade of rights in those copyrighted materials without explicit authorization of AOLTW, I believe Nullsoft management would have acted as legal agents of Nullsoft Inc. with respect to copyrighted materials when they posted them on the Nullsoft web site with license and copyright notices attached. If AOL failed to put greater contractual and procedural controls in place, that's their problem, and they could take it up in court with the individual managers or corporate personage of Nullsoft, Inc.
Then again, after the Gnutella fiasco, if AOLTW _didn't_ have explicit contracts in place giving them assignment and control of all copyrighted Nullsoft works, they are idiots.
You clearly are missing the point here. Obviously, showing that trade secrets were not kept secret does not obviate the potential for copyright infringement. Luckily, as you point out in problem 3, we are fairly confident that there was no copyright infringement - and if there was, it is fairly straightforward to remedy it.
The point is that if we knock out the possibility of a cogent trade secret argument being made against IBM/Linux, that forces SCO to prove copyright infringement and make a case around that. We don't think they can, because they refuse to explain where it happened and because there's really no evidence that anybody needed to copy lines of code from SCO to make Linux work. The fear is that they might try to swerve this into a trade secret violation case, which would make it easier to cast doubt over the IP heritage of Linux, because to make a logical argument they would only need to show that 1) Somebody at IBM who worked on AIX was subject to NDA with SCO, 2) Said person worked with people who contributed code to parts of Linux that gave it features it didn't previously have, and 3) Said features constituted the use of "trade secret" material from SCO/UnixWare.
Note that the vagueness of SCOs claims make this a rather founded fear IMO, since they keep mentioning "IP violations" more often than they specifically have mentioned copyright violations, though they have also mentioned those.
Cypher: Well you have to. The compilers work for the construct program. But there's way too much information to decode the Matrix. You get used to it. I...I don't even see the code. All I see is an array, function pointer, integer. Hey, you uh... want a drink?
Neo: Sure.
Cypher: You know, I know what you're thinking, because right now I'm thinking the same thing. Actually, I've been thinking it ever since I got here. Why, oh why didn't I sell my VA Linux stock?... Good shit, huh? Cowboy Neal makes it. It's good for two things, degreasing Perl code and killing brain cells.
Yes, but the question is did they continue to redistribute the GPLed code in question AFTER they knew their "stolen code" was in there. That implies that they continued to accept the GPL license on that code, since they were "aware" of their own copyrighted code in the other parts of the work. Whether or not they "marked" it with a copyright notice is irrelevant as I understand the law (IANAL), but whether they were aware of and intended to license said code under the GPL is eminently relevant. Their continued distribution under the GPL of a work they knew to be composed of their own copyrighted code and others copyrighted code implies that they intended to place that code under the GPL at that time, and only after the fact, decided that it was more beneficial for them to take the "it was stolen, let's sue everybody" approach. They can only do that if they can show they never intended to release the code under the GPL - and it seems like admitting that they waited many months before stopping their own GPLed distribution of the relevant source code certainly does not strengthen their case. It likely hinges on whether a judge deems that they took reasonable efforts to prevent further dissemination of that code by their own organization under the GPL - how many months of lag time is reasonable in your mind?
Right, which is why I don't quite agree with the grandparent post. You do have to be a legal agent of a company with respect to their held copyrights to release copyrighted material _owned by the company_ under any license. Remember that the GPL only holds if in fact the licensing entity holds the copyright. As to who is a legally designated agent of a company with respect to the company's copyrighted materials, I'm sure there's some existing jurisprudence out there on it. Obviously if a company was selling a product in stores that included GPLed code, and distributing the source code for said product from an FTP server, it would be next to impossible to argue convincingly in court that the company as an entity did not approve that distribution - there was no rogue employee who misappropriated copyrighted material and slapped a GPL on it in this case.
In any case, if the copyrighted material they supposedly gave up rights to had been previously surreptitiously added to a GPLed work without their consent, I don't think the fact that they later redistributed an amalgamated work under the GPL will be considered their consenting to the GPL licensing of their earlier copyrighted work, only of the work they contributed during the period they were distributing the GPLed work. IANAL, but that's my best guess as to how a (reasonable) court would interpret copyright law and the GPL in this case. A lot of the law revolves around how a "reasonable person" would interpret the situation, and I don't think any reasonable person would believe that by distributing a GPLed work, they could be relinquishing rights to other works of their that had been previously misappropriated and inserted surreptitiously into the GPLed work by another party.
