I suspect that this will play out as a trend in Open Source and quasi-open source development projects lead by the likes of Sun and Netscape(Aol). Large companies will decide to release the source to something, and fail to see the value which Open Source projects employ in the release-early-release-often model. This is unfortunate, since releasing a not-quite-there-yet, but promising version as "alpha" would get other people finishing features and fixing bugs that they needed to scratch. This is what I think Mozilla has done wrong. They release these non-functional Mx releases as code benchmarks, but fail to work on getting a nice looking, feature light version out the door so that large masses of people will USE it, and eventually have some of them join the project. They've already forgotten how fast the Australians were able to impliment real crypto back when everyone thought that Mozilla would be a usable browser in 2 months.
NOTE: I'm not involved with either effort, but if my statements are incorrect because of my lack of inside knowledge, then they are likely to -- even more accurately -- reflect what the average Linux/Open Source user is thinking....
why set up a Linux is not X/BSD/PD/Artistic/GNU/Linux page when nobody claimed that Linux is all those things? all he said was that Linux is a kernel
Go read the document. He's not saying "since we can call it anything why not add GNU in." That would be damn near reasonable. He says "you're running a modified version of the GNU system."
This is doublespeak worthy of Ed Muth, not RMS. There never was a GNU system, and this statement discounts the many large chunks of your average Linux distribution that are not derived from GNU's attempts to create a system (e.g. X, wu-ftpd, tcsh, ftp, rlogin, inn, nvi, vim, etc, etc, etc). If we discount "GNU homages" like GNOME, which are not actually part of the GNU effort, but use the name, very little of your average distribution is GNU. There's the compiler, basic C library, EMACS and a slew of utilities. That's really about it. Wonderful contributions, all, and I respect them. Before RMS started on this rant, I was happily pointing out the importance of GNU to Linux, but now I downplay it because this territorial attitude really turns me off.
I've never been able to think of a reasonable justification for calling Linux GNU/Linux. The only reasonable way to modify what we call it to properly attribute the external efforts that have helped Linux get started would be to call it some silly, long string of names that cited the BSD, Public Domain, MIT/X, Perl/Artistic, GNU and many other contributions. Let's just not go there, Ok. The thing is Linux, and it runs a lot of damn fine programs. If Stallman wants to name Linux, he should create a distribution (which, arguably, he has done, since the Debian folks were happy to take up his re-naming).
I can just see the "next step" where we try to rename "Solaris" to "GNU/Solaris" because so many people install GNU utilities on every Solaris box they touch.... Just scary.
Woefully, what started out as an excellent rebuttal by Stallman (some of the most coherent writing I've seen by him so far) falls all over itself in the end with these few holes:
Fortunately, Eric Raymond's doctrinaire, almost Marxian vision of economic determinism is not realistic
I actually had to do a double-take through the document to see if he'd prefaced this at all. Nope, this is just a random, unexplained cheap-shot at ESR, who is not mentioned anywhere else in the document.
If you are using Linux, Linus Torvalds' kernel, you are most likely using it in conjunction with the GNU system. This combination, the GNU/Linux operating system
This is getting very old. I'm tempted to set up a "Linux is not X/BSD/PD/Artistic/GNU/Linux" page and just post the URL in every forum that refers to yet another Stallman article. I'll not go into it more, here, though. I assume people have better things to do.
I suggest avoiding the term intellectual property [...] It takes for granted that these things should be a kind of property [...] It is too big a generalization, and is likely to tempt all nonlawyers into assuming that patents and copyrights are similar
The term does not imply, state or suggest that there is a "should" involved. In fact I can quite easily say "some forms of intellectual property, such as software patents, are foolish and should not be property in the first place". Look Ma, I used a word!
Patents and copyrights are similar. If they weren't we would not be in the mess that we are now. The only differences are scope, enforcability and duration. The basic idea is that a non-tangible can be property, which in the interest of innovation and creativity is protected by the federal government. Go ahead tell me that I just exclusively defined copyrights or patents.
In the end, it's classic Stallman. The man can write damn fine code, but I wish he'd stop trying to talk to people. Others such as ESR, Larry Wall and Linus Torvalds really do do a much better job of getting the points across. Overall, though the early points in the article were fair. Stallman does get a lot of flack over the GPL, which is quite a reasonable document when you consider things like the MS EULA (which I will never agree to, even though I use their software as an end-user at work) define the rest of the playing field.
I know this is percieved as "political correctness", and maybe it is, but are there other angles that can benefit us all?
Imagine, if you will, a World Wide Web where everything was delivered in a, say, XML format where enough "structured content" information was delivered to a browser that the browser could reasonably read the content to a blind user (who could ask for things like "read section headings and/or headlines from this page".
What else could you possibly use this for...? How about a dynamic browser that you can configure to do things like "hide all "Article" content" so that sites like Slashdot would show up as just headlines. How about asking your browser to display no images, but still having enough structure information to display the page correctly.... The blind will end up pushing the Net back into what it was intended to be: an information delivery mechanism, not a hypertext ad magazine.
Clearly content structure is not universal, and not every site is a collection of articles like Slashdot, but I think the example is powerful enough to merit finding a way to make the general case work....
Hidding the existance of a message is called steganography. Its more common to high the message in a single image or MP3 as they have more bits to obscure the payload.
Absolutely. And, they have more bits to analyze. I'll bet I could write a program that would identify images very quickly that have been so modified. How? Just a simple statistical analysis against a baseline sample of such images. Gee, the low bits of this image are very unlikely for a JPEG. I'll just take a closer look....
