Nanotech is not about making big things small. People watch Star Trek or read The Diamond Age and assume that nano is about little robots with little CPU's and little mechanical parts. And that just ain't so. We already have tiny molecular cutting devices: they're called enzymes. Yep, it's the squishy science of chemistry, where we deal with weird funny smelling liquids rather than neat shiny MicroMachines. The kind of nanolithography Feynmann talks about is rather similar to how we make chips now, and they're infinitely more flexible than a micro-encyclopedia. There's not nearly as much utility in putting all the manufacturing effort. toward creating single-purpose micro-devices when you can create a general-purpose one like a CPU.
> Even Moore himself admits that his law can't hold forever, only he is wise enough not to put a timetable on it.
Moore wasn't some digerati pundit making grandiose prognostications, unlike the current crop of wannabe net.prophets. He only predicted that a trend toward doubling the number of transistors on a chip every 18 months would continue for at least the next 10 years -- that was over 20 years ago.
Actually, didn't the move to RISC actually result in LESS transistors?
Re:Why Mozilla 5.0 will die. (At least on the Mac)
on
Whither Netscape 5.0?
·
· Score: 2
So why aren't you predicting the death of Quicktime 4 while you're at it? The interface for it is so bad, it got itself a whole column in the Interface Hall of Shame
> How about standard compliance? Mozilla is completely standard compliant, easing web developers a real lot on developing applications which works on all browsers. > How about XUL?... > How about being built as a library... > How about XML support? > How about a more open architecture?... > How about a more open development model?...
Why yes, just look at the thriving Chinese, North Korean, and Cuban software outfits. Oh and you can spare me the claptrap about how those aren't "true" communism/socialism/whatever, because central control in the name of "the people" has corrupted and failed every last time.
> My thinking is that they died out for a reason, and they'll die out again
It seems most likely they were hunted to extinction. They'd die out again now because there simply isn't enough gene stock for them to survive. That and it's just not cold where elephants would tend to feed now.
1.The originator may combine distributed derived works produced by others and released under TGPL with the originator's proprietary code and redistribute the result.
Redistribute under what licenses? If a patch is submitted without specifying whether it's for the GPL'd or TGPL'd fork, does it automatically go into the TGPL'd fork?
And I say "fork", because that's exactly what will happen the instant the code is released. Someone will grab it, rename it, put it under GPL, and isolate all further improvements to the GPL'd version. Which leads one to wonder why you wouldn't just create the fork in the first place by simply dual-licensing it.
Better would be a license that creates a publisher/author relationship, where copyright of any modifications is automatically assigned to the originator. Some sort of "good faith" enforcement clause would be needed to ensure that redistribution rights are not terminated. The FSF has this written into their charter. If they decided to revoke the GPL for every GNU package and turn it into closed-source commercialware, the rights of the consumer (redistributor) under the GPL would be upheld because the FSF would be acting in bad faith. Well, probably: "good faith" is always a judgement call. And they can play with the license as they please, within the boundaries of good faith: they OWN every line of code in official GNU software thanks to the copyright assignment agreements they require.
The charter of the KDE Free Qt foundation would be one place to look for a charter spelled out explicitly. If Troll ever goes out of business, Qt becomes BSD licensed.
But all this said, licenses are not computer programs. They are subject to interpretation, loopholes, and flat-out nullification in the case of bad faith, violation of existing law, invalidity under statutes of contract law, or just the amount of money one can spend on lawyers. If you do not trust the license issuer, you cannot ever really trust the license.
Tip when using a thumb trackball: don't use your thumb. At least not all the time. You should be able to sorta glide all your fingers over it if you don't mash your hand down on it, which is worse to your wrists than using a mouse, which at least gives you lateral movement. Move the trackball close to you so you don't feel like you're reaching out to use it. It should feel like your hand just drops on the trackball, you should neither have to reach out for it nor mash your hand down on it.