Of course, I think it's moot anyway, because SCO has failed utterly to explain exactly where such code is. And none of this applies if it's not a copyright they claim was infringed but a trade secret or patent, which are governed by other laws entirely. Unfortunately, everything is speculation since SCO refuses to disclose what actual IP violations have occured.
Without cabs, the city would not be manageable. I agree that private individuals with cars aren't really needed (except for getting into and out of NYC, which is the only thing I use mine for). But not every place in the city is very accessible by public transportation, and because our transit system grew organically, it is vastly more efficient between certain locales than between others, without regard for distances involved.
I'm not saying NYC couldn't be adapted to make it truly car free (it could), but I don't think it's quite as trivial as you make it sound. Of course, you could never eliminate the trucks and delivery vehicles that supply the city. And people like having the convenience of cabs - especially if you need to get more stuff than you can carry from point A to point B, or if you are in a hurry (a reality of the modern world, that people just never had to deal with in the pre-automobile era, because expectations were different).
In short, I think you could effectively get rid of private passenger car traffic in NYC, with some work, without immensely inconveniencing those who live in the city. But I'm not sure how much it would really improve things, and it would clearly inconvenience immensely those who visit the city via car, or who commute into the city from areas not served by public transportation. Don't forget that even NYC in large part relies on people who don't reside within its boundaries to work its many service jobs.
P2P is not infringement. P2P is a distribution technology for copying data. In reality, are P2P networks exploited to get access to and share access to copyrighted materials? Yes. But that's different, and I mean EXTREMELY different, from making running a server with copyrighted content illegal.
In that case, you are talking about making copyright infringement illegal in all cases, tweaking a law that had a noteable loophole in it. In this case, you are talking about making illegal a technology or distribution mechanism, rather than the people who break the law with it. Nobody ever suggested making ftp illegal, though there are tons of warez and MP3 ftp sites out there. Nobody ever suggested making HTTP illegal or banning Usenet because of the large quantity of kiddie porn out there (jebus man, look at some of those Usenet newsgroups - used for nothing but moving around illegal files). Why is P2P different? How would you propose fighting it legislatively? Making indexing illegal? Well, it already is, see Napster for a case in point. Make it illegal to produce software that allows anonymous distribution of files? Making algorithms illegal doesn't work. Just see what happened with the RSA shit - people moved all work on it offshore until it came off patent and the US government got their head outta their asses on the encryption legislation stuff. Yes, technology can be scary, and it can be used for illegal purposes, but making bits and bytes illegal because they can be used to infringe copyright? Making a whole class of protocols illegal because they involve indexing information and transfering data between a large number of participants instead of centralizing all data? There are a lot of people who will fight that to the bitter end, and I am glad to count myself among them.
Why does everybody in C++ land feel the need to roll their own language every time? Why does everything have to be a FLOAT instead of a float? I mean, I understand that in many cases, it's for portability, but it seems like some of these features are completely overused, and absolutely mystical.
Preprocessors are, I think, way overused. do {... }while(0) my ass.
Agree with you 100% here. Generics are a huge improvement, and I cannot possibly say how many times I have longed for them in Java.
Recently I've been working in C++ again after many years away from it, and the things I miss most are the simplicity of learning a new API (nice, beautiful, clear, simple, hyperlinked Javadocs - and when the fuck will C++ people learn that you don't have to use every feature of the language in a public API), and beauty of IntelliJ. I mean, Visual C++ has some nice features (I find its debugging to be simple, and powerful), but god do I miss the beautiful syntax highlighting, drop-down completion and refactoring capabilities of IntelliJ. I, for one, would appreciate having an "IntelliC++". Having to manually reorganize code again feels like moving back to the fricking stone age. I'm sure the syntactic complexity of C++ makes this harder to achieve in a generalizable programmatic fashion. I think IntelliJ has made one of the strongest cases yet for Java by showing how the language lends itself to powerful IDE features that I haven't seen anywhere else (outside of the Java world).
I gotta say, I'm positively impressed. C# and Java are feeding features off of each other. Iterators, enumerators and autoboxing/unboxing - hmm, I think I've seen those before. Java 1.5 starts to look more and more like C#.