Yeah, that kind of analysis is probably part of the NSA's automatic filtering already, and you can bet that they've got a full USENET feed watching the binary newsgroups for just this sort of thing. Funny that we would pay so much money in taxes for our government to collect and analyze porn;-)
The NSA has been around (we think, and have good reason to believe) just about as long as the CIA (e.g. since late WWII). Some posts contradict this with suggestions that the NSA has been around as long as Morse. This may be true if you count other covert agencies, and there may be some legacy of monitoring infrastructure, there, but I don't think so. The NSA was the result of a joit US/UK attempt to make sure that the kinds of things that happened in WWII could never happen again (e.g. Pearl Harbor and the V2s). A funny thing to note is that, while I think that the NSA needs to be slapped down, and hard; I really do think that this kind of monitoring is generally a good thing. It just needs much tighter control and oversight. Privacy groups like the EFF should certainly be consulted, and backdoors like the FBI proposals should just be shot; but the idea is sound. We can't afford to miss out on a slip-up, and if we're ever in another large-scale war, we certainly want the kind of SIGINT and code-breaking capabilities that we had during WWII.
The NSA's mandate is international, and efforts like FedNET underscore the likelyhood that the NSA does not monitor STRICTLY domestic communications. Of course definitions get tricky, here. It's most likely that any communications that involve long-distance radio or microwave including satellite are monitored. Also, if your packets happen to bounce through Canada or other countries on their way, they will almost certainly get bagged.
Voice is monitored. This much is almost certain these days. Imagine, if you will, a regular-expression against patterns in an audio stream. That's probably what's going on. If you say "Iraq" over an international phone-call, your call will be flagged, and transcribed as well as the best speach-to-text that money can buy will get you.
Never trust that using encryption makes you safe. It doesn't. In fact it identifies you as a target for code-cracking.
The only semi-safe way to go is to meet in person with someone that you want to communicate with, transfer a phrasebook or list of one-time pads, and then use those later on. I've been thinking of writing something that uses postings to things like USENET, Slashdot, and so on to subtly encode things into. This would look just like ordinary traffic, but you could manipulate, say, the timestamp in the message header to get a small amount of data through. This would be very low bandwidth, but when combined with automation would allow short messages to be turned into several dozen "Hey, check out this article" type messages....
The shells are all a poor way to code anything more than very simple commands. For example, I've done some timing of roughly the scenario you describe:
It takes about 4 minutes and 22 seconds to go over every C file in the GNOME source tree, searching for the inline keyword using find/grep (grep only once via xargs). This is about as efficient as I can get with the shell.
It took 3 minutes and 39 seconds in perl (depth-first, one stat per file, simple print "$file:$.:$_" if/inline/ type statement.
I used sort/diff on the resulting files and they were the same. Perl's just faster by about 1.2 times. Of course, that gap widens considerably as soon as you start doing anything complex, and the gap between perl and C/C++ starts narrowing.
The reason is that the shell can only communicate by moving around and parsing strings. Perl is managing real datastructures like file and directory handles, and never has to copy a filename from one process to another (e.g. the way find has to send filenames to grep).
Now, try making it "find all strings that look like an author's name inside of a C source file, inside of a comment (either C or C++ style)". I can put money on perl vs. shell there, and C is going to take a lot of work to beat Perl by just a little (you'd basically have to write a simple, single-purpose scanner that matched C comments. Perl can do this with its built-in regular expressions (e.g. search_for_author(\$&) while m|/\*.*?\*/|g).
If you spend a majority of your time on a project waiting for it to compile rather than writing code, you either have a really slow processor or a really small project.
I just compiled GNOME from scratch, including glib, gtk+, ORBIt, all of the other support libraries, GIMP and a number of other applications. It took all morning and some of the afternoon. I have a six-month-old machine with 64MB of RAM. Would you care to re-think that statement?
merely a cut down Outer Limits which uses Star Trek backgrounds.
Interesting. When someone refers to an anthology series as being "like" something, they usually go for The Twilight Zone.
Yes, I'm proposing an anthology series based loosely on the Trek universe.
setting is irrelevant to the success or failure of a TV show (or movie).
Setting is not irrelevant for many reasons. The first, and least important, is that setting draws people into the show. If the writing overshadows this, then you are essentially right as far as the viewer is concerned. However, setting is also critical in a TV show because it determines how likely you will be to succede in making the show. The Twilight Zone worked because they just used whatever sets were lying around. They dressed them up a little and shot. Nowadays, a series really needs to have its own space, and that means that you have to have some kind of stable sets. I was suggesting a way to go half-way and have semi-stable sets. This is how you get a "differnet" setting each week, and avoid being tied to "The Ship" or "The Station", etc. This makes the series bible simpler and less restrictive, which is a hook for authors, which leads into...
Your other comments can be summed up as "it's the writing, stupid!"
I cannot agree more. This is why I was pointing out a way to get the Paramount writers out of the loop and open it up wide to any authors who want to do mass-market SF. Does it have to be a Starfleet/Federation story? Not really. The likelyhood that someone's going to be able to make the episode will probably be increased if you use the core characters, but there's no reason that a story that only has 1 or 2 characters needs to use the regulars.... This could be a chance for Paramount/Viacom to really pump some life back into the series. Of course, based on past performance, I doubt it would happen.
The other way I could see it going is to have a fairly loose bible, and a dictator-like staff-editor/writer who does the Straczynski thing (e.g. weaving a "background" plot which stretches over several seasons by slightly modifying the contributed scripts). This would have less likelyhood of working for Paramount, because they have no one like JMS to do this with an iron hand, and yet crank out top-notch backplot. If they could get someone, then it could work. Breman (is that the right spelling) would just screw this up. If you got someone who was really brilliant, though, you could even have the episodes out of chronological order, and slowly put the pieces together in such a way that only someone who sees the entire series even knows it's happening.
I think, though, that the only reason this worked so well in the first and second seasons of B5 was that viewers who didn't "get" the uber-plot were still grabbed by the individual stories. If you want to do that for Trek, you need to rip it out of the current "writing team" hands. Writing by commitee does not work, and ST is now a great example of the pitfalls.
