lots and lots of developers choose Qt instead of GTK+
I already covered this! This isn't a useful rebuttal to his position.
the fact is that Qt is the leading toolkit for commercial developement on Linux, and he shut it out. Not very smart IMO.
History will show whether UserLinux is really huge or a failed dead-end. If you are right, you will be able to look at the failure of UserLinux and say, "I was right."
Meanwhile, I wouldn't be shocked if someone started up KUserLinux. (Or call it "BusinessLinux". "UserLinux" is a strange name for an enterprise Linux.)
I had an idea for a great Trek-based show. I wanted to call it _Star Trek: First Contact_ but then they used that title for a movie.
Here's the idea: we know that a civilisation can join the Federation after it invents warp drive. Before inventing warp drive, it must be left alone (Prime Directive). But we also know that at least some of the time, the Federation might send agents undercover to a civilisation that is just on the verge of discovering warp drive, to give it a nudge or two. (Remember the episode where Riker was undercover doing this?)
So, the show would be story-arc based. The First Contact Team would go, stash the ship somewhere hard to detect (far side of a moon, say) and then start doing undercover stuff. They can't rely on magic technology to solve all their problems, because they have to stay undercover. (They can't even rely on the Universal Translator; they need to blend in and they must learn the local language without an accent!) It could take many episodes to wrap up one new planet and see it admitted to the Federation.
And sure, hire known good SF authors to write storylines.
They are still around. In the most recent episode, they helped retake the ship.
In classic "modern Trek" tradition, none of them got even a single line of dialog; I guess it's because you can pay actors much less when you don't give them any dialog. But anyway, there they were running around shooting phasers.
Well, Bruce Perens is wrong. If he was right, we would have lots and lots of commercial GTK+-apps
Please show me where Bruce Perens said "no one will use a library that isn't licensed under LGPL." Or even "people will so prefer LGPL that as soon as we release GTK everyone will flock to it."
What he actually said was something like "GNOME and KDE are evenly matched in most important ways; when one is behind the other, it soon catches up. If their licenses were the same I would have had to flip a coin to choose one. Since their licenses are not the same, I decided to go with the one that's LGPL." And then he listed reasons why he felt the LGPL is a win for an enterprise OS.
If you don't accept his premises, you won't accept his conclusion. But I don't think you can usefully refute his position by saying that there are few proprietary GTK+ apps.
But it seems to me that Qt is more popular when it comes to commercial software than GTK+ is.
Yes, I hear this a lot. I have several points about this.
0) From everything I hear, Qt is great and people like it. Many companies will consider it worth the money.
1) Qt apps integrate well with KDE, and KDE is popular. (It certainly helps that KDE was already working and useful before GNOME was more than an idea. In other words, KDE had a head start on GNOME, and also Qt had a head start on GTK+.)
2) TrollTech has a PR department, and makes sure we all hear about it when companies license Qt. GTK+ doesn't have a PR department. How many companies really write GTK+ apps? Who can really say?
3) The effects of a license can be subtle and take a long time to really become obvious. GTK+ is a relative newcomer compared to Qt, and who can really say how popular the two libraries will be in the future?
4) There is no reason why someone couldn't make a nice layer over GTK+ that would be easy to work with like Qt (perhaps even written in C++). If that were to happen, GTK+ would likely have a sharp uptick in popularity.
And it's not like you have to sell your testicles to pay for Qt. It's not THAT expensive.
Depends on who you are and what you want. If you are IBM or AT&T, the cost of Qt is a non-issue. For smaller companies, it could be an issue.
Some projects might use GTK+ purely because of budget, even within a large company: if you need to get permission to spend money on Qt, you might go with GTK+ because you don't need to get permission.
I would cheerfully pay $10 to $20 for a fun shareware game. I doubt that very many college kids will want to spend the money for a full Qt dev kit, so they probably won't use Qt to write cheap shareware games. (Unless they use GPL for their games, of course.)
But still, lots and lots of developer choose to pay for Qt, instead of getting GTK+ for free, why?
Because they like Qt and feel it is worth the money, of course. I never said Qt was worthless, and neither did Bruce Perens.
Sun - Now there is a company that can pick winning user interfaces (is CDE ugly or what).
Dude -- CDE has been around since before KDE or GNOME. Sun ships CDE because they have always shipped CDE. They are now shipping both GNOME and CDE, and someday they will ship only GNOME.
And Sun paid money for lots of usability testing that helped GNOME get better. GNOME 2 is way better than GNOME 1 (at least for newbies), and Sun gets some of the credit.
UserLinux - In spite of what they say, it is a religious issue with them.
Why should I believe you instead of believing Bruce Perens? He explained his reasons, and they make sense to me.