Some of this other stuff has been going on for ages - I'm curious about the metadata/declarative programming features. I've developed complete code generation systems for Java - the number of situations that require reams of very repetitive code in your average large-scale Java app is huge, and much of that can be automated. It would be great to see a consistent, standard system for doing this built in as a language feature rather than having hundreds of home-rolled systems everywhere, but the nature of many problems lends themselves to home-rolling. I can't tell you how many times I've written simple SAX parsers that spew out Java code, and rolled up an ant target to run it on some schema and package up the result. It's not clear from this brief example whether this is only for XML/RPC or whether there's an extensible metadata system that you can build on top of.
Then again, we should be careful not to roll too much into the language itself. I think generics, autoboxing, and enumerators are fabulous. Iterators I could give less of a shit over. Other stuff is great, but I question whether extending the language is the right mechanism. Much of the power and flexibility of Java comes from its simplicity. And most importantly, the ease of reading other people's code - there's far less stylistic variation because there are only N ways to do a task, rather than N! ways to do it, like in C++ (don't get me wrong, for a lot of tasks, I'd never dream of doing them in Java, like 3D programming). But it takes me about 1/4 to 1/3 the time to assimilate and learn a new Java API or library as it does to learn your average C++ API or library, and that's the appeal of Java.
Of course, I'll be thanking the gods that I never have to deal with the fugliness of casting and wrapping to move stuff between typed arrays and untyped Lists again.
I agree with you. If they called it the "Software Freedom Foundation" instead of "Free Software Foundation", I think it would get a clearer point across to those who haven't been subjected to the RMS doctrine. Of course, you can't really call it "Freedom Software" (well you can, but then people will laugh at you and ask if you want some French... er... Freedom Fries with that).
Yeah, I mean nobody will ever know who wrote ReiserFS - it sounds like this guy has a bit of a chip on his shoulder and ego problem going on. I think the whole "mandatory credit" thing was tried with the original BSD license, and it just becomes an onerous task because of the large number of contributers, copyright holders, and so on.
I am all for giving credit where it is due, and I think commercial Linux distros should make a serious effort to thank those who've made their products possible, and contribute back to the community. Of course ego fulfillment is one of the major motivators for working on Free Software projects - but it's not the only one, and making the Free Software so onerous to use that it becomes onerous to use is not the right way forward. The motivation to work on something that other people will build on top of and make into something better as a whole than one person could do themselves is a big part of the appeal of Free Software. Let's not pollute important projects with licenses demanding all sorts of overblown ego fulfillment and stroking. A brief copyright note in the kernel message, or a Help->About dialog in a GUI app is plenty of acknowledgement for the majority of Free Software authors.
Not a strawman. You seemed to be suggesting that Iraqis should have the right to "vote in" (or have foisted upon them by a violent and vocal group) an Islamic fundamentalist regime, which I was pointing out is a totalitarian regime in its own right, and thus one that disallows moral choice on the part of individuals. Are we really serving the interests of the Iraqi people by letting them run back into the embrace of a different kind of nice simplistic totalitarianism? What these people need is education and awareness of alternatives, and they need to have a military force keeping democratic institutions intact (clearly, this won't be our military after we leave, it will be their own). It's not really that different from what Ataturk did in Turkey.
It's true that our forefathers set up their own government. Our forefathers were also well educated intellectuals. Religious fundamentalists had no place in our government. Furthermore, our forefathers were NOT unabashed majority-rule democracy fans. That's why there are so many checks and balances on raw democracy in an American-style Republic - the unwashed masses really don't always know what's best for them, and that's with a real secular public education system (admittedly a very sucky one).
I certainly think the Iraqis should set up their own government (and I don't think I ever said otherwise). The question is which Iraqis are part of that process - the educated intellectuals, or the people who run around waving AK-47s at the behest of their Mullah? Because the latter have no problem using fear and intimidation and groupthink to prevent the masses from voting for the former. Letting the strongest win in Iraq without a thought as to how to build a sustainable democracy that represents the people's will while protecting the minority from the majority (i.e. protecting everyone's rights) is hardly an example of colonialism or imperialism. Saying the Iraqis can't set it up entirely by themselves is not patronizing, it just recognizes honestly the condition of an ill-educated populace combined with the anarchic aftermath from the collapse of years of totalitarian rule.
I said nothing about being beholden to the US. Would you say the Japanese government is beholden to the US? What about Germany? We forced the disarmament and democratization of these countries after they lost WWII. It's our right to prevent them from threatening us and the rest of the world, but that doesn't mean installing puppet governments, it means setting bounds. And it's our responsibility to ensure that for the first several years at least, democracy gets a reasonable chance in Iraq.