TNG was uneven, and those who refuse to admit that should be made to watch the "planet of somewhat clad bodybuilders" episode from the first season until they cry "uncle". TNG had some great episodes. The first half of "The Best of Both Worlds" was some of the most exciting TV SF I've seen. The second half was a sorry excuse for the beginning of a season (make them go to sleep? This is a fearsome enemy?) But that's OK. Unevenness at least proved they were trying, aspiring to SOMETHING. Voyager is not... uneven. It's quite bland all the way through. Sigh. Sometimes I turn on the TV and find Voyager and hope too much. I watch what might be a good episode, and my roommate says "it's all going to be a dream, you know. They can't actually be allowed to change things in this series." He's always right, of course.
TOS was campy and sometimes truely bad (e.g. "Spock's Brain"), but there were times that that show succeded at something that TV SF had never done before (in the US anyhow): it made people THINK. There was a hue and cry over the inter-racial kiss. They even tried to explore (a little poorly, I admit) the phenomenon of hippies, which was a very touchy subject at the time (unless you were just slapping them down outright). They touched on the subjects of hatred, slavery and compassion. And, in the end it was just a much better show than TNG could ever have been because the studio was convinced it would fail. With TNG they would never have been allowed to show an actual gay couple kissing (in fact even a man-turned-woman kissing the Dr. goodbye was changed to a kiss on the wrist), but given the political climate of the times, that would be the exact political equivalent of the Kirk/Uhura kiss. I am not saying that this political topic had to be explored, but the fact that it could not be was quite telling.
Why couldn't TNG forge into unknown territory politically? Because the studio thought the show could be a success. It did not have TOS' luxury of assumed failure.
Same with Babylon 5 / Crusade. Straczynski was left alone to do Babylon 5, and challenged the viewers to re-invent TV SF in their minds. When he went to do Crusade, he found that the industry had never changed. He was just under their radar screen the first time around.
It's not really Rick Bremman or Gene Roddenbury or JMS or Majel Barrette-Roddenbury (pardon misspellings) who have caused any of this. They try to varying extents to make their shows great. But, Hollywood is a scary town and perceived profit is like chum in the water for sharks. The Ferrengi start to look like the good guys....
Star Trek is essentially dead. Voyager is stringing along just well enough to keep some minimal fan base from storming the studio with pitchforks and torches, but.....
What if Paramount created a very loosely themed series with nothing but guest writers and a team of staff-editors? You could have a group of 5-10 "regulars". Kind of an anthology show, maybe even with 1/2-hour shorts grouped 2 per episode. Don't give it a ship. We've already established that the ships look pretty similar inside, so there's no reason to tie the show down. Just build a bunch of sets that can be quickly dressed to be different ships.
You could then tackle everything from the oft mentioned Star Fleet Academy (still an idea I would need proved to me) to exploration to simple character pieces to huge Starfleet-moving plots. Hell, at that point you might even get a few good SF writers to write an episode or two. Not so much share-cropper novels as mildly-constrained free-form SF.
The best way to go would probably be to avoid setting any ground-rules. Just write up the core cast for the series bible and set the writers loose. If they write something that you would never have allowed, but it just works, go with it.
This is really about all I can imagine that would save Star Trek. You'll notice that I've given no actual story ideas here. If someone's on the inside, please feel free to show this around, as there's nothing here that any real claim could be laid to.
Star Trek was a wonderful phenomenon to watch. The original series could really stand on its own, even given the extreme level of schmaltz. The fan-base was an unheard of phenomenon for a TV show.
But it's been a long, slow downhill slide. The movies were always uneven, but the recent offerings even make The Great Bird's self-indugent outing in ST:TMP look like art. Voyager episode quality can be measured in terms of how much of the plot can be said to have actually happened at the end of the episode, and no one on the set of any recent ST production (TV or movie) actually takes it seriously beyond their own task. It's just a business, now (this last from various sources at media cons and through the Net).
Look at what's been going on meanwhile: Babylon 5, Earth: Final Conflict, Farscape, X-Files. Even Reboot has plots that Voyager could learn a few things from. I'm not saying that other shows are perfect. Babylon 5's "Grey 17 Is Missing" provoked creator J. Michael Straczynski to lament not being able to apologize to each and every fan individually. But these shows try to explore what can be done with TV science fiction. Star Trek is interested in finding a way to increase the franchise market-share, and not much else.
I'm not actually trying to bash Star Trek, here. What I'm trying to say is that the good people involved in that show would be far better served trying to create something new and innovative. Those who desperately cling to the franchise because it's the only way to assure a market should be quietly taken out and shot. It's just more humane that way. After all, they shoot producers, don't they?
Perhaps in 20 years, the fan base will rise again to provoke a new generation of studio execs to re-create the mythos again. Until then, even the hint that ST's days may be numbered is welcome news in these quarters...:-|
It is hard to see how open source would be less acceptable in situations with security concerns. The GPL doesn't mandate releasing derived code to anyone unless you are distributing it outside your organization.
First, in your later comments you confuse this use of "security" (as in "national security") with "security" as in how secure the code is.
Second, absolutely, the Government should never use GPLed (or LGPLed) code where matters of national security will require that the code not be revealed (again, unless the FSF is willing to modify the GPL for this case). Why? Because, given your example, if the DoD wants to have that weapons system's code modified, they may choose to give it to a contracting firm to do (this is very standard practice). They cannot do this while meeting the terms of the GPL and the terms of the presumed security rating of this software. They would have to require that the contractor not distribute the source to third parties (say, newspapers), but the GPL requires that this be allowed, and says that if you cannot meet this condition, you may not distribute the source!
Clearly, restricting source code to the point where it cannot be distributed at all was not the point, here. Of course BSD and MIT/X style licenses allow for this, as they have no clause which requires free redistribution, they only allow for it. Thus, your example nuke could be built with, say, the BSD C library printf code, and all that would be required is that the BSD copyright is maintained on the code (this, even, may be a problem, as govt. code is required to be public domain, but IANAL).
