Yes, Sun could offer Java under a dual license. However, once IBM takes the GPL version and adds all the performance tricks from their JVM, the GPL version will jump ahead of the proprietary, and no one will want to use the latter.
Even if IBM would agree to let Sun take their contributions dual (which is possible), most of the open source community wouldn't be very happy about it. Part of the point of going open is to gather the support of the larger community.
Another possibility is to BSD it, in which case everyone probably will contribute under BSD and let Sun take the code in, but that would mean MS could use everything, and Sun would never accept that.
If you go to RMS's personal webpage you'll see quite a few opinions that he regards as orthogonal to GNU. Apart from that page, I've never heard about RMS's involvement with this stuff beyond a few jokes. RMS promotes GNU, but not the rest of himself.
ESR really does promote himself, not just the OSI. Everyone's heard about his guns, his sex life, his politics.... He considers it valueable for everyone to know about him. This is different.
In ESR's defence, he claims to have "developed an entire theory of media manipulation, which I then proceeded to apply. The theory centers around the use of what I call ``attractive dissonance'' to fan an itchy curiosity about the evangelist, and then exploiting that itch for all it's worth in promoting the ideas." In other words, he needs to be a primadonna to get the mainstream media to notice him, then he can tell them about what matters.
He has taken quotes given by Scott McNealy to analysts and attacked them as if they were spoken to the Open Source community.
I didn't believe it until the article finished loading for me. Yes, Phipps really did say that. Forgive my doubt.
Is he saying that everyone lies to analysts so that's all right? Is he admitting that Sun lies to the open-source community, and that we shouldn't try to find out the truth? Is he saying that all statements are true, but we should stick with the ones that are meant for us?
Seriously, if Sun has been telling different listeners contradictory things, that's bad. If they think it's all right, that's worse.
Firstly, as you said, there must be 'substancial similarity'. Reading code once does not make everything you write forever afterwards a derivative work.
Secondly, the 'access' rule is easy to get around. If you have one engineer read the code, and document the format/protocol, and then another engineer looks only at the documentation, this shouldn't count as 'access', right?
What's a lot worse than copyrights are trade secrets. Why I'm obligated to defend Microsoft's secrets I have no idea, but that seems to be the way the law is structured. If someone's willing to take the risk of posting documentation to usenet, however, that information will no longer be secret, and that won't be an issue any more.
Patents are irrelevant: having seen code has no effect on the binding power of a patent.
If you combine the RFID with a smartcard system, this can be secured
The store transmits a random number. The smartcard digitally signs it. The stores checks the digital signiture with the card's public key. The private key is never transmitted anywhere.
Not that they'll get it right the first time, but after a few high-profile thefts, they'll bring in programmers who knows what they're doing.
Everything the old dialog does is still supported in the new one. Part of the advantage of working with old code is getting all this for free.
Incidentally, did you know that you can tab-expand a glob in the old (and new) dialog? Type something like *.jpg into the filename box and hit tab. A useful little feature that I only discovered by reading the source (Hmm, that might suggest a problem...).
Check out the work in progress on it. It's quite usable, but we're still adding features.
The fealing on the GTK list seems to be that there's a need for an entire new widget GtkFileChooser, and programs will eventually convert to this new API. IMHO, this is a very bad idea, as the oldstyle will never really go away any more than the win3.1 style has in the windows world. I think we ought to just add the new features and protect future APIs with preprocessor flags. Code for that might look like:
From the mplayer site, haven't checked it personally:
2004.01.03, Saturday:: Kiss Tech comment
posted by Gabucino
Before I get another 10 mails about this: the GPL.ZIP file which they offer for download on their site contains only the Linux kernel and busybox sources, not MPlayer's!
Add or remove words to ignore (including any markup that gets in) from the grep -E -v line. Replace tail with head to get the least used words.
You can get a very dim idea of what an article is about, but that's really about it.
Re:Best examples of heresy I can think of
on
What You Can't Say
·
· Score: 1
The worst part of it is that these statements are false! (or at least misleading).
Yes, the "evidence" implicating Israel is mostly that they knew something was coming. We know they knew because they warned us. Yes, Israel opporates an effective intelligence service. Wish we did.
The war on drugs kills people in a racist way, but the death count is at most in the thousands -- probably the hundreds. That's not genocide. (It's not acceptable either, but that's a different argument.)
I'm not sure what you think feminism has done to America, but the richest, most powerful, and 17th freeist nation on earth doesn't "ruined" to me.