I know nothing about forcing American management onto Iraqi oil fields - obviously, I don't support that. I never defended George Bush et. al. in terms of original motivations, since I think it's a rather complex issue, and there are clearly quite a mix of motivations, and they vary from individual to individual (I certainly don't think you can talk about the administration as having monolithically "good" or "bad" motivations for this war). But the colonialism stuff is getting quite tiresome. Let's save it for the actually bad stuff the US government does, not for trying to ensure that Iraqi democracy gets a fair shake.
Slashdot Mirror
An app server should provide reliability and scalability to an enterprise application without costing so much in terms of performance overhead for base-case scenarios that the system is just a resource hog. Maybe the EJB spec is better these days, I gave up on it after 1.1.
Aaron Brown used the word "mensch" on CNN last night (to describe Sammy Sosa, I believe), which indicates that it's seeping into the common English language. In English, it originated as a loan word from Yiddish. It's commonly used in New York City, among Jews and non-Jews alike, and can be heard in areas where a lot of New Yorkers live or where a lot of Jews live, like South Florida. If you live in other areas, like the West Coast, or the Midwest, you may never have heard it before. Its use in English (and I think in Yiddish) generally implies that somebody is a nice, kind-hearted, likeable person, not just a "person" or "man" which is the literal meaning in German, I believe.
I think the argument to make to your managers or to the legal department is "this code we worked on provides a useful component of some project or projects but A) It doesn't represent intellectual property that's core to our business itself and B) If we don't contribute the changes back to the community, we incur substantial additional costs of merging and maintaining our own separate code base for our version of the product." Point B is most likely of interest to your manager, and Point A is most likely of interest to whoever is responsible for legal things in the company (this will depend greatly on the nature and size of the company).
Trinity: Was it an orange bar or did it only look orange?
Neo: I don't know.
Trinity: An orange bar is a glitch in the Slashcode. It happens when the machines change things.
Often, the person who installs and fixes simple stuff is Joe Sixpack's son (kids seem much more ept at figuring out how to follow basic installation instructions for programs, installing drivers and the like). If it's not his son, maybe it's his neighbor. The problem is that figuring out how to do these things on Linux is often still just TOO damned hard (unless Joe is lucky enough to have a serious geek as a son or neighbor).
If you want to know who to blame I'll tell you. For hardware, it's partially the vendors faults for not supporting Linux, and partially the kernel hackers faults for not making stable binary drivers possible (there should be ONE AND ONLY ONE binary driver interface per major kernel release). Open Source is fabulous, but forcing people to compile their own drivers (or forcing different pre-compiled drivers for each and every minor distro release) is sheer idiocy.
For software, I would point to the library hell that is most modern Linuces. An application installer should just fucking work - not have one RPM for Mandrake 8.1, one for Mandrake 9, one for RedHat 9, a DEB for Woody, a DEB for.... and so on. This is WAY too confusing. When I go to download.com I download one version of an application that works on Windows 98+. Or maybe there are two version - one for Win 95/98/ME and one for Win 2k/XP. If Microsoft can version their OS without breaking binary compatibility in all their apps, why can't Linux? Yes, it's rhetorical, I know that it's perfectly possible, it's just that GLIBC changes its interface with each version, and the UI widgets change between GNOME 1.0 and 1.2, and so on. Ugh. The lack of universal versioning combined with complete lack of concern for binary compatibility ("oh, just recompile it!") creates a packaging and distribution nightmare for Joe Sixpack's neighbor or son when he wants to install something that didn't come on the Linux CD. He concludes Linux is cool looking, but just too complicated ("if I can't figure out how to do X, how the hell is my dad/neighbor going to use this").
Actually, my final analysis is that some judge will toss all these things up in the air, and see which weighs the most. If it came to court, which I doubt.
:) And the other part of my assessment was that it's almost certainly cheaper and less effort to rewrite the damned thing than to litigate with AOL (just get Justin to rewrite it when he's outta there, or if he can't for contractual reasons, then via clean room re-engineering). From my past experience, once you've written something like this once, you can generally recreate the whole thing from scratch in about 1/8th the time it took you the first time around.
That was my assessment too.
I think this case is also less clear, because there's the question of whether the source code's appearance for a few hours on the website is sufficient to create contractually binding apparent authority. As other people have mentioned, clearly if a rogue employee sticks some source code up on a website with an Open Source license (the Microsoft Windows source code example people keep throwing around), I doubt any court would rule that its temporary presence on a website created apparent authority.