And he's not doing anything to break the Debian KDE, so companies that really want KDE will just install it.
If you want to see a good looking, easy to use, smooth running application, look at K3b.
I've never used it, but I hear good things about it and it looks really nice. There are at least two projects for GNOME to produce something roughly similar, and probably one of those will jell into a real product one of these days.
Pretty much, if KDE does something better than GNOME, GNOME will improve to match -- and vice versa. This competition is a good thing; it drives progress.
Gnome advocates now use QT's GPL licensing as a weapon against it
Well, some of them do. Others don't.
The reason the FSF wrote the LGPL, and not just the GPL, is that there are times when LGPL is appropriate. Many people (including Bruce Perens) think that a Linux distro intended for enterprise use should make it easy to write proprietary software, for those who want to do it... and thus think the LGPL would be a better license for the toolkit used in such a Linux distro.
Hard to say. It really depends what is in the 60 pages.
The document we are discussing here, by itself, could not be considered to be what IBM asked for. IBM asked for a specific list of all code SCO claims rights to. IBM also asked for a specific list of all code SCO claims IBM did anything improper with. This document contains a lot of "our engineers think" and "it appears", and if that's all there is to SCO's evidence, IBM's lawyers will have a field day with it. The judge won't be amused.
SCO claimed to have direct proof of "millions of lines" of infringing code, and they claimed to be showing that code (under draconian NDA) to people like Laura DiDio. The odd thing is that they don't seem to be in any hurry to simply take all that direct proof and hand it over in the courtroom.
SCO is now forked: either they turn that over and it gets debunked, or else they don't turn it over and IBM pounds them for unfair competition (spreading lies about infringing code).
I don't consider "SCO introduces the evidence and it holds up in court" as a real possibility.
Funny you should say that. I remember, years ago (around the time of Episode V), George Lucas was asked if the main actors would be back for episodes VII, VIII, and IX; his reply was "sure, if they can be made to look old enough." The implication was that he wanted to set the final trilogy a few decades after the end of Episode VI.
On the other hand, his more recent comment was that he "Never really had a story" for the final trilogy.
On the gripping hand, it's clear that not having a story never really stops him. I refuse to believe that he planned all along for young Skywalker to build C-3PO, for example. I refuse to believe that during filming of Episode IV that he had already planned that Darth Vader would have grown up on Tattooine. I strongly suspect that the whole Luke/Leia being brother/sister was invented after he already had finished Episode IV. He just makes stuff up and sticks it on.
Star Wars is great not because of George Lucas, who got the ball rolling, but because of the contributions of so many people. The sets, costumes, and effects represent man-years of work by many people. At this point you could take Star Wars out of George Lucas's hands and make new movies. (Sadly, they might be better movies for that; George Lucas's recent track record isn't great.)
I predict we'll see Episodes VII, VIII, and IX someday, no matter what George Lucas says now. And we'll probably see other movies as well. (I'd like to see some "Jedi Academy" movies.)
I cant see companies offering downloads in lossless format anytime in the near future
You can if you visit magnatune.com, since they offer FLAC and even WAV files, as well as Ogg Vorbis and MP3. You could, if you were insane, even download all four formats.
I read your article, but I am not as worried as you are.
First, my credentials: I haven't run an organized study of spam, as you have, but I did set up a Bayesian filter, SpamProbe, on my mail server (and I wrote an article about it). I get about 150 spam messages per day, and I only see the ones that get past my Bayesian filter. So I have looked over dozens of spams to see why they fooled my filter. (My filter is about 95% effective, and once I had it trained, I haven't observed any false positives.)
Yes, if a spammer works carefully, he can craft a message that will have a better chance of slipping past a Bayesian filter. But my Bayesian filter is not 100% effective anyway; as long as I only have to manually handle 5-15 messages per day, I'd say the filter is working. So the question is not whether the spammers can ever slip a message past the filter; the question is whether the spammers can completely destroy the usefulness of Bayesian filters, as you fear.
Bayesian filters look at the whole message, and they can learn to recognize spam in unexpected ways. For example, HTML font tags that set large red letters are a good spam indicator. HTML font tags that set white-on-white text are another. So Bayesian filters will force spammers to change the format of their spam.
Most spammers want you to call a phone number or view a URL. Since the Bayesian filter will learn the phone numbers and URLs are spam flags, Bayesian filters will force spammers to keep setting up new phone numbers and servers.
The "from" addresses of my friends will quickly become good ham indicators, and that will be difficult for the spammers to exploit (since everyone has different friends).
Also, my understanding is that you cannot really "poison" a word for Bayesian filtering; all you can do is lessen its usefulness as a spam/ham indicator. If spammers use different hammy words for each spam, the poison's dosage will be diluted; while if they use the same hammy words for each spam, those words will then be a legitimate spam flag.