Slashdot is nice because it's a very specific forum for us to talk about computers and OSes (read: Linux). If we bog it down with less salient issues, it will become garbage.
See http://slashdot.org/search.pl?topic=movi es for a lot more examples of non-technical news. Oh yeah, this is just a tech news site. This is a site dedicated to a particular set of cross-cultural interests held by the new technical sub-culture (oooh! I just made that up;-)
If what you want is to have a site that caters to your specific tech bent, then exclude the movie info (that's under preferences). Slashdot is Slashdot. If you see it here, that's Slashdot content. What you think is Slashdot "style" is rather beside the point.
This is an unfortunate trend. Sandman is Gaiman's most acclaimed literary credit, but we sort of trip over the idea of calling it "Literature". Sure, there are words, but there are... GASP... PICTURES! It's a COMIC BOOK!
Books like Sandman will continue to challenge the assumption that works with storyboard-style pictures are not Literature. This is a Good Thing(tm).
Or, to paraphrase Gaiman from a talk he gave at MIT: If it's a picture we call it art, and it's respected. If it's a book with words, it's literature and it's respected. If it's a book with words and pictures, it's crap for kids.
I could see an Open Source initiative for the US government, but we have some oddities that would require special casing. Here's my thought:
A general policy (executive order, sense of congress?) which states that the use of Open Source software is acceptable in any situation where the resulting software is not to be constrained by national security. BSD and MIT/X style software would still be fine in those situations, but GPL would be right out, unless the FSF was willing to make a general exception in writing for government projects thus constrained. The key, here, is that right now, most government contractors "feel" (and sometimes are told) that the government will not accept projects based on Open Source.
Require that for every contract bid over X number of dollars, an Executive Office of Open Source be asked to review the Open Source world for existing solutions. This is sort of like requiring that there be an Open Source bid, but without there having to be a company to bid it. This office could be VERY small, and need not be a beaurecratic albetros around the bidding process' neck. OTOH, has that ever stopped the executive branch?
Perhaps an ARPA-style grant could go out to research all non-classified government source code and determine what pieces would best be cleaned up and contributed back into the general Open Source pool (note, this would be as public domain software, as that is what ALL US government-owned software is by default, unless classified).
I've been consistantly impressed with MSNBC's objectivity when it comes to Microsoft. They were among the first mainstream News outlets to tout the joys of Linux, they've had objective articles about the Anti-Trust case, and now this.
This is very important in journalism, and I'm quite impressed that MS has not subverted the objectivity of the site. OTOH, who knows what's gone on behind the scenes to maintain that....
I've gone and checked out their site, and they don't mention patenting the genome sequences, but if you want to contact them, their contact page is at:
And, on a technical note: ".shtm"?! What the hell is that?! People started using.htm back when DOS/Windows could not handle more than 3-character extensions reliably, but this is just silliness! Sigh.
Of course, even if you get convicted, you still have rights. Something that the United States seems to want to forget is that prisoners have rights. We overlook some on a routine basis (unreasonable search and seizure for example), but others are respected on an ad-hoc basis (e.g. freedom of speech and the right to vote).
Tying this into another article (I really wish you could cross-post on/.), this is also one of the largest problems with schools. Students' rights are routinely overlooked in the interest of "crowd control"....
Gee, it's been 12 years, and I'm still bitter about high school. Go figure.
You're exactly right, and I wish people wouldn't knee-jerk like this. But, I also feel even more strongly than you that this is a Bad Thing(tm).
Why? Because it will occasionally be correct, and when it is, it will mess some kid up good. I have a friend who is fairly unstable. He knows this, and has known it for many years. Still, he does what he needs to to keep himself under control and never takes out his agressive tendancies on others. Would he show up as violent on this test? Probably yes. Would it have pushed him over the edge if people started tagging him as a potential threat in High School? You betcha! Instead, he is now a respected individual who I am happy to be able to say I know. Go figure.
What we need to do, here, is not try to pick out the trouble-makers and "protect ourselves" from them. We need to a) figure out if there really is a problem (someone goes nuts in a school several times a year, most times it doesn't even make the news) b) assuming that there is a problem, we need to find and adress it. Not the kids. They're no more the source of the problem than the USPS was back in the early 90's. Nor are the schools the problem. We, as a society are encouraging people to behave in the ways that they do. If we don't like the way that people are behaving, we need to change the incentives.
I don't know if there is a wide-spread problem, or just a few good excuses for news ratings. But, if there is, we will not solve it by spending more money on police, beauracracy, personality tests and other controls. We will only solve for it if we pull together and decide what we want to do with ourselves.
Hmmm... Yes and no. I see your point, but keep in mind that what we're seeing now is the business equivalent of an auto-immune reaction in the human body. The corporate world is trying to figure out if we're a good thing or a bad thing, and until they do so, they cannot afford to let their guard down.
This may sound melodramatic, but keep in mind that many businesses fail exactly because they are not paranoid enough. Once open source software has been accepted as harmless, you see a lot less of this sort of thing, but that will take some really amazing success stories.
This is the two-edged sword of business involvement. Several years ago, we all wanted to see how far all this could go. Now, we see how far it will go, and some of us, I suspect, are getting cold feet. Yes, the future of Linux is going to be more in marketing than in what the code actually does. Yes, Apache's biggest hurdle in the next 5 years will be building "vertical markets" (whatever the heck those are), and you'll hear no end of it all if you can still stand to read the trade rags.
Nature of the beast. People like RMS thought they could change the world. What they all failed to consider is how much the world would absorb their work and go right on acting like nothing had happend. That's what humans do. Otherwise, we'd still be sitting in a corner kibbying over the implications of the wheel.
Open-source hackers are accustomed to hearing the likes of evangelist Eric Raymond declaiming authoritatively upon the whys and wherefores of open source, but there's a paucity of hard data to work from.