My point is more general than this though. Once people do encounter heresies, the memes stick in the back of their heads unchallenged. Most people are as reluctant to deny a heresy as to speak it (discussions like this one weaken the taboos). A strongly dogmatic society knows nothing and suspects everything.
If it is, we don't have to put up with it. Most of the code is in our hands. If the gui code is ugly, we'll clean it up. We did it with Mozilla. This is much smaller, and will be much quicker.
If they actually go and cram cruft into the binary codec itself, well then we'll drop the whole thing. mplayer supports real format, both audio and video (x86 only, but I suspect so's this). mplayer's very sleak -- no problem there.
Most spam has some way to contact them (after all, they need your credit card). Often it's a URL. For example, the most recent spam I received (advertising a 'magic lubricant' for enhanced sexual performance) encourages me to visit href=http://dottry.biz/vpoil/ a site which could easily be DDOSed into oblivion
(it can hardly handle normal traffic). What might be more appropriate, however, would be to use their comments or their ortder form to overwhelm them with bogus entries. This ought to be scriptable.
This should be attempted with caution. Spammers may be working with crackers, virus-writers, and possibly meat-space criminals. Operate from behind a firewall, in an open-source browser, with active content diabled. Never give your true name or similar information. If there are strange codes in the URL, remove them, they may encode your e-mail address.
So that's my little guide to vigilante anti-spam activity. Not that I'd ever encourage you to do anything illegal:-)
And what would that number mean even if it comes from somewhere meaningful?
Almost nothing.
The number of 'chemicals' is a silly thing to measure in the first place. Chemically speaking, every protein is a different chemical. Every strand of DNA with a different gene on it is two different chemicals (3-5 side and 5-3 side, H-bonds don't count).
Very few foods are chemically simple: water, salt, sugar, vinegar, maybe grain alcohol-- that's probably it. Even these are a strech, as almost no one consumes straight salt, vinegar, or grain alcohol.
What's more important -- and a major flaw in the sited study, or at least its public face -- is that quantities matter. There are very few chemicals for which the smallest trace we can measure is dangerous. Any chemical is dangerous in sufficient dosage (even the infamous dihidrogen monoxide) and, except for self-reproducing molecules like BSE, none is harmful in insufficient dosage. Some molecules (such as PCBs) cause serious danger at concentrations below 10^-6. These are a real cause for concern, and there's a real need to watch for them. But we can detect them at concentrations orders of magnitude below that.
As a general rule, the number of chemicals with a given effect means nothing. The concentration of those chemicals means something. The concentration compared to the threshold for an effect means almost everything. Ignore this study; there are real ones out there.
If you contact your water department, I think they're obligated to at least tell you what precautions they're already taking regarding your water, including the results of the last round of tests. I strongly suspect they'll send you a long, detailed, and boring packet of information. Generally, water departments are quite thorough, and federal regulations are mostly pretty strict.
If your water does not appear to match what the water department says it's sending out, then there are three possibilities: it actually does, just doesn't seem to; the water company messed up; or the pipes in your house are contaminating it.
Check with a neighbor and a chemical test kit. Photocopy the relevant pages from the Handbook of Chemistry and Physics at your local college library. In the end, you may need to check with a plumber (mnow *that's* expensive!).
Not a problem, the FSF is already 501-c3 (it's not all *that* hard). They certainly accept (tax-deductable) donations which they use to hire programmers (and occationally lawyers). I'm sure they'd take one with acceptable strings (e.g. "Here's $100K, spend it on Gnome"). Obviously they won't take money earmarked for non-free software, but I doubt they insist on copyleft or GNU.
I think there are similar organizations dedicated to KDE, Apache, Mozilla, and many other large projects. For that matter, a serious e-mail to a dev list offering money can probably get a 501-c3 set up.
Speed (and acceleration) are settings which can be changed. This can be done while X is running. I've seen a nice GUI app for it, but I can't remember where. Google is probably your friend. Otherwise, it's probably in xset somewhere.
Oddly, IIRC, Slackware and Debian ship with really nice defaults here, but Redhat and Mandrake go way to slow. Weird.
Plot hole: How did Shelob's sting get through Frodo if he was wearing his mithril vest? Its obvious he wasn't stung higher or lower than where the vest was.
But Tolkien said:
'Yes, I can walk,' said Frodo, getting up slowly, 'I am not hurt, Sam. Only I feel very tired, and I've got a pain here.' He put his hand to the back of his neck above his left shoulder
So presumably that's where Schlob got him. Since he was facing her, this means she reached around a bit, but she has long arms:-). I haven't sseen the movie yet, but presumably the back of Frodo's neck isn't shown. Maybe it's covered by his hair?