There probably are a few refinements that could be made to spam filters. I'd like to see a spam filter that, if there is both an HTML part and a plain text part, only checks the HTML. That way the spammers can include ham in the text part and it won't affect the filtering.
In summary, I am reasonably hopeful that there is no way for spammers to completely defeat Bayesian filtering. The best they can hope to do is to sneak some mildly-phrased messages by the filters.
P.S. I agree with you that the ultimate anti-spam measure would be a "for-pay" mail system. I envision a mail protocol that allows you to specify how much it costs to send you an email: you put your friends on the free list, and otherwise it costs 5 cents or whatever. If you are really famous you might raise the cost up to reduce the volume of email you receive. There should be a mechanism in place to quickly refund the costs, and friends should be identified with a digital signature, not by an easily forged string. Spam only works because it's so cheap to send many messages, so a 0.001% response rate is enough. At even 5 cents per message, spam wouldn't be cost-effective anymore. You would still get ads in the mail, but they would be less obnoxious and more carefully targeted. Send me an ad for Mexican Viagra and you won't get your 5 cents back, but send me an ad for something I actually want and I'll consider it.
all lawyers should be paid by the government with our tax dollars.
Good grief, no! No! Bad idea!
If you want more lawyers and more lawsuits, have government pay for it. If you want less lawsuits, do something else.
The best idea would be to adopt the "loser pays" system, where the loser of a court case has to pay the legal fees of the winner. There should be a down side to a stupid lawsuit, so companies like SCO will think twice before trying them.
If you adopt "loser pays" you need a few other tweaks to the system, to make sure that the rights of the poor don't get stepped on. (It would be bad if the truly poor did not dare risk a lawsuit since they could not risk having to pay for lawyers.)
I'm affraid of a system that allows corporations like SCO to exist.
Never fear, the system in the US will not allow SCO to continue to exist indefinitely. It's going to take a long time, much longer than I would prefer, but when it's over SCO will be a smoking crater.
here in the US our legal system offers no protection to businesses or consumers from corporations that behave like SCO.
The defense against stupid lawsuits from companies like SCO is the "countersuit". Note that IBM has already filed a countersuit against SCO. (SCO is now trapped: they can't just walk away from the lawsuit, because they can drop their suit against IBM but they cannot make IBM drop the countersuit.)
This is a cycle of continuous manipulation that's called capitalism.
Nonsense. Capitalism is where people are allowed to sell things to each other; look it up. SCO is trying to use the legal system to prey on other companies, and that is not called capitalism.
According to the specs in the Google cache (thanks AKAImBatman, there is a docking cable. I quote:
OQO docking cable includes
* 3D accelerated 1280 x 1024 video output (VGA and LVDS)
* Serialized PCI
* Additional USB
* Additional FireWire (IEEE1394)
* Ethernet
* DC power
* Audio out
So my guess is that the weird port is the port for the docking cable.
"Serialized PCI". Cool. You could make a docking port with PCI cards in it!
Maybe they will sue, but I doubt they will win. The Ogg guys were careful. They had lawyers looking over their work at every step. They worked hard to make sure that they were 100% in the clear on all patents.
Someone could sue them anyway. But they have certainly done their homework on the patents issue, and I'm confident they won't lose.
It's possible that IBM is just posturing to get a better price from MS, but I doubt it. IBM offers consulting services, and they sell support to companies adopting Linux. It makes sense for them to "eat their own dogfood", and once it's done they can point to themselves as a success story.
However, IBM isn't a monolith, and various groups inside IBM might go off in their own directions. It's possible that some parts of IBM will take the deal from MS and go with the cheaper licenses for Windows.
But IBM would be an ideal company for rolling out Linux everywhere: they have so many employees that they stand to save a whole bunch of money (on license fees they no longer have to pay), they can get computer consulting from their own consultants, and they can use the resulting success as a marketing tool (to help them sell consulting services). I think the only real question is "when", not "if".
P.S. Naturally they will always have some Windows desktops running somewhere. As long as they sell computers running Windows they will need to have Windows in-house for testing, for one thing.
0) AAC doesn't have any DRM. Apple wraps AAC with DRM anyhow. Ogg Vorbis can be wrapped in DRM, easily.
1) What makes Ogg Vorbis better than AAC or WMA? Mainly, it's not covered by patents, so it is legal for people to write free versions. (There may be an open source AAC but it will still be covered by patents.)
Should we assume that this will have the exact same internal software, or is there a chance that HP will change things around? I'd love an iPod that could play my Ogg Vorbis tunes.