While, I think the idea of a survey of Open Source data-points (e.g. employment, work habits, motivation, etc) would be great, I don't get this. If you actually read Eric Raymond's writing, most of it is based on real-world experience with the projects that he has worked on and those that he has had direct contact with. Netscape, for example, provides a wealth of data in their release of the Mozilla source. How much work can be harnessed from the community? Look at the changelogs....
Same goes for things like GNOME. We know exactly how fast a given sized group of hackers can put together a large Open Source software effort, because we've seen it.
What we don't have numbers for, and I don't think that a survey can establish this any better than ESR can, is this: what happens if your company wants to create an Open Source effort around your product? Answering that requires the answers to these questions:
Do you have a product that will entice developers to work on it?
Are you willing to spend time and energy in the beginning to assure the developer community that you're for real, and actually care about what happens to the code?
Is there a competing effort (or reason to create one) that more developers will be interested in?
For example, if Sun opened up the source to their C++ compiler, I'm sure the GCC/EGCS folks would enjoy getting a look at how Sun handles some of the SPARC optimization, but I can't see a lot of developers clustering in to help Sun develop their product. It's just not technically interesting enough. It's not portable, it has less than 5 language front-ends and it just doesn't have the clout in the community that GCC has.
On the other hand, if Adobe were to open up the source to Illustrator, and really convince the community that they wanted to jump in with both feet, the way Netscape did, they would have a huge developer interest.
In the end, Open Source is not so much a "phenomenon". This is the way software worked pre-80's. Hiding source seemed to make about as much sense to most people as trying to hide how a lightbulb works. Now, we're coming full circle, and people are cluing in to that. The "Open Source Phenomenon" is just a bunch of people trying to figure out how to make the intervening 20 years of industry make sense....
Slashdot Mirror
I suspect that this will play out as a trend in Open Source and quasi-open source development projects lead by the likes of Sun and Netscape(Aol). Large companies will decide to release the source to something, and fail to see the value which Open Source projects employ in the release-early-release-often model. This is unfortunate, since releasing a not-quite-there-yet, but promising version as "alpha" would get other people finishing features and fixing bugs that they needed to scratch. This is what I think Mozilla has done wrong. They release these non-functional Mx releases as code benchmarks, but fail to work on getting a nice looking, feature light version out the door so that large masses of people will USE it, and eventually have some of them join the project. They've already forgotten how fast the Australians were able to impliment real crypto back when everyone thought that Mozilla would be a usable browser in 2 months.
NOTE: I'm not involved with either effort, but if my statements are incorrect because of my lack of inside knowledge, then they are likely to -- even more accurately -- reflect what the average Linux/Open Source user is thinking....
The email address of the lawyers in question has been post ed to the list as well.
So, if I want to get a patent, all I have to do is take a product that currently exists, and remove stuff? Ok, here's one:
A microwave oven.... with no buttons!
Wow, this is fun! Ok, one more:
Windows... without the bugs!
Oh damn, I'm gonna get rich!
why set up a Linux is not X/BSD/PD/Artistic/GNU/Linux page when nobody claimed that Linux is all those things? all he said was that Linux is a kernel
Go read the document. He's not saying "since we can call it anything why not add GNU in." That would be damn near reasonable. He says "you're running a modified version of the GNU system."
This is doublespeak worthy of Ed Muth, not RMS. There never was a GNU system, and this statement discounts the many large chunks of your average Linux distribution that are not derived from GNU's attempts to create a system (e.g. X, wu-ftpd, tcsh, ftp, rlogin, inn, nvi, vim, etc, etc, etc). If we discount "GNU homages" like GNOME, which are not actually part of the GNU effort, but use the name, very little of your average distribution is GNU. There's the compiler, basic C library, EMACS and a slew of utilities. That's really about it. Wonderful contributions, all, and I respect them. Before RMS started on this rant, I was happily pointing out the importance of GNU to Linux, but now I downplay it because this territorial attitude really turns me off.
I've never been able to think of a reasonable justification for calling Linux GNU/Linux. The only reasonable way to modify what we call it to properly attribute the external efforts that have helped Linux get started would be to call it some silly, long string of names that cited the BSD, Public Domain, MIT/X, Perl/Artistic, GNU and many other contributions. Let's just not go there, Ok. The thing is Linux, and it runs a lot of damn fine programs. If Stallman wants to name Linux, he should create a distribution (which, arguably, he has done, since the Debian folks were happy to take up his re-naming).
I can just see the "next step" where we try to rename "Solaris" to "GNU/Solaris" because so many people install GNU utilities on every Solaris box they touch.... Just scary.
I actually had to do a double-take through the document to see if he'd prefaced this at all. Nope, this is just a random, unexplained cheap-shot at ESR, who is not mentioned anywhere else in the document.
This is getting very old. I'm tempted to set up a "Linux is not X/BSD/PD/Artistic/GNU/Linux" page and just post the URL in every forum that refers to yet another Stallman article. I'll not go into it more, here, though. I assume people have better things to do.
In the end, it's classic Stallman. The man can write damn fine code, but I wish he'd stop trying to talk to people. Others such as ESR, Larry Wall and Linus Torvalds really do do a much better job of getting the points across. Overall, though the early points in the article were fair. Stallman does get a lot of flack over the GPL, which is quite a reasonable document when you consider things like the MS EULA (which I will never agree to, even though I use their software as an end-user at work) define the rest of the playing field.
I know this is percieved as "political correctness", and maybe it is, but are there other angles that can benefit us all?
Imagine, if you will, a World Wide Web where everything was delivered in a, say, XML format where enough "structured content" information was delivered to a browser that the browser could reasonably read the content to a blind user (who could ask for things like "read section headings and/or headlines from this page".