As cheap as memory is nowadays, there's no excuse to not upgrade.
Now that's not true.
First of all, a whole lot of windows users aren't somfortable doing hardware upgrades. Ram's especially difficult, at least on those motherboards (which used to be common) where you need a lot of brute force and you're squeezed right between the chip-fan and the power supply, so you have no room to work. For that matter, many old motherboards have weird requirenemnts about RAM (mandatory pairs, forbidden pairs, ordered slots, configurations forbidden because the BIOS-maker was in a bad mood...) and some left off the upper wires in the FSB, putting artificial caps on it.
Now combine these difficulties with someone who doesn't know how to take the case off in the first place, or the need to act on hundreds of computers, and sticking with what you have sounds like a good idea after all.
Yes, literally speaking, it is easy to find excellent programmers. All you need to do is list 2 of them (say, Linus and RMS) and stalk them.
What I think the OP meant is that it's hard to recognize good programmers. Just because someone's degree says "MIT" doesn't guarentee they're any good, nor does the lack of any degree mean they aren't. Nor, as a handful of people suspect, is the opposite true. I suppose another programmer could tell a good programmer by reading his/her code, assuming a large body of that code was available for examination. Even if one is, it doesn't describe how long it took to write, and it will take a long time to study.
AFAICT, no one has found a solution to this yet. Education, experience, certifications, reputation.... Nothing seems to reliably seperate good programmers from bad -- and there are an awful lot of bad. Maybe someone'll come up with something soon.
I can't leave this one uncorrected. The OP did not refer to _Second_Foundation_, the third book of the Foundation trilogy, but to the Second Foundation Trilogy, three books accidentally authorized by Asimov's estate. The first one is a parrelel novel to _Forwar_the_Foundation_ called "Foundation's Fear", which ignores the themes of the original trilogy in favor of some very clumsy and superficial cyberpunk. It re-writes numerous basic premises (including replacing hyperspace with wormwholes, WTF?) and turns Hari Seldon into::shudder:: an action hero.
It's written by Gregory Benford, who (according to reputation) should know better. Maybe he felt that he couldn't use any of his own good ideas in Asimov's universe, but no one else can write Asimov, so he was left with his own bad ideas. Or maybe Benford just sucks -- I haven't read anything else of his.
Avoid this book. Do read the original foundation trilogy (and, IMHO, all the other books in the sequence, except maybe _Foundation's_Edge_). Don't mix them up. That's like confusing Gentry Lee with Arthur C. Clarke.::shudder::
Re:been there and done that
on
AOL's $299 PC
·
· Score: 1
On the other hand, is that really a $699 computer?
I doubt it very much. They don't give full specs, so I wouldn't be at all surprised if it has a 100MHz bus, 24X CD-ROM, and generally low-end components for everything not listed. You can get a comparable system from microtel for $330. (I'm figuring that a 1.6GHz Athlon is at least as good as a 1.7GHz Celeron.) Admittedly, that's without moniter or printer, but that still doesn't cover the difference. Of course, one year of dial-up internet counts for something too. It costs about $150.
So, no, this isn't a particularly great deal. It's not a massive rip-off, either, though (assuming you want a year of dial-up, and don't already have a moniter), so it might make sense for people who don't know how to put things together (even on a very simple scale).
Verbatim copying and distribution of this entire article is permitted without royalty in any medium provided this notice is preserved.
When you're copying an entire essay, is it really too much to include a few lines at the end, so that people know who wrote it and what they're allowed to do with it? It's not like you have to copy-type it, we have copy-and-paste working reliably now?:-)
As long as we're plugging Vinge, someone should mention _Fat_Times_at_Fairmont_High_ (I think I just did), a novella he just wrote playing with these themes in a very near-future context. Someone upthread mentioned how these things will eventually build up as chemical polution -- eventually I guess we'll need them biodegradeable (one of the plot-lines in the story). That novella is becoming true at an alarming rate (not long ago, I took a 'local' test) so I highly recommend it to everyone here.
Slashdot Mirror
Yes, Sun could offer Java under a dual license. However, once IBM takes the GPL version and adds all the performance tricks from their JVM, the GPL version will jump ahead of the proprietary, and no one will want to use the latter.
Even if IBM would agree to let Sun take their contributions dual (which is possible), most of the open source community wouldn't be very happy about it. Part of the point of going open is to gather the support of the larger community.
Another possibility is to BSD it, in which case everyone probably will contribute under BSD and let Sun take the code in, but that would mean MS could use everything, and Sun would never accept that.