If anybody has more info on limitations on apparent authority and any lawyers have any thoughts on whether it is likely to apply here, I'm curious to hear it.
In short, it's IP situation is muddied enough now that the code is tainted for good. Easier and cheaper to reverse engineer and rewrite than to hassle with AOL about it.
It's standard CYA talk. They have to disclaim that loudly and clearly, because eventually somebody will bite on their NDA offer and review the material under NDA, and they may very well say "we see no merit to the SCO case". In which case SCOX crashes and burns. I think SCO believes that by preemptively discrediting any experts who disagree with them that they can just invoke this shit again to calm investor panic if somebody gives their "evidence" the thumbs down (I'm sure they'll find some SCO/MS shills to give their evidence the thumbs up in any case).
I don't really see how those comments are damning. First of all, AFAIK, IBM created and owns the copyright to their own journaled file system and print driver - those were certainly cleaned of anybody else's IP before they were released as Linux kernel patches. There is some small chance that something got through that looks like parts of standard Unix header files or something, and this may be where SCO thinks they see a breach. But I sincerely doubt IBM would have taken chunks of SCO code without noticing it. Secondly, and equally importantly, the breach of contract that SCO seems to be alleging (once you dig through the layers upon layers of bullshit) is apparently with regards to Project Monterey IP jointly developed by IBM and SCO - not old, old SysV files. Unless those chunks just described were developed as part of Project Monterey and later folded into AIX, there's nothing terribly concerning. I don't know the history of this work, so I'm not certain by any means, but like Bruce said, IBM is usually VERY anal with approving things they open source - they don't usually "Nullsoft" shit out there and then change their mind a day later.
Anyway, The Prisoner is worth seeing, but don't see it out of order, and don't expect it all to make sense.
Most likely he has some golden handcuff clause with clawback provisions (i.e. ownership of AOLTW shares reverts to AOLTW) if he leaves of his own free will, but that don't kick in if AOLTW terminates him without cause (maybe even with cause). Maybe he wants them to terminate him so he gets to walk _and_ keep his candy (err... shares).
As far as I know, in the absence of overriding contracts regarding copyright holdings of Nullsoft, Inc. that automatically assign such copyrights to AOLTW and prohibit sale or trade of rights in those copyrighted materials without explicit authorization of AOLTW, I believe Nullsoft management would have acted as legal agents of Nullsoft Inc. with respect to copyrighted materials when they posted them on the Nullsoft web site with license and copyright notices attached. If AOL failed to put greater contractual and procedural controls in place, that's their problem, and they could take it up in court with the individual managers or corporate personage of Nullsoft, Inc.
Then again, after the Gnutella fiasco, if AOLTW _didn't_ have explicit contracts in place giving them assignment and control of all copyrighted Nullsoft works, they are idiots.
The point is that if we knock out the possibility of a cogent trade secret argument being made against IBM/Linux, that forces SCO to prove copyright infringement and make a case around that. We don't think they can, because they refuse to explain where it happened and because there's really no evidence that anybody needed to copy lines of code from SCO to make Linux work. The fear is that they might try to swerve this into a trade secret violation case, which would make it easier to cast doubt over the IP heritage of Linux, because to make a logical argument they would only need to show that 1) Somebody at IBM who worked on AIX was subject to NDA with SCO, 2) Said person worked with people who contributed code to parts of Linux that gave it features it didn't previously have, and 3) Said features constituted the use of "trade secret" material from SCO/UnixWare.
Note that the vagueness of SCOs claims make this a rather founded fear IMO, since they keep mentioning "IP violations" more often than they specifically have mentioned copyright violations, though they have also mentioned those.
Cypher: Well you have to. The compilers work for the construct program. But there's way too much information to decode the Matrix. You get used to it. I...I don't even see the code. All I see is an array, function pointer, integer. Hey, you uh... want a drink?
Neo: Sure.
Cypher: You know, I know what you're thinking, because right now I'm thinking the same thing. Actually, I've been thinking it ever since I got here. Why, oh why didn't I sell my VA Linux stock?... Good shit, huh? Cowboy Neal makes it. It's good for two things, degreasing Perl code and killing brain cells.