I wonder if the contract from Apple would even allow this.
Perhaps, for the PC market, HP would want to support Windows Media Audio files... and if as they are doing that, they might as well add Ogg Vorbis support.
I also wonder if HP will put FireWire on all their computers now, or whether they will just depend on the USB 2.0 support Apple already has for the Windows version of the iPod.
0) I don't think IBM expected the PC to be the monster hit it turned out to be. If they had realized ahead of time how huge it would be, they would have done things differently.
1) Remember how awful the keyboard was? Great IBM clackity-clack keys, but important keys were in weird places, and I was always typing '/' when I wanted to hit the Shift. One theory says that IBM was worried that the PC would eat into sales of IBM word processors (i.e., machines sold by IBM that did only one thing, word processing). Of course the PC totally wiped out the market for dedicated word processors, and of course third-party keyboards were instantly available.
2) There is a lot of evidence that IBM wasn't just taking a long time in product development, but actually was trying to slow the pace of PC development. The IBM PC AT, when it shipped, had a 286 with a socketed 6 MHz clock chip. The circuits were all fast enough for an 8 MHz clock. It appears that IBM made a last-minute decision to ship with 6 MHz instead of 8. Then, as you noted, Compaq actually beat IBM to market with the first 386. I think IBM was having internal wars, because the 386 was where the PC was starting to really be a threat to IBM's minis. (The 386 is the first chip that could run *NIX properly!)
It does appear that IBM has learned their lessons. They aren't trying to hold back the industry anymore.
Some people may wonder whether we can ever truly trust a large corporation. With free software, we can welcome their contributions without needing to trust them. And as the old saying goes: "When you don't have to trust someone, you know you can trust them."
GNOME 2.0 had geeks on Slashdot shrieking in agony over all the features that were cut from GNOME 1.x. Did you know that GNOME 2.x, by default, only has one way to maximize a window? Shocking!
So no, I don't think GNOME is bloated with featuritis. I think the GNOME guys have done a great job of paring things down to where you can quickly find the features you actually want to use.
steveha
P.S. If you actually miss the "maximize horizontally only" or "maximize vertically only" commands, you can choose to run GNOME with Sawfish and get them back.
I wanted a very quiet PC. I bought a huge, solid, steel case, customized with a 120mm fan on the back (theory: big and slow == quiet). I got a Zalman "flower" heatsink/fan (HSF), and mounted the big Zalman fan to blow over it. I got an Enermax power supply with a speed adjustment. It's pretty quiet; I hear a fan, but I think that's the GeForce 4 card's fan.
I wanted to make a quiet PC for my wife. I bought a Lian Li aluminum case, an Arctic Cooling HSF, and a similar Enermax power supply. No fancy 120mm fan on the back, just a standard 80mm fan, but I used a very quiet one with thermostatic control so it is very slow and quiet when the system is cool. This computer, as it turns out, is almost completely silent! Much quieter than my computer. I did use a GeForce4 MX board in her system, because it has just a passive heat sink (no cooling fan), so perhaps that explains it.
I loved working with the Lian Li case. It's a PC-60 model with USB. I also much preferred the Arctic Cooling HSF. I got my Arctic Cooling HSF form SVC.com:
P.S. About the quiet power supply: I got a 365 Watt power supply with two cooling fans, one with a speed control and one with a thermal control (automatically runs faster when hot). This power supply has "Active PFC", which I don't completely understand, but I gather it is a more efficient way to convert AC to DC and thus makes less waste heat. It has a 3-pin jumper to attach to the motherboard, so the motherboard can monitor the speed of the power supply's main fan, and also so that the motherboard can signal to the power supply that it wants all fans powered down for sleep mode. (I don't think either computer is ever really sleeping at the moment. I ought to play around with ACPI and get that working, but they are quiet enough that it hasn't been a priority.) I ordered the power supplies from Directron. This one isn't the exact same model but it has the same features:
P.S. Why is it really a tale of three Athlon XPs? Because I crunched one trying to put on the HSF. With an Athlon XP, be very, very careful when putting on the HSF. You can make a very expensive mistake! I'm looking forward to Athlon64 and Opteron with a heat spreader protecting the chip.
Slashdot Mirror
lots and lots of developers choose Qt instead of GTK+
I already covered this! This isn't a useful rebuttal to his position.
the fact is that Qt is the leading toolkit for commercial developement on Linux, and he shut it out. Not very smart IMO.
History will show whether UserLinux is really huge or a failed dead-end. If you are right, you will be able to look at the failure of UserLinux and say, "I was right."
Meanwhile, I wouldn't be shocked if someone started up KUserLinux. (Or call it "BusinessLinux". "UserLinux" is a strange name for an enterprise Linux.)
steveha
I had an idea for a great Trek-based show. I wanted to call it _Star Trek: First Contact_ but then they used that title for a movie.