As for your mouse, similar advice. Keep the heel of your hand off the thing and control it with your thumb and little finger. A light mouse with a slim profile helps, which pretty much rules out a MS Mouse. Of course I have huge hands so this is easy for me to do. But a light touch is the key. Now i just need to make better keymaps to rid myself of emacs pinky (no i will not use vi)
> Many spy thrillers have claimed there is another classification above Top Secret, without needing to shoot me, can you confirm or deny that?:)
SCI: Special Compartmentalized Intelligence.
It's not any more secret than Top Secret, but it has more stringent rules concerning its distribution. Having Top Secret clearance doesn't automatically clear you for SCI. It's the codified definition of "need to know". SCI information viewed on computers is done in a separate room on separate wiring where even the nearby water pipes are electrically isolated. Very secure stuff. But otherwise a well-known level of security.
The stuff more secret than that is the stuff that doesn't have a classification. It's the stuff the president or the director of the NSA or CIA says to another aide "don't tell this to anyone, ANYONE, got it?" In other words, pretty much all your extralegal stuff.
> It's up to window manager, not desktop environment to deal with global key bindings.
Grog no understand, Grog have desktop, but desktop no work way other desktop work. Grog want feature in desktop and Grog not care what part of desktop do what part.
Now if SawMill can do that, then it already sounds a lot more powerful than E, and I might suggest that gnome ship with some keybindings that do all that. Matching a window by its title does not seem to be the most elegant way of doing this. Something on the order of xtoolwait that gives me the windowid of the newly mapped window would be far more preferable.
> Is there a way to defend yourself from a (at least frivolous) lawsuit without spending thousands of dollars you can't get back even when you win?
If you're really confident that it's frivolous, you ask for it to be summarily thrown out of court. This usually doesn't happen, in which case, hope for a lawyer that does pro bono work or you're screwed. And you still get to miss a lot of work. Welcome to justice, American style.
I was using kde since the beta versions, and tried out gnome at the 1.0 release. After many cries of disgust and peals of ridiculing laughter, I swore off gnome forever. Recently I played musical distributions, going to debian then back to redhat 6.0 (I won't get into why) and in the process, installed gnome and not kde, as I wanted to work with KDE from CVS only.
It's not bad now. It's definitely a lot faster. E is even respectable looking, though I have plans to replace that with Sawmill. What with the million and a half language bindings for gtk+, I definitely can't hold the language card over its head. But I still have some issues with gnome. Mind you I have 'em with KDE, but this isn't a KDE forum at the moment:
DOCUMENTATION!!!: I can't say it enough. The state of GNOME's documentation, from gtk to everything else, is pathetic. The KDE classes themselves aren't always terrifically documented, but Qt's documentation is phenomenal. Make no mistake, GNOME has irretrievably lost potential developers because they couldn't get the documentation they needed. This has to be addressed. Raw API docs don't cut it, examples and introductory texts are essential.
If you're shamelessly ripping off items from other desktops, here's something to take from CDE: I'd like a panel setup that had several drawers of grouped functions ala CDE instead of one awful Cascading Menu From Hell known as the "Start Menu". Take this from CDE: create drawers that have an icon that when clicked, start the app, and an arrow *above* them that opens the drawer. Have drop targets in all the drawers that can add a new item to the drawer. Allow setting any item in the drawer as the launcher icon. CDE is otherwise an awful cumbersome piece of junk, but this feature is just so amazingly ergonomic and useful. You work for Redhat, they have to have an old copy of TriTeal CDE sitting around: boot it and learn it.
Another thing I'd like to see is a popup "run" menu, like KDE's alt-f2 or window's meta-R. See, I bound this in enlightenment to a key, but it tended to launch an instance of the launcher every time I hit the key, in a random place, then didn't give it focus. If I set E to give it focus, I would have to make it give *all* popups focus. Alt-F2 is probably the single reason I still run KDE (and now it's gotten buggy with focus problems that it's losing its worth).
IANAL, but AFAIK, you can't countersue for frivolous lawsuits unless they constitute a pattern of harrassment. One suit don't do it. You can challenge the suit as being without merit, but that's for the judge to decide, and to the defendant, every suit is without merit.