Yes, but the question is did they continue to redistribute the GPLed code in question AFTER they knew their "stolen code" was in there. That implies that they continued to accept the GPL license on that code, since they were "aware" of their own copyrighted code in the other parts of the work. Whether or not they "marked" it with a copyright notice is irrelevant as I understand the law (IANAL), but whether they were aware of and intended to license said code under the GPL is eminently relevant. Their continued distribution under the GPL of a work they knew to be composed of their own copyrighted code and others copyrighted code implies that they intended to place that code under the GPL at that time, and only after the fact, decided that it was more beneficial for them to take the "it was stolen, let's sue everybody" approach. They can only do that if they can show they never intended to release the code under the GPL - and it seems like admitting that they waited many months before stopping their own GPLed distribution of the relevant source code certainly does not strengthen their case. It likely hinges on whether a judge deems that they took reasonable efforts to prevent further dissemination of that code by their own organization under the GPL - how many months of lag time is reasonable in your mind?
In any case, if the copyrighted material they supposedly gave up rights to had been previously surreptitiously added to a GPLed work without their consent, I don't think the fact that they later redistributed an amalgamated work under the GPL will be considered their consenting to the GPL licensing of their earlier copyrighted work, only of the work they contributed during the period they were distributing the GPLed work. IANAL, but that's my best guess as to how a (reasonable) court would interpret copyright law and the GPL in this case. A lot of the law revolves around how a "reasonable person" would interpret the situation, and I don't think any reasonable person would believe that by distributing a GPLed work, they could be relinquishing rights to other works of their that had been previously misappropriated and inserted surreptitiously into the GPLed work by another party.
Of course, I think it's moot anyway, because SCO has failed utterly to explain exactly where such code is. And none of this applies if it's not a copyright they claim was infringed but a trade secret or patent, which are governed by other laws entirely. Unfortunately, everything is speculation since SCO refuses to disclose what actual IP violations have occured.
I'm not saying NYC couldn't be adapted to make it truly car free (it could), but I don't think it's quite as trivial as you make it sound. Of course, you could never eliminate the trucks and delivery vehicles that supply the city. And people like having the convenience of cabs - especially if you need to get more stuff than you can carry from point A to point B, or if you are in a hurry (a reality of the modern world, that people just never had to deal with in the pre-automobile era, because expectations were different).
In short, I think you could effectively get rid of private passenger car traffic in NYC, with some work, without immensely inconveniencing those who live in the city. But I'm not sure how much it would really improve things, and it would clearly inconvenience immensely those who visit the city via car, or who commute into the city from areas not served by public transportation. Don't forget that even NYC in large part relies on people who don't reside within its boundaries to work its many service jobs.
In that case, you are talking about making copyright infringement illegal in all cases, tweaking a law that had a noteable loophole in it. In this case, you are talking about making illegal a technology or distribution mechanism, rather than the people who break the law with it. Nobody ever suggested making ftp illegal, though there are tons of warez and MP3 ftp sites out there. Nobody ever suggested making HTTP illegal or banning Usenet because of the large quantity of kiddie porn out there (jebus man, look at some of those Usenet newsgroups - used for nothing but moving around illegal files). Why is P2P different? How would you propose fighting it legislatively? Making indexing illegal? Well, it already is, see Napster for a case in point. Make it illegal to produce software that allows anonymous distribution of files? Making algorithms illegal doesn't work. Just see what happened with the RSA shit - people moved all work on it offshore until it came off patent and the US government got their head outta their asses on the encryption legislation stuff. Yes, technology can be scary, and it can be used for illegal purposes, but making bits and bytes illegal because they can be used to infringe copyright? Making a whole class of protocols illegal because they involve indexing information and transfering data between a large number of participants instead of centralizing all data? There are a lot of people who will fight that to the bitter end, and I am glad to count myself among them.
Preprocessors are, I think, way overused. do {
Recently I've been working in C++ again after many years away from it, and the things I miss most are the simplicity of learning a new API (nice, beautiful, clear, simple, hyperlinked Javadocs - and when the fuck will C++ people learn that you don't have to use every feature of the language in a public API), and beauty of IntelliJ. I mean, Visual C++ has some nice features (I find its debugging to be simple, and powerful), but god do I miss the beautiful syntax highlighting, drop-down completion and refactoring capabilities of IntelliJ. I, for one, would appreciate having an "IntelliC++". Having to manually reorganize code again feels like moving back to the fricking stone age. I'm sure the syntactic complexity of C++ makes this harder to achieve in a generalizable programmatic fashion. I think IntelliJ has made one of the strongest cases yet for Java by showing how the language lends itself to powerful IDE features that I haven't seen anywhere else (outside of the Java world).