Here's the idea: we know that a civilisation can join the Federation after it invents warp drive. Before inventing warp drive, it must be left alone (Prime Directive). But we also know that at least some of the time, the Federation might send agents undercover to a civilisation that is just on the verge of discovering warp drive, to give it a nudge or two. (Remember the episode where Riker was undercover doing this?)
So, the show would be story-arc based. The First Contact Team would go, stash the ship somewhere hard to detect (far side of a moon, say) and then start doing undercover stuff. They can't rely on magic technology to solve all their problems, because they have to stay undercover. (They can't even rely on the Universal Translator; they need to blend in and they must learn the local language without an accent!) It could take many episodes to wrap up one new planet and see it admitted to the Federation.
And sure, hire known good SF authors to write storylines.
steveha
many shows in japan are canceled while they are hot. Evangelion, anyone?
Huh? There was only one season of Evangelion, and it wasn't cancelled (that one season ran all episodes). Not a good example.
steveha
Whatever happened to the Space Delta Force guys?
They are still around. In the most recent episode, they helped retake the ship.
In classic "modern Trek" tradition, none of them got even a single line of dialog; I guess it's because you can pay actors much less when you don't give them any dialog. But anyway, there they were running around shooting phasers.
steveha
Well, Bruce Perens is wrong. If he was right, we would have lots and lots of commercial GTK+-apps
Please show me where Bruce Perens said "no one will use a library that isn't licensed under LGPL." Or even "people will so prefer LGPL that as soon as we release GTK everyone will flock to it."
What he actually said was something like "GNOME and KDE are evenly matched in most important ways; when one is behind the other, it soon catches up. If their licenses were the same I would have had to flip a coin to choose one. Since their licenses are not the same, I decided to go with the one that's LGPL." And then he listed reasons why he felt the LGPL is a win for an enterprise OS.
If you don't accept his premises, you won't accept his conclusion. But I don't think you can usefully refute his position by saying that there are few proprietary GTK+ apps.
But it seems to me that Qt is more popular when it comes to commercial software than GTK+ is.
Yes, I hear this a lot. I have several points about this.
0) From everything I hear, Qt is great and people like it. Many companies will consider it worth the money.
1) Qt apps integrate well with KDE, and KDE is popular. (It certainly helps that KDE was already working and useful before GNOME was more than an idea. In other words, KDE had a head start on GNOME, and also Qt had a head start on GTK+.)
2) TrollTech has a PR department, and makes sure we all hear about it when companies license Qt. GTK+ doesn't have a PR department. How many companies really write GTK+ apps? Who can really say?
3) The effects of a license can be subtle and take a long time to really become obvious. GTK+ is a relative newcomer compared to Qt, and who can really say how popular the two libraries will be in the future?
4) There is no reason why someone couldn't make a nice layer over GTK+ that would be easy to work with like Qt (perhaps even written in C++). If that were to happen, GTK+ would likely have a sharp uptick in popularity.
And it's not like you have to sell your testicles to pay for Qt. It's not THAT expensive.
Depends on who you are and what you want. If you are IBM or AT&T, the cost of Qt is a non-issue. For smaller companies, it could be an issue.
Some projects might use GTK+ purely because of budget, even within a large company: if you need to get permission to spend money on Qt, you might go with GTK+ because you don't need to get permission.
I would cheerfully pay $10 to $20 for a fun shareware game. I doubt that very many college kids will want to spend the money for a full Qt dev kit, so they probably won't use Qt to write cheap shareware games. (Unless they use GPL for their games, of course.)
But still, lots and lots of developer choose to pay for Qt, instead of getting GTK+ for free, why?
Because they like Qt and feel it is worth the money, of course. I never said Qt was worthless, and neither did Bruce Perens.
steveha
Sun - Now there is a company that can pick winning user interfaces (is CDE ugly or what).
Dude -- CDE has been around since before KDE or GNOME. Sun ships CDE because they have always shipped CDE. They are now shipping both GNOME and CDE, and someday they will ship only GNOME.
And Sun paid money for lots of usability testing that helped GNOME get better. GNOME 2 is way better than GNOME 1 (at least for newbies), and Sun gets some of the credit.
UserLinux - In spite of what they say, it is a religious issue with them.
Why should I believe you instead of believing Bruce Perens? He explained his reasons, and they make sense to me.
And he's not doing anything to break the Debian KDE, so companies that really want KDE will just install it.
If you want to see a good looking, easy to use, smooth running application, look at K3b.