"have got" is one of those grammatical constructions that is becoming "acceptable use" but hasn't quite arrived at that status yet. Give it a couple more generations, and english teachers won't even blink at it.
The common its/it's confusion isn't likely to become acceptable for a long time, since the two spellings actually impart different meanings, as opposed to "have got" which is just redundant.
It's as bogus as a blue bear, but it still costs massive bucks to defend against frivolous suits. This is where the legal system needs reform: you lose, you pay.
> Patents should only be possible to claim on truly revolutionary methods of doing things.
Nonsense. Unique would be perfectly suffucient criteria. Uniqueness then tends to run into natural law, think of it it "God's Prior Art". You can't patent a mathematical formula for instance, but you can still patent the implementation of it in an algorithm, even if the language is nothing more than the specification of the formula. Then there's patenting of file formats, which smacks to me like patenting a shape of a drawing (something you could no doubt get copyright or trademark on, not a patent).
Patents on algorithms are based on the ludicrous concept of a physical machine that executes the algorithm, otherwise it's not patentable. Even though you have no IP rights over the language, runtime, or silicon that executes it, it doesn't matter because you've patented the hypothetical "database-o-tron" machine that runs the algorithm and has its file formats carved in hardware somehow. It's truly the twilight zone.
> They include the GoType keyboard, wireless modems, voice recorders, vibrating modules, extended IR transmitters (longer range than the dinky port on top), and even thermometers.
All of a sudden those pilot ads with the naked woman and her pilot make a certain amount of sense...
> I don't know about now, but five years ago, a state-of-the-art code for weather forcasting used spectral approximations (Fourier or Chebychev expansion functions) in the X- and Y-directions (Latitude and Longitude, say) and some high-order...
Dude... I think you just compressed an entire episode of Star Trek into six sentences.:)
Activision might have spirited away the Civ name, but the overall quality of AC over CTP is Sid's very sweet revenge. AC has very nice big icons for selecting units, CTP has this miniscule MFC-based interface at the very bottom of the screen. Diplomacy in AC actually consists of intelligent complete sentences, the chinese-menu quality of it is far less evident. Every single advance has a highly relevant and cogent quotation behind it. The effects of projects actually make sense. And just *listen* to AC, the music and sound effects are eerie, very sci-fi. This is what I call "production value", when a game feels like it all fits together. CTP on the other hand looks and feels slapped together, the interface especially so.
I think Sid sleeps well at night now. But the person to credit is really Brian Reynolds, who is responsible for most of Civ II and AC.
Uses disklabel instead of partitions. Basically all of freebsd's partitions are extended partitions. Pain in the ass reading them in linux (I just gave up trying to mount one), but overall a rather nice feature.
Had to hand-edit a config file of nasty little abbreviated names to configure the kernel for reinstall. That just wasn't terribly fun.
No configurator/wizard for ppp, had to pretty much set that up by hand. Didn't take me long, but it sure was a speedbump.
Had some funny ideas about its root device when booted, had to fiddle with the boot loader to get it working. Comes up so often it's a FAQ, but maybe it should install the boot loader with the right parameters to begin with. Just a minor problem tho, and probably what i get for installing it on a secondary slave IDE drive in the first place.
Ports are great, but it also has a package manager that looked adequate at any rate. KDE installed as a package, worked nicely out of the box.
Now I'm wondering, will I be able to use the GLX 3d driver for my TNT card, and can I get sblive support for freebsd? I don't play many games on linux, but I like the possibility, and I do play the occasional mods or mp3's through it. Oh I wasn't too thrilled how aha152x support is specifically left out of the later kernels BTW. I know it's a crappy card, but I don't see it interfering with anything.
Slashdot Mirror
> God does not play dice - Einstein
> Not only does God play dice, he sometimes throws them where they can't be seen. - Hawking
How about this one concerning parity violation:
Not only does God play dice, the dice are loaded.