If you go to RMS's personal webpage you'll see quite a few opinions that he regards as orthogonal to GNU. Apart from that page, I've never heard about RMS's involvement with this stuff beyond a few jokes. RMS promotes GNU, but not the rest of himself.
ESR really does promote himself, not just the OSI. Everyone's heard about his guns, his sex life, his politics.... He considers it valueable for everyone to know about him. This is different.
In ESR's defence, he claims to have "developed an entire theory of media manipulation, which I then proceeded to apply. The theory centers around the use of what I call ``attractive dissonance'' to fan an itchy curiosity about the evangelist, and then exploiting that itch for all it's worth in promoting the ideas." In other words, he needs to be a primadonna to get the mainstream media to notice him, then he can tell them about what matters.
I didn't believe it until the article finished loading for me. Yes, Phipps really did say that. Forgive my doubt.
Is he saying that everyone lies to analysts so that's all right? Is he admitting that Sun lies to the open-source community, and that we shouldn't try to find out the truth? Is he saying that all statements are true, but we should stick with the ones that are meant for us?
Seriously, if Sun has been telling different listeners contradictory things, that's bad. If they think it's all right, that's worse.
Firstly, as you said, there must be 'substancial similarity'. Reading code once does not make everything you write forever afterwards a derivative work.
Secondly, the 'access' rule is easy to get around. If you have one engineer read the code, and document the format/protocol, and then another engineer looks only at the documentation, this shouldn't count as 'access', right?
What's a lot worse than copyrights are trade secrets. Why I'm obligated to defend Microsoft's secrets I have no idea, but that seems to be the way the law is structured. If someone's willing to take the risk of posting documentation to usenet, however, that information will no longer be secret, and that won't be an issue any more.
Patents are irrelevant: having seen code has no effect on the binding power of a patent.
The store transmits a random number. The smartcard digitally signs it. The stores checks the digital signiture with the card's public key. The private key is never transmitted anywhere.
Not that they'll get it right the first time, but after a few high-profile thefts, they'll bring in programmers who knows what they're doing.
Everything the old dialog does is still supported in the new one. Part of the advantage of working with old code is getting all this for free.
Incidentally, did you know that you can tab-expand a glob in the old (and new) dialog? Type something like *.jpg into the filename box and hit tab. A useful little feature that I only discovered by reading the source (Hmm, that might suggest a problem...).
The fealing on the GTK list seems to be that there's a need for an entire new widget GtkFileChooser, and programs will eventually convert to this new API. IMHO, this is a very bad idea, as the oldstyle will never really go away any more than the win3.1 style has in the windows world. I think we ought to just add the new features and protect future APIs with preprocessor flags. Code for that might look like:
But that's for later. For now, the code that's up there works, and it might make your GTK-related life a lot more pleasanttr "A-Z" "a-z"|tr " -A" "\n" | sort | uniq -c | grep [a-z] | sort -n |grep -E -v ' the$| a$| and$| or$'|tail -n 20
Add or remove words to ignore (including any markup that gets in) from the grep -E -v line. Replace tail with head to get the least used words.
You can get a very dim idea of what an article is about, but that's really about it.
Yes, the "evidence" implicating Israel is mostly that they knew something was coming. We know they knew because they warned us. Yes, Israel opporates an effective intelligence service. Wish we did.
The war on drugs kills people in a racist way, but the death count is at most in the thousands -- probably the hundreds. That's not genocide. (It's not acceptable either, but that's a different argument.)
I'm not sure what you think feminism has done to America, but the richest, most powerful, and 17th freeist nation on earth doesn't "ruined" to me.
My point is more general than this though. Once people do encounter heresies, the memes stick in the back of their heads unchallenged. Most people are as reluctant to deny a heresy as to speak it (discussions like this one weaken the taboos). A strongly dogmatic society knows nothing and suspects everything.
If they actually go and cram cruft into the binary codec itself, well then we'll drop the whole thing. mplayer supports real format, both audio and video (x86 only, but I suspect so's this). mplayer's very sleak -- no problem there.
This should be attempted with caution. Spammers may be working with crackers, virus-writers, and possibly meat-space criminals. Operate from behind a firewall, in an open-source browser, with active content diabled. Never give your true name or similar information. If there are strange codes in the URL, remove them, they may encode your e-mail address.
So that's my little guide to vigilante anti-spam activity. Not that I'd ever encourage you to do anything illegal :-)
Almost nothing.
The number of 'chemicals' is a silly thing to measure in the first place. Chemically speaking, every protein is a different chemical. Every strand of DNA with a different gene on it is two different chemicals (3-5 side and 5-3 side, H-bonds don't count).