I've never used it, but I hear good things about it and it looks really nice. There are at least two projects for GNOME to produce something roughly similar, and probably one of those will jell into a real product one of these days.
Pretty much, if KDE does something better than GNOME, GNOME will improve to match -- and vice versa. This competition is a good thing; it drives progress.
steveha
Gnome advocates now use QT's GPL licensing as a weapon against it
Well, some of them do. Others don't.
The reason the FSF wrote the LGPL, and not just the GPL, is that there are times when LGPL is appropriate. Many people (including Bruce Perens) think that a Linux distro intended for enterprise use should make it easy to write proprietary software, for those who want to do it... and thus think the LGPL would be a better license for the toolkit used in such a Linux distro.
steveha
i'm an MCP
:-)
Do you have some guy named Tron throwing a glowing frisbee at you now?
Sorry, couldn't resist.
steveha
SCO _have_ complied for the most part
Hard to say. It really depends what is in the 60 pages.
The document we are discussing here, by itself, could not be considered to be what IBM asked for. IBM asked for a specific list of all code SCO claims rights to. IBM also asked for a specific list of all code SCO claims IBM did anything improper with. This document contains a lot of "our engineers think" and "it appears", and if that's all there is to SCO's evidence, IBM's lawyers will have a field day with it. The judge won't be amused.
SCO claimed to have direct proof of "millions of lines" of infringing code, and they claimed to be showing that code (under draconian NDA) to people like Laura DiDio. The odd thing is that they don't seem to be in any hurry to simply take all that direct proof and hand it over in the courtroom.
SCO is now forked: either they turn that over and it gets debunked, or else they don't turn it over and IBM pounds them for unfair competition (spreading lies about infringing code).
I don't consider "SCO introduces the evidence and it holds up in court" as a real possibility.
steveha
All the main actors have not aged well
Funny you should say that. I remember, years ago (around the time of Episode V), George Lucas was asked if the main actors would be back for episodes VII, VIII, and IX; his reply was "sure, if they can be made to look old enough." The implication was that he wanted to set the final trilogy a few decades after the end of Episode VI.
On the other hand, his more recent comment was that he "Never really had a story" for the final trilogy.
On the gripping hand, it's clear that not having a story never really stops him. I refuse to believe that he planned all along for young Skywalker to build C-3PO, for example. I refuse to believe that during filming of Episode IV that he had already planned that Darth Vader would have grown up on Tattooine. I strongly suspect that the whole Luke/Leia being brother/sister was invented after he already had finished Episode IV. He just makes stuff up and sticks it on.
Star Wars is great not because of George Lucas, who got the ball rolling, but because of the contributions of so many people. The sets, costumes, and effects represent man-years of work by many people. At this point you could take Star Wars out of George Lucas's hands and make new movies. (Sadly, they might be better movies for that; George Lucas's recent track record isn't great.)
I predict we'll see Episodes VII, VIII, and IX someday, no matter what George Lucas says now. And we'll probably see other movies as well. (I'd like to see some "Jedi Academy" movies.)
steveha
I cant see companies offering downloads in lossless format anytime in the near future
You can if you visit magnatune.com, since they offer FLAC and even WAV files, as well as Ogg Vorbis and MP3. You could, if you were insane, even download all four formats.
steveha
This is one of the reasons I use SpamProbe. It uses two-word pairs.
steveha
I read your article, but I am not as worried as you are.
First, my credentials: I haven't run an organized study of spam, as you have, but I did set up a Bayesian filter, SpamProbe, on my mail server (and I wrote an article about it). I get about 150 spam messages per day, and I only see the ones that get past my Bayesian filter. So I have looked over dozens of spams to see why they fooled my filter. (My filter is about 95% effective, and once I had it trained, I haven't observed any false positives.)
Yes, if a spammer works carefully, he can craft a message that will have a better chance of slipping past a Bayesian filter. But my Bayesian filter is not 100% effective anyway; as long as I only have to manually handle 5-15 messages per day, I'd say the filter is working. So the question is not whether the spammers can ever slip a message past the filter; the question is whether the spammers can completely destroy the usefulness of Bayesian filters, as you fear.
Bayesian filters look at the whole message, and they can learn to recognize spam in unexpected ways. For example, HTML font tags that set large red letters are a good spam indicator. HTML font tags that set white-on-white text are another. So Bayesian filters will force spammers to change the format of their spam.
Most spammers want you to call a phone number or view a URL. Since the Bayesian filter will learn the phone numbers and URLs are spam flags, Bayesian filters will force spammers to keep setting up new phone numbers and servers.
The "from" addresses of my friends will quickly become good ham indicators, and that will be difficult for the spammers to exploit (since everyone has different friends).