What else could you possibly use this for...? How about a dynamic browser that you can configure to do things like "hide all "Article" content" so that sites like Slashdot would show up as just headlines. How about asking your browser to display no images, but still having enough structure information to display the page correctly.... The blind will end up pushing the Net back into what it was intended to be: an information delivery mechanism, not a hypertext ad magazine.
Clearly content structure is not universal, and not every site is a collection of articles like Slashdot, but I think the example is powerful enough to merit finding a way to make the general case work....
Hidding the existance of a message is called steganography. Its more common to high the message in a single image or MP3 as they have more bits to obscure the payload.
;-)
Absolutely. And, they have more bits to analyze. I'll bet I could write a program that would identify images very quickly that have been so modified. How? Just a simple statistical analysis against a baseline sample of such images. Gee, the low bits of this image are very unlikely for a JPEG. I'll just take a closer look....
Yeah, that kind of analysis is probably part of the NSA's automatic filtering already, and you can bet that they've got a full USENET feed watching the binary newsgroups for just this sort of thing. Funny that we would pay so much money in taxes for our government to collect and analyze porn
The only semi-safe way to go is to meet in person with someone that you want to communicate with, transfer a phrasebook or list of one-time pads, and then use those later on. I've been thinking of writing something that uses postings to things like USENET, Slashdot, and so on to subtly encode things into. This would look just like ordinary traffic, but you could manipulate, say, the timestamp in the message header to get a small amount of data through. This would be very low bandwidth, but when combined with automation would allow short messages to be turned into several dozen "Hey, check out this article" type messages....
The shells are all a poor way to code anything more than very simple commands. For example, I've done some timing of roughly the scenario you describe:
/inline/ type statement.
It takes about 4 minutes and 22 seconds to go over every C file in the GNOME source tree, searching for the inline keyword using find/grep (grep only once via xargs). This is about as efficient as I can get with the shell.
It took 3 minutes and 39 seconds in perl (depth-first, one stat per file, simple print "$file:$.:$_" if
I used sort/diff on the resulting files and they were the same. Perl's just faster by about 1.2 times. Of course, that gap widens considerably as soon as you start doing anything complex, and the gap between perl and C/C++ starts narrowing.
The reason is that the shell can only communicate by moving around and parsing strings. Perl is managing real datastructures like file and directory handles, and never has to copy a filename from one process to another (e.g. the way find has to send filenames to grep).
Now, try making it "find all strings that look like an author's name inside of a C source file, inside of a comment (either C or C++ style)". I can put money on perl vs. shell there, and C is going to take a lot of work to beat Perl by just a little (you'd basically have to write a simple, single-purpose scanner that matched C comments. Perl can do this with its built-in regular expressions (e.g. search_for_author(\$&) while m|/\*.*?\*/|g).
If you spend a majority of your time on a project waiting for it to compile rather than writing code, you either have a really slow processor or a really small project.
I just compiled GNOME from scratch, including glib, gtk+, ORBIt, all of the other support libraries, GIMP and a number of other applications. It took all morning and some of the afternoon. I have a six-month-old machine with 64MB of RAM. Would you care to re-think that statement?
merely a cut down Outer Limits which uses Star Trek backgrounds.
Interesting. When someone refers to an anthology series as being "like" something, they usually go for The Twilight Zone.
Yes, I'm proposing an anthology series based loosely on the Trek universe.
setting is irrelevant to the success or failure of a TV show (or movie).
Setting is not irrelevant for many reasons. The first, and least important, is that setting draws people into the show. If the writing overshadows this, then you are essentially right as far as the viewer is concerned. However, setting is also critical in a TV show because it determines how likely you will be to succede in making the show. The Twilight Zone worked because they just used whatever sets were lying around. They dressed them up a little and shot. Nowadays, a series really needs to have its own space, and that means that you have to have some kind of stable sets. I was suggesting a way to go half-way and have semi-stable sets. This is how you get a "differnet" setting each week, and avoid being tied to "The Ship" or "The Station", etc. This makes the series bible simpler and less restrictive, which is a hook for authors, which leads into...
Your other comments can be summed up as "it's the writing, stupid!"
I cannot agree more. This is why I was pointing out a way to get the Paramount writers out of the loop and open it up wide to any authors who want to do mass-market SF. Does it have to be a Starfleet/Federation story? Not really. The likelyhood that someone's going to be able to make the episode will probably be increased if you use the core characters, but there's no reason that a story that only has 1 or 2 characters needs to use the regulars.... This could be a chance for Paramount/Viacom to really pump some life back into the series. Of course, based on past performance, I doubt it would happen.
The other way I could see it going is to have a fairly loose bible, and a dictator-like staff-editor/writer who does the Straczynski thing (e.g. weaving a "background" plot which stretches over several seasons by slightly modifying the contributed scripts). This would have less likelyhood of working for Paramount, because they have no one like JMS to do this with an iron hand, and yet crank out top-notch backplot. If they could get someone, then it could work. Breman (is that the right spelling) would just screw this up. If you got someone who was really brilliant, though, you could even have the episodes out of chronological order, and slowly put the pieces together in such a way that only someone who sees the entire series even knows it's happening.
I think, though, that the only reason this worked so well in the first and second seasons of B5 was that viewers who didn't "get" the uber-plot were still grabbed by the individual stories. If you want to do that for Trek, you need to rip it out of the current "writing team" hands. Writing by commitee does not work, and ST is now a great example of the pitfalls.
TNG was uneven, and those who refuse to admit that should be made to watch the "planet of somewhat clad bodybuilders" episode from the first season until they cry "uncle". TNG had some great episodes. The first half of "The Best of Both Worlds" was some of the most exciting TV SF I've seen. The second half was a sorry excuse for the beginning of a season (make them go to sleep? This is a fearsome enemy?) But that's OK. Unevenness at least proved they were trying, aspiring to SOMETHING. Voyager is not... uneven. It's quite bland all the way through. Sigh. Sometimes I turn on the TV and find Voyager and hope too much. I watch what might be a good episode, and my roommate says "it's all going to be a dream, you know. They can't actually be allowed to change things in this series." He's always right, of course.