Very few foods are chemically simple: water, salt, sugar, vinegar, maybe grain alcohol-- that's probably it. Even these are a strech, as almost no one consumes straight salt, vinegar, or grain alcohol.
What's more important -- and a major flaw in the sited study, or at least its public face -- is that quantities matter. There are very few chemicals for which the smallest trace we can measure is dangerous. Any chemical is dangerous in sufficient dosage (even the infamous dihidrogen monoxide) and, except for self-reproducing molecules like BSE, none is harmful in insufficient dosage. Some molecules (such as PCBs) cause serious danger at concentrations below 10^-6. These are a real cause for concern, and there's a real need to watch for them. But we can detect them at concentrations orders of magnitude below that.
As a general rule, the number of chemicals with a given effect means nothing. The concentration of those chemicals means something. The concentration compared to the threshold for an effect means almost everything. Ignore this study; there are real ones out there.
If your water does not appear to match what the water department says it's sending out, then there are three possibilities: it actually does, just doesn't seem to; the water company messed up; or the pipes in your house are contaminating it.
Check with a neighbor and a chemical test kit. Photocopy the relevant pages from the Handbook of Chemistry and Physics at your local college library. In the end, you may need to check with a plumber (mnow *that's* expensive!).
Good luck!
I think there are similar organizations dedicated to KDE, Apache, Mozilla, and many other large projects. For that matter, a serious e-mail to a dev list offering money can probably get a 501-c3 set up.
Actually, SCO's case relies entirely on non-existant law. That and non-existant facts.
Oddly, IIRC, Slackware and Debian ship with really nice defaults here, but Redhat and Mandrake go way to slow. Weird.
First of all, a whole lot of windows users aren't somfortable doing hardware upgrades. Ram's especially difficult, at least on those motherboards (which used to be common) where you need a lot of brute force and you're squeezed right between the chip-fan and the power supply, so you have no room to work. For that matter, many old motherboards have weird requirenemnts about RAM (mandatory pairs, forbidden pairs, ordered slots, configurations forbidden because the BIOS-maker was in a bad mood...) and some left off the upper wires in the FSB, putting artificial caps on it.
Now combine these difficulties with someone who doesn't know how to take the case off in the first place, or the need to act on hundreds of computers, and sticking with what you have sounds like a good idea after all.
What I think the OP meant is that it's hard to recognize good programmers. Just because someone's degree says "MIT" doesn't guarentee they're any good, nor does the lack of any degree mean they aren't. Nor, as a handful of people suspect, is the opposite true. I suppose another programmer could tell a good programmer by reading his/her code, assuming a large body of that code was available for examination. Even if one is, it doesn't describe how long it took to write, and it will take a long time to study.
AFAICT, no one has found a solution to this yet. Education, experience, certifications, reputation.... Nothing seems to reliably seperate good programmers from bad -- and there are an awful lot of bad. Maybe someone'll come up with something soon.
if (Hsco->Hcode = Hlinux->Hcode)
{
}
Really, it explains so much.
It's written by Gregory Benford, who (according to reputation) should know better. Maybe he felt that he couldn't use any of his own good ideas in Asimov's universe, but no one else can write Asimov, so he was left with his own bad ideas. Or maybe Benford just sucks -- I haven't read anything else of his.
Avoid this book. Do read the original foundation trilogy (and, IMHO, all the other books in the sequence, except maybe _Foundation's_Edge_). Don't mix them up. That's like confusing Gentry Lee with Arthur C. Clarke. ::shudder::
I doubt it very much. They don't give full specs, so I wouldn't be at all surprised if it has a 100MHz bus, 24X CD-ROM, and generally low-end components for everything not listed. You can get a comparable system from microtel for $330. (I'm figuring that a 1.6GHz Athlon is at least as good as a 1.7GHz Celeron.) Admittedly, that's without moniter or printer, but that still doesn't cover the difference. Of course, one year of dial-up internet counts for something too. It costs about $150.
So, no, this isn't a particularly great deal. It's not a massive rip-off, either, though (assuming you want a year of dial-up, and don't already have a moniter), so it might make sense for people who don't know how to put things together (even on a very simple scale).
Incidentally, the original article included a few footnotes, and is available on GNU's site.
As long as we're plugging Vinge, someone should mention _Fat_Times_at_Fairmont_High_ (I think I just did), a novella he just wrote playing with these themes in a very near-future context. Someone upthread mentioned how these things will eventually build up as chemical polution -- eventually I guess we'll need them biodegradeable (one of the plot-lines in the story). That novella is becoming true at an alarming rate (not long ago, I took a 'local' test) so I highly recommend it to everyone here.