Also, my understanding is that you cannot really "poison" a word for Bayesian filtering; all you can do is lessen its usefulness as a spam/ham indicator. If spammers use different hammy words for each spam, the poison's dosage will be diluted; while if they use the same hammy words for each spam, those words will then be a legitimate spam flag.
There probably are a few refinements that could be made to spam filters. I'd like to see a spam filter that, if there is both an HTML part and a plain text part, only checks the HTML. That way the spammers can include ham in the text part and it won't affect the filtering.
In summary, I am reasonably hopeful that there is no way for spammers to completely defeat Bayesian filtering. The best they can hope to do is to sneak some mildly-phrased messages by the filters.
P.S. I agree with you that the ultimate anti-spam measure would be a "for-pay" mail system. I envision a mail protocol that allows you to specify how much it costs to send you an email: you put your friends on the free list, and otherwise it costs 5 cents or whatever. If you are really famous you might raise the cost up to reduce the volume of email you receive. There should be a mechanism in place to quickly refund the costs, and friends should be identified with a digital signature, not by an easily forged string. Spam only works because it's so cheap to send many messages, so a 0.001% response rate is enough. At even 5 cents per message, spam wouldn't be cost-effective anymore. You would still get ads in the mail, but they would be less obnoxious and more carefully targeted. Send me an ad for Mexican Viagra and you won't get your 5 cents back, but send me an ad for something I actually want and I'll consider it.
steveha
all lawyers should be paid by the government with our tax dollars.
Good grief, no! No! Bad idea!
If you want more lawyers and more lawsuits, have government pay for it. If you want less lawsuits, do something else.
The best idea would be to adopt the "loser pays" system, where the loser of a court case has to pay the legal fees of the winner. There should be a down side to a stupid lawsuit, so companies like SCO will think twice before trying them.
If you adopt "loser pays" you need a few other tweaks to the system, to make sure that the rights of the poor don't get stepped on. (It would be bad if the truly poor did not dare risk a lawsuit since they could not risk having to pay for lawyers.)
Eh, enough off-topic for today.
steveha
I'm affraid of a system that allows corporations like SCO to exist.
Never fear, the system in the US will not allow SCO to continue to exist indefinitely. It's going to take a long time, much longer than I would prefer, but when it's over SCO will be a smoking crater.
here in the US our legal system offers no protection to businesses or consumers from corporations that behave like SCO.
The defense against stupid lawsuits from companies like SCO is the "countersuit". Note that IBM has already filed a countersuit against SCO. (SCO is now trapped: they can't just walk away from the lawsuit, because they can drop their suit against IBM but they cannot make IBM drop the countersuit.)
This is a cycle of continuous manipulation that's called capitalism.
Nonsense. Capitalism is where people are allowed to sell things to each other; look it up. SCO is trying to use the legal system to prey on other companies, and that is not called capitalism.
steveha
So my guess is that the weird port is the port for the docking cable.
"Serialized PCI". Cool. You could make a docking port with PCI cards in it!
steveha
Maybe they will sue, but I doubt they will win. The Ogg guys were careful. They had lawyers looking over their work at every step. They worked hard to make sure that they were 100% in the clear on all patents.
Someone could sue them anyway. But they have certainly done their homework on the patents issue, and I'm confident they won't lose.
steveha
It's possible that IBM is just posturing to get a better price from MS, but I doubt it. IBM offers consulting services, and they sell support to companies adopting Linux. It makes sense for them to "eat their own dogfood", and once it's done they can point to themselves as a success story.
However, IBM isn't a monolith, and various groups inside IBM might go off in their own directions. It's possible that some parts of IBM will take the deal from MS and go with the cheaper licenses for Windows.
But IBM would be an ideal company for rolling out Linux everywhere: they have so many employees that they stand to save a whole bunch of money (on license fees they no longer have to pay), they can get computer consulting from their own consultants, and they can use the resulting success as a marketing tool (to help them sell consulting services). I think the only real question is "when", not "if".
P.S. Naturally they will always have some Windows desktops running somewhere. As long as they sell computers running Windows they will need to have Windows in-house for testing, for one thing.
steveha
0) AAC doesn't have any DRM. Apple wraps AAC with DRM anyhow. Ogg Vorbis can be wrapped in DRM, easily.
1) What makes Ogg Vorbis better than AAC or WMA? Mainly, it's not covered by patents, so it is legal for people to write free versions. (There may be an open source AAC but it will still be covered by patents.)
steveha
as bad as you want Ogg, you will settle for AAC and buy an iPod
Actually, no, I won't.
if you won't, then you are a market minority so small that Apple doesn't have the time and money to spend reaching you.