TOS was campy and sometimes truely bad (e.g. "Spock's Brain"), but there were times that that show succeded at something that TV SF had never done before (in the US anyhow): it made people THINK. There was a hue and cry over the inter-racial kiss. They even tried to explore (a little poorly, I admit) the phenomenon of hippies, which was a very touchy subject at the time (unless you were just slapping them down outright). They touched on the subjects of hatred, slavery and compassion. And, in the end it was just a much better show than TNG could ever have been because the studio was convinced it would fail. With TNG they would never have been allowed to show an actual gay couple kissing (in fact even a man-turned-woman kissing the Dr. goodbye was changed to a kiss on the wrist), but given the political climate of the times, that would be the exact political equivalent of the Kirk/Uhura kiss. I am not saying that this political topic had to be explored, but the fact that it could not be was quite telling.
Why couldn't TNG forge into unknown territory politically? Because the studio thought the show could be a success. It did not have TOS' luxury of assumed failure.
Same with Babylon 5 / Crusade. Straczynski was left alone to do Babylon 5, and challenged the viewers to re-invent TV SF in their minds. When he went to do Crusade, he found that the industry had never changed. He was just under their radar screen the first time around.
It's not really Rick Bremman or Gene Roddenbury or JMS or Majel Barrette-Roddenbury (pardon misspellings) who have caused any of this. They try to varying extents to make their shows great. But, Hollywood is a scary town and perceived profit is like chum in the water for sharks. The Ferrengi start to look like the good guys....
Star Trek is essentially dead. Voyager is stringing along just well
enough to keep some minimal fan base from storming the studio with
pitchforks and torches, but.....
What if Paramount created a very loosely themed series with nothing
but guest writers and a team of staff-editors? You could have a group
of 5-10 "regulars". Kind of an anthology show, maybe even with
1/2-hour shorts grouped 2 per episode. Don't give it a ship. We've
already established that the ships look pretty similar inside, so
there's no reason to tie the show down. Just build a bunch of sets
that can be quickly dressed to be different ships.
You could then tackle everything from the oft mentioned Star Fleet
Academy (still an idea I would need proved to me) to exploration to
simple character pieces to huge Starfleet-moving plots. Hell, at that
point you might even get a few good SF writers to write an episode or
two. Not so much share-cropper novels as mildly-constrained free-form
SF.
The best way to go would probably be to avoid setting any
ground-rules. Just write up the core cast for the series bible and set
the writers loose. If they write something that you would never have
allowed, but it just works, go with it.
This is really about all I can imagine that would save Star
Trek. You'll notice that I've given no actual story ideas here. If
someone's on the inside, please feel free to show this around, as
there's nothing here that any real claim could be laid to.
Star Trek was a wonderful phenomenon to watch. The original series could really stand on its own, even given the extreme level of schmaltz. The fan-base was an unheard of phenomenon for a TV show.
:-|
But it's been a long, slow downhill slide. The movies were always uneven, but the recent offerings even make The Great Bird's self-indugent outing in ST:TMP look like art. Voyager episode quality can be measured in terms of how much of the plot can be said to have actually happened at the end of the episode, and no one on the set of any recent ST production (TV or movie) actually takes it seriously beyond their own task. It's just a business, now (this last from various sources at media cons and through the Net).
Look at what's been going on meanwhile: Babylon 5, Earth: Final Conflict, Farscape, X-Files. Even Reboot has plots that Voyager could learn a few things from. I'm not saying that other shows are perfect. Babylon 5's "Grey 17 Is Missing" provoked creator J. Michael Straczynski to lament not being able to apologize to each and every fan individually. But these shows try to explore what can be done with TV science fiction. Star Trek is interested in finding a way to increase the franchise market-share, and not much else.
I'm not actually trying to bash Star Trek, here. What I'm trying to say is that the good people involved in that show would be far better served trying to create something new and innovative. Those who desperately cling to the franchise because it's the only way to assure a market should be quietly taken out and shot. It's just more humane that way. After all, they shoot producers, don't they?
Perhaps in 20 years, the fan base will rise again to provoke a new generation of studio execs to re-create the mythos again. Until then, even the hint that ST's days may be numbered is welcome news in these quarters...
It is hard to see how open source would be less acceptable in situations with security concerns. The GPL doesn't mandate releasing derived code to anyone unless you are distributing it outside your organization.
First, in your later comments you confuse this use of "security" (as in "national security") with "security" as in how secure the code is.
Second, absolutely, the Government should never use GPLed (or LGPLed) code where matters of national security will require that the code not be revealed (again, unless the FSF is willing to modify the GPL for this case). Why? Because, given your example, if the DoD wants to have that weapons system's code modified, they may choose to give it to a contracting firm to do (this is very standard practice). They cannot do this while meeting the terms of the GPL and the terms of the presumed security rating of this software. They would have to require that the contractor not distribute the source to third parties (say, newspapers), but the GPL requires that this be allowed, and says that if you cannot meet this condition, you may not distribute the source!
Clearly, restricting source code to the point where it cannot be distributed at all was not the point, here. Of course BSD and MIT/X style licenses allow for this, as they have no clause which requires free redistribution, they only allow for it. Thus, your example nuke could be built with, say, the BSD C library printf code, and all that would be required is that the BSD copyright is maintained on the code (this, even, may be a problem, as govt. code is required to be public domain, but IANAL).
Let's just look at the trend here:
See http://slashdot.org/search.pl?topic=movi es for a lot more examples of non-technical news. Oh yeah, this is just a tech news site. This is a site dedicated to a particular set of cross-cultural interests held by the new technical sub-culture (oooh! I just made that up
If what you want is to have a site that caters to your specific tech bent, then exclude the movie info (that's under preferences). Slashdot is Slashdot. If you see it here, that's Slashdot content. What you think is Slashdot "style" is rather beside the point.