(apologies for once again snipping quotes from Alpha Centauri)
Nanotech is not about making big things small. People watch Star Trek or read The Diamond Age and assume that nano is about little robots with little CPU's and little mechanical parts. And that just ain't so. We already have tiny molecular cutting devices: they're called enzymes. Yep, it's the squishy science of chemistry, where we deal with weird funny smelling liquids rather than neat shiny MicroMachines. The kind of nanolithography Feynmann talks about is rather similar to how we make chips now, and they're infinitely more flexible than a micro-encyclopedia. There's not nearly as much utility in putting all the manufacturing effort. toward creating single-purpose micro-devices when you can create a general-purpose one like a CPU.
> Even Moore himself admits that his law can't hold forever, only he is wise enough not to put a timetable on it.
Moore wasn't some digerati pundit making grandiose prognostications, unlike the current crop of wannabe net.prophets. He only predicted that a trend toward doubling the number of transistors on a chip every 18 months would continue for at least the next 10 years -- that was over 20 years ago.
Actually, didn't the move to RISC actually result in LESS transistors?
So why aren't you predicting the death of Quicktime 4 while you're at it? The interface for it is so bad, it got itself a whole column in the Interface Hall of Shame
> How about standard compliance? Mozilla is completely standard compliant, easing web developers a real lot on developing applications which works on all browsers. ... ...
> How about XUL?
> How about being built as a library...
> How about XML support?
> How about a more open architecture?
> How about a more open development model?...
How about a browser?
Why yes, just look at the thriving Chinese, North Korean, and Cuban software outfits. Oh and you can spare me the claptrap about how those aren't "true" communism/socialism/whatever, because central control in the name of "the people" has corrupted and failed every last time.
Shove it.
> THat's what happened with QT , and why to this DAY KDE is still considered non-free.
Except by RMS, who has personally called it free software.
stupid git.
> My thinking is that they died out for a reason, and they'll die out again
It seems most likely they were hunted to extinction. They'd die out again now because there simply isn't enough gene stock for them to survive. That and it's just not cold where elephants would tend to feed now.
Redistribute under what licenses? If a patch is submitted without specifying whether it's for the GPL'd or TGPL'd fork, does it automatically go into the TGPL'd fork?
And I say "fork", because that's exactly what will happen the instant the code is released. Someone will grab it, rename it, put it under GPL, and isolate all further improvements to the GPL'd version. Which leads one to wonder why you wouldn't just create the fork in the first place by simply dual-licensing it.
Better would be a license that creates a publisher/author relationship, where copyright of any modifications is automatically assigned to the originator. Some sort of "good faith" enforcement clause would be needed to ensure that redistribution rights are not terminated. The FSF has this written into their charter. If they decided to revoke the GPL for every GNU package and turn it into closed-source commercialware, the rights of the consumer (redistributor) under the GPL would be upheld because the FSF would be acting in bad faith. Well, probably: "good faith" is always a judgement call. And they can play with the license as they please, within the boundaries of good faith: they OWN every line of code in official GNU software thanks to the copyright assignment agreements they require.
The charter of the KDE Free Qt foundation would be one place to look for a charter spelled out explicitly. If Troll ever goes out of business, Qt becomes BSD licensed.
But all this said, licenses are not computer programs. They are subject to interpretation, loopholes, and flat-out nullification in the case of bad faith, violation of existing law, invalidity under statutes of contract law, or just the amount of money one can spend on lawyers. If you do not trust the license issuer, you cannot ever really trust the license.
I'm not a lawyer, but I play one on Slashdot.
Tip when using a thumb trackball: don't use your thumb. At least not all the time. You should be able to sorta glide all your fingers over it if you don't mash your hand down on it, which is worse to your wrists than using a mouse, which at least gives you lateral movement. Move the trackball close to you so you don't feel like you're reaching out to use it. It should feel like your hand just drops on the trackball, you should neither have to reach out for it nor mash your hand down on it.