But companies much smaller than Apple, such as iRiver, do have the time and money to spend reaching me.
steveha
Should we assume that this will have the exact same internal software, or is there a chance that HP will change things around? I'd love an iPod that could play my Ogg Vorbis tunes.
I wonder if the contract from Apple would even allow this.
Perhaps, for the PC market, HP would want to support Windows Media Audio files... and if as they are doing that, they might as well add Ogg Vorbis support.
I also wonder if HP will put FireWire on all their computers now, or whether they will just depend on the USB 2.0 support Apple already has for the Windows version of the iPod.
steveha
I mostly agree with your comments. A few notes:
0) I don't think IBM expected the PC to be the monster hit it turned out to be. If they had realized ahead of time how huge it would be, they would have done things differently.
1) Remember how awful the keyboard was? Great IBM clackity-clack keys, but important keys were in weird places, and I was always typing '/' when I wanted to hit the Shift. One theory says that IBM was worried that the PC would eat into sales of IBM word processors (i.e., machines sold by IBM that did only one thing, word processing). Of course the PC totally wiped out the market for dedicated word processors, and of course third-party keyboards were instantly available.
2) There is a lot of evidence that IBM wasn't just taking a long time in product development, but actually was trying to slow the pace of PC development. The IBM PC AT, when it shipped, had a 286 with a socketed 6 MHz clock chip. The circuits were all fast enough for an 8 MHz clock. It appears that IBM made a last-minute decision to ship with 6 MHz instead of 8. Then, as you noted, Compaq actually beat IBM to market with the first 386. I think IBM was having internal wars, because the 386 was where the PC was starting to really be a threat to IBM's minis. (The 386 is the first chip that could run *NIX properly!)
It does appear that IBM has learned their lessons. They aren't trying to hold back the industry anymore.
Some people may wonder whether we can ever truly trust a large corporation. With free software, we can welcome their contributions without needing to trust them. And as the old saying goes: "When you don't have to trust someone, you know you can trust them."
steveha
If you think Gnome is bloated from featuritis now
Huh?
GNOME 2.0 had geeks on Slashdot shrieking in agony over all the features that were cut from GNOME 1.x. Did you know that GNOME 2.x, by default, only has one way to maximize a window? Shocking!
So no, I don't think GNOME is bloated with featuritis. I think the GNOME guys have done a great job of paring things down to where you can quickly find the features you actually want to use.
steveha
P.S. If you actually miss the "maximize horizontally only" or "maximize vertically only" commands, you can choose to run GNOME with Sawfish and get them back.
I think the trick is to have the heat sink perfectly straight on. If it is perfectly square to the top of the CPU package, it cannot crush the chip.
steveha
Well, three really.
I wanted a very quiet PC. I bought a huge, solid, steel case, customized with a 120mm fan on the back (theory: big and slow == quiet). I got a Zalman "flower" heatsink/fan (HSF), and mounted the big Zalman fan to blow over it. I got an Enermax power supply with a speed adjustment. It's pretty quiet; I hear a fan, but I think that's the GeForce 4 card's fan.
I wanted to make a quiet PC for my wife. I bought a Lian Li aluminum case, an Arctic Cooling HSF, and a similar Enermax power supply. No fancy 120mm fan on the back, just a standard 80mm fan, but I used a very quiet one with thermostatic control so it is very slow and quiet when the system is cool. This computer, as it turns out, is almost completely silent! Much quieter than my computer. I did use a GeForce4 MX board in her system, because it has just a passive heat sink (no cooling fan), so perhaps that explains it.
I loved working with the Lian Li case. It's a PC-60 model with USB. I also much preferred the Arctic Cooling HSF. I got my Arctic Cooling HSF form SVC.com:
http://svc.com/arcoolsupsil.html
P.S. About the quiet power supply: I got a 365 Watt power supply with two cooling fans, one with a speed control and one with a thermal control (automatically runs faster when hot). This power supply has "Active PFC", which I don't completely understand, but I gather it is a more efficient way to convert AC to DC and thus makes less waste heat. It has a 3-pin jumper to attach to the motherboard, so the motherboard can monitor the speed of the power supply's main fan, and also so that the motherboard can signal to the power supply that it wants all fans powered down for sleep mode. (I don't think either computer is ever really sleeping at the moment. I ought to play around with ACPI and get that working, but they are quiet enough that it hasn't been a priority.) I ordered the power supplies from Directron. This one isn't the exact same model but it has the same features:
http://www.directron.com/eg465axve.html
P.S. Why is it really a tale of three Athlon XPs? Because I crunched one trying to put on the HSF. With an Athlon XP, be very, very careful when putting on the HSF. You can make a very expensive mistake! I'm looking forward to Athlon64 and Opteron with a heat spreader protecting the chip.
steveha