Neil Gaiman of Sandman and literary fame...
This is an unfortunate trend. Sandman is Gaiman's most acclaimed literary credit, but we sort of trip over the idea of calling it "Literature". Sure, there are words, but there are... GASP... PICTURES! It's a COMIC BOOK!
Books like Sandman will continue to challenge the assumption that works with storyboard-style pictures are not Literature. This is a Good Thing(tm).
Or, to paraphrase Gaiman from a talk he gave at MIT: If it's a picture we call it art, and it's respected. If it's a book with words, it's literature and it's respected. If it's a book with words and pictures, it's crap for kids.
Thoughts?
Looks like three are finished already, with more coming
;-)
Haven't I heard this about Knuth's work before?
I've been consistantly impressed with MSNBC's objectivity when it comes to Microsoft. They were among the first mainstream News outlets to tout the joys of Linux, they've had objective articles about the Anti-Trust case, and now this.
This is very important in journalism, and I'm quite impressed that MS has not subverted the objectivity of the site. OTOH, who knows what's gone on behind the scenes to maintain that....
I've gone and checked out their site, and they don't mention patenting the genome sequences, but if you want to contact them, their contact page is at:
.htm back when DOS/Windows could not handle more than 3-character extensions reliably, but this is just silliness! Sigh.
http://www.celera.com/Compan yInformation/Contacts.shtm
And, on a technical note: ".shtm"?! What the hell is that?! People started using
Of course, even if you get convicted, you still have rights. Something that the United States seems to want to forget is that prisoners have rights. We overlook some on a routine basis (unreasonable search and seizure for example), but others are respected on an ad-hoc basis (e.g. freedom of speech and the right to vote).
/.), this is also one of the largest problems with schools. Students' rights are routinely overlooked in the interest of "crowd control"....
Tying this into another article (I really wish you could cross-post on
Gee, it's been 12 years, and I'm still bitter about high school. Go figure.
You're exactly right, and I wish people wouldn't knee-jerk like this. But, I also feel even more strongly than you that this is a Bad Thing(tm).
Why? Because it will occasionally be correct, and when it is, it will mess some kid up good. I have a friend who is fairly unstable. He knows this, and has known it for many years. Still, he does what he needs to to keep himself under control and never takes out his agressive tendancies on others. Would he show up as violent on this test? Probably yes. Would it have pushed him over the edge if people started tagging him as a potential threat in High School? You betcha! Instead, he is now a respected individual who I am happy to be able to say I know. Go figure.
What we need to do, here, is not try to pick out the trouble-makers and "protect ourselves" from them. We need to a) figure out if there really is a problem (someone goes nuts in a school several times a year, most times it doesn't even make the news) b) assuming that there is a problem, we need to find and adress it. Not the kids. They're no more the source of the problem than the USPS was back in the early 90's. Nor are the schools the problem. We, as a society are encouraging people to behave in the ways that they do. If we don't like the way that people are behaving, we need to change the incentives.
I don't know if there is a wide-spread problem, or just a few good excuses for news ratings. But, if there is, we will not solve it by spending more money on police, beauracracy, personality tests and other controls. We will only solve for it if we pull together and decide what we want to do with ourselves.
Hmmm... Yes and no. I see your point, but keep in mind that what we're seeing now is the business equivalent of an auto-immune reaction in the human body. The corporate world is trying to figure out if we're a good thing or a bad thing, and until they do so, they cannot afford to let their guard down.
This may sound melodramatic, but keep in mind that many businesses fail exactly because they are not paranoid enough. Once open source software has been accepted as harmless, you see a lot less of this sort of thing, but that will take some really amazing success stories.
This is the two-edged sword of business involvement. Several years ago, we all wanted to see how far all this could go. Now, we see how far it will go, and some of us, I suspect, are getting cold feet. Yes, the future of Linux is going to be more in marketing than in what the code actually does. Yes, Apache's biggest hurdle in the next 5 years will be building "vertical markets" (whatever the heck those are), and you'll hear no end of it all if you can still stand to read the trade rags.
Nature of the beast. People like RMS thought they could change the world. What they all failed to consider is how much the world would absorb their work and go right on acting like nothing had happend. That's what humans do. Otherwise, we'd still be sitting in a corner kibbying over the implications of the wheel.
While, I think the idea of a survey of Open Source data-points (e.g. employment, work habits, motivation, etc) would be great, I don't get this. If you actually read Eric Raymond's writing, most of it is based on real-world experience with the projects that he has worked on and those that he has had direct contact with. Netscape, for example, provides a wealth of data in their release of the Mozilla source. How much work can be harnessed from the community? Look at the changelogs....
Same goes for things like GNOME. We know exactly how fast a given sized group of hackers can put together a large Open Source software effort, because we've seen it.
What we don't have numbers for, and I don't think that a survey can establish this any better than ESR can, is this: what happens if your company wants to create an Open Source effort around your product? Answering that requires the answers to these questions:
For example, if Sun opened up the source to their C++ compiler, I'm sure the GCC/EGCS folks would enjoy getting a look at how Sun handles some of the SPARC optimization, but I can't see a lot of developers clustering in to help Sun develop their product. It's just not technically interesting enough. It's not portable, it has less than 5 language front-ends and it just doesn't have the clout in the community that GCC has.
On the other hand, if Adobe were to open up the source to Illustrator, and really convince the community that they wanted to jump in with both feet, the way Netscape did, they would have a huge developer interest.
In the end, Open Source is not so much a "phenomenon". This is the way software worked pre-80's. Hiding source seemed to make about as much sense to most people as trying to hide how a lightbulb works. Now, we're coming full circle, and people are cluing in to that. The "Open Source Phenomenon" is just a bunch of people trying to figure out how to make the intervening 20 years of industry make sense....