Some of this other stuff has been going on for ages - I'm curious about the metadata/declarative programming features. I've developed complete code generation systems for Java - the number of situations that require reams of very repetitive code in your average large-scale Java app is huge, and much of that can be automated. It would be great to see a consistent, standard system for doing this built in as a language feature rather than having hundreds of home-rolled systems everywhere, but the nature of many problems lends themselves to home-rolling. I can't tell you how many times I've written simple SAX parsers that spew out Java code, and rolled up an ant target to run it on some schema and package up the result. It's not clear from this brief example whether this is only for XML/RPC or whether there's an extensible metadata system that you can build on top of.
Then again, we should be careful not to roll too much into the language itself. I think generics, autoboxing, and enumerators are fabulous. Iterators I could give less of a shit over. Other stuff is great, but I question whether extending the language is the right mechanism. Much of the power and flexibility of Java comes from its simplicity. And most importantly, the ease of reading other people's code - there's far less stylistic variation because there are only N ways to do a task, rather than N! ways to do it, like in C++ (don't get me wrong, for a lot of tasks, I'd never dream of doing them in Java, like 3D programming). But it takes me about 1/4 to 1/3 the time to assimilate and learn a new Java API or library as it does to learn your average C++ API or library, and that's the appeal of Java.
Of course, I'll be thanking the gods that I never have to deal with the fugliness of casting and wrapping to move stuff between typed arrays and untyped Lists again.
I agree with you. If they called it the "Software Freedom Foundation" instead of "Free Software Foundation", I think it would get a clearer point across to those who haven't been subjected to the RMS doctrine. Of course, you can't really call it "Freedom Software" (well you can, but then people will laugh at you and ask if you want some French... er... Freedom Fries with that).
I am all for giving credit where it is due, and I think commercial Linux distros should make a serious effort to thank those who've made their products possible, and contribute back to the community. Of course ego fulfillment is one of the major motivators for working on Free Software projects - but it's not the only one, and making the Free Software so onerous to use that it becomes onerous to use is not the right way forward. The motivation to work on something that other people will build on top of and make into something better as a whole than one person could do themselves is a big part of the appeal of Free Software. Let's not pollute important projects with licenses demanding all sorts of overblown ego fulfillment and stroking. A brief copyright note in the kernel message, or a Help->About dialog in a GUI app is plenty of acknowledgement for the majority of Free Software authors.
It's true that our forefathers set up their own government. Our forefathers were also well educated intellectuals. Religious fundamentalists had no place in our government. Furthermore, our forefathers were NOT unabashed majority-rule democracy fans. That's why there are so many checks and balances on raw democracy in an American-style Republic - the unwashed masses really don't always know what's best for them, and that's with a real secular public education system (admittedly a very sucky one).
I certainly think the Iraqis should set up their own government (and I don't think I ever said otherwise). The question is which Iraqis are part of that process - the educated intellectuals, or the people who run around waving AK-47s at the behest of their Mullah? Because the latter have no problem using fear and intimidation and groupthink to prevent the masses from voting for the former. Letting the strongest win in Iraq without a thought as to how to build a sustainable democracy that represents the people's will while protecting the minority from the majority (i.e. protecting everyone's rights) is hardly an example of colonialism or imperialism. Saying the Iraqis can't set it up entirely by themselves is not patronizing, it just recognizes honestly the condition of an ill-educated populace combined with the anarchic aftermath from the collapse of years of totalitarian rule.
I said nothing about being beholden to the US. Would you say the Japanese government is beholden to the US? What about Germany? We forced the disarmament and democratization of these countries after they lost WWII. It's our right to prevent them from threatening us and the rest of the world, but that doesn't mean installing puppet governments, it means setting bounds. And it's our responsibility to ensure that for the first several years at least, democracy gets a reasonable chance in Iraq.
I know nothing about forcing American management onto Iraqi oil fields - obviously, I don't support that. I never defended George Bush et. al. in terms of original motivations, since I think it's a rather complex issue, and there are clearly quite a mix of motivations, and they vary from individual to individual (I certainly don't think you can talk about the administration as having monolithically "good" or "bad" motivations for this war). But the colonialism stuff is getting quite tiresome. Let's save it for the actually bad stuff the US government does, not for trying to ensure that Iraqi democracy gets a fair shake.