As for your mouse, similar advice. Keep the heel of your hand off the thing and control it with your thumb and little finger. A light mouse with a slim profile helps, which pretty much rules out a MS Mouse. Of course I have huge hands so this is easy for me to do. But a light touch is the key. Now i just need to make better keymaps to rid myself of emacs pinky (no i will not use vi)
> Many spy thrillers have claimed there is another classification above Top Secret, without needing to shoot me, can you confirm or deny that? :)
SCI: Special Compartmentalized Intelligence.
It's not any more secret than Top Secret, but it has more stringent rules concerning its distribution. Having Top Secret clearance doesn't automatically clear you for SCI. It's the codified definition of "need to know". SCI information viewed on computers is done in a separate room on separate wiring where even the nearby water pipes are electrically isolated. Very secure stuff. But otherwise a well-known level of security.
The stuff more secret than that is the stuff that doesn't have a classification. It's the stuff the president or the director of the NSA or CIA says to another aide "don't tell this to anyone, ANYONE, got it?" In other words, pretty much all your extralegal stuff.
> It's up to window manager, not desktop environment to deal with global key bindings.
Grog no understand, Grog have desktop, but desktop no work way other desktop work. Grog want feature in desktop and Grog not care what part of desktop do what part.
Now if SawMill can do that, then it already sounds a lot more powerful than E, and I might suggest that gnome ship with some keybindings that do all that. Matching a window by its title does not seem to be the most elegant way of doing this. Something on the order of xtoolwait that gives me the windowid of the newly mapped window would be far more preferable.
> Is there a way to defend yourself from a (at least frivolous) lawsuit without spending thousands of dollars you can't get back even when you win?
If you're really confident that it's frivolous, you ask for it to be summarily thrown out of court. This usually doesn't happen, in which case, hope for a lawyer that does pro bono work or you're screwed. And you still get to miss a lot of work. Welcome to justice, American style.
I was using kde since the beta versions, and tried out gnome at the 1.0 release. After many cries of disgust and peals of ridiculing laughter, I swore off gnome forever. Recently I played musical distributions, going to debian then back to redhat 6.0 (I won't get into why) and in the process, installed gnome and not kde, as I wanted to work with KDE from CVS only.
It's not bad now. It's definitely a lot faster. E is even respectable looking, though I have plans to replace that with Sawmill. What with the million and a half language bindings for gtk+, I definitely can't hold the language card over its head. But I still have some issues with gnome. Mind you I have 'em with KDE, but this isn't a KDE forum at the moment:
DOCUMENTATION!!!: I can't say it enough. The state of GNOME's documentation, from gtk to everything else, is pathetic. The KDE classes themselves aren't always terrifically documented, but Qt's documentation is phenomenal. Make no mistake, GNOME has irretrievably lost potential developers because they couldn't get the documentation they needed. This has to be addressed. Raw API docs don't cut it, examples and introductory texts are essential.
If you're shamelessly ripping off items from other desktops, here's something to take from CDE: I'd like a panel setup that had several drawers of grouped functions ala CDE instead of one awful Cascading Menu From Hell known as the "Start Menu". Take this from CDE: create drawers that have an icon that when clicked, start the app, and an arrow *above* them that opens the drawer. Have drop targets in all the drawers that can add a new item to the drawer. Allow setting any item in the drawer as the launcher icon. CDE is otherwise an awful cumbersome piece of junk, but this feature is just so amazingly ergonomic and useful. You work for Redhat, they have to have an old copy of TriTeal CDE sitting around: boot it and learn it.
Another thing I'd like to see is a popup "run" menu, like KDE's alt-f2 or window's meta-R. See, I bound this in enlightenment to a key, but it tended to launch an instance of the launcher every time I hit the key, in a random place, then didn't give it focus. If I set E to give it focus, I would have to make it give *all* popups focus. Alt-F2 is probably the single reason I still run KDE (and now it's gotten buggy with focus problems that it's losing its worth).
IANAL, but AFAIK, you can't countersue for frivolous lawsuits unless they constitute a pattern of harrassment. One suit don't do it. You can challenge the suit as being without merit, but that's for the judge to decide, and to the defendant, every suit is without merit.
