To me, Kansas seems to be a great place for a trade show... Not because Kansas is such a great place to visit that people will come from all over the country to go to it, but because there are probably a lot of people in the area that can't make it to trade shows in other parts of the country.
If you're fighting a battle to use Linux in Kansas, I don't imagine many companies would send you to Las Vegas for a Linux trade show on their expense tab.
From what I've seen of this fiasco, the main problem was that they tried to scale a new trade show in a new area into something much bigger than the local Linux communities could support in its first year.
Start small, and build it into something greater, not the other way around.
As far as where to publish, I'm sure there are many online content publishers such as fatbrain.com's MightyWords.com that will allow you to publish your docs online and will sell them for you, while you collect royalties.
Now you just have to hope that people will find the sites.
If you decide to go with a SCSI adapter in addition to the onboard IDE, I'd recommend going with the Apple installed one or adding in an Adaptec board yourself.
When I got my G3, I opted for the cheapest SCSI card I could find, to drive my older scanner and an extra drive from my older PowerMac. (It's a Orange Micro connector). And I've had nothing but problems with it... It didn't support my drive even with the FWB drivers that came with the drive, and I've had intermittent problems with SCSI timeouts on the scanner. (Plus it's not compatible with LinuxPPC.)
Everything I've heard from other users is that the Adaptec card is worth the extra cash.
Personally, when I took my first compilers class I used the book Crafting a Compiler with C by Charles Fischer and Richard Leblanc Jr. IMHO, it was also a bit dated, but really hit on the basic concepts, such as LL and LALR parsing. I think it may even go as far as peephole optimizations, but my class didn't get that far into the text.
On the other hand, the recent incarnations of that class seem to be using Modern Compiler Implementation in {C/Java} by Andrew Appel, but I can't speak for it firsthand.
With a bit of hindsight, and the approaches I learned in my second Compilers class, I would say that real compiler frameworks are built using tools like lex, yacc, bison, or Eli. But still, you need the basics from handwritting a parser to really appreciate how these tools work.
I used to work in the Albany County Public Library in Laramie, Wyoming, where I was the sole computer support person. We ran into these issues in 1997, when we got internet access at public terminals. Suprisingly enough, it was very easy, and publically supported when we stated that we would NOT allow any censorware on our machines. Here's why:
The ACPL already had a policy that stated that children should be under responsible adult supervision. This includes their use of the internet terminals. Handy posters next to every terminal reiterated this fact.
As a member of the American Library Association, the ACPL had a responsibility to prevent any kind of censorship of library resources. See the ALA Code of ethics at http://www.ala.org/