"have got" is one of those grammatical constructions that is becoming "acceptable use" but hasn't quite arrived at that status yet. Give it a couple more generations, and english teachers won't even blink at it.
The common its/it's confusion isn't likely to become acceptable for a long time, since the two spellings actually impart different meanings, as opposed to "have got" which is just redundant.
It's as bogus as a blue bear, but it still costs massive bucks to defend against frivolous suits. This is where the legal system needs reform: you lose, you pay.
> Patents should only be possible to claim on truly revolutionary methods of doing things.
Nonsense. Unique would be perfectly suffucient criteria. Uniqueness then tends to run into natural law, think of it it "God's Prior Art". You can't patent a mathematical formula for instance, but you can still patent the implementation of it in an algorithm, even if the language is nothing more than the specification of the formula. Then there's patenting of file formats, which smacks to me like patenting a shape of a drawing (something you could no doubt get copyright or trademark on, not a patent).
Patents on algorithms are based on the ludicrous concept of a physical machine that executes the algorithm, otherwise it's not patentable. Even though you have no IP rights over the language, runtime, or silicon that executes it, it doesn't matter because you've patented the hypothetical "database-o-tron" machine that runs the algorithm and has its file formats carved in hardware somehow. It's truly the twilight zone.
> They include the GoType keyboard, wireless modems, voice recorders, vibrating modules, extended IR transmitters (longer range than the dinky port on top), and even thermometers.
All of a sudden those pilot ads with the naked woman and her pilot make a certain amount of sense...
McNamara seemed more like a General Ripper to me, not a Doctor Strangelove...
> I don't know about now, but five years ago, a state-of-the-art code for weather forcasting used spectral approximations (Fourier or Chebychev expansion functions) in the X- and Y-directions (Latitude and Longitude, say) and some high-order ...
:)
Dude... I think you just compressed an entire episode of Star Trek into six sentences.
Activision might have spirited away the Civ name, but the overall quality of AC over CTP is Sid's very sweet revenge. AC has very nice big icons for selecting units, CTP has this miniscule MFC-based interface at the very bottom of the screen. Diplomacy in AC actually consists of intelligent complete sentences, the chinese-menu quality of it is far less evident. Every single advance has a highly relevant and cogent quotation behind it. The effects of projects actually make sense. And just *listen* to AC, the music and sound effects are eerie, very sci-fi. This is what I call "production value", when a game feels like it all fits together. CTP on the other hand looks and feels slapped together, the interface especially so.
I think Sid sleeps well at night now. But the person to credit is really Brian Reynolds, who is responsible for most of Civ II and AC.
Just a nitpick: Windows© is a complete misnomer. You never see the copyright symbol after a trademark, because they aren't at all the same thing. It's Windows®
Uses disklabel instead of partitions. Basically all of freebsd's partitions are extended partitions. Pain in the ass reading them in linux (I just gave up trying to mount one), but overall a rather nice feature.
Had to hand-edit a config file of nasty little abbreviated names to configure the kernel for reinstall. That just wasn't terribly fun.
No configurator/wizard for ppp, had to pretty much set that up by hand. Didn't take me long, but it sure was a speedbump.
Had some funny ideas about its root device when booted, had to fiddle with the boot loader to get it working. Comes up so often it's a FAQ, but maybe it should install the boot loader with the right parameters to begin with. Just a minor problem tho, and probably what i get for installing it on a secondary slave IDE drive in the first place.
Ports are great, but it also has a package manager that looked adequate at any rate. KDE installed as a package, worked nicely out of the box.
Now I'm wondering, will I be able to use the GLX 3d driver for my TNT card, and can I get sblive support for freebsd? I don't play many games on linux, but I like the possibility, and I do play the occasional mods or mp3's through it. Oh I wasn't too thrilled how aha152x support is specifically left out of the later kernels BTW. I know it's a crappy card, but I don't see it interfering with anything.
That's fading from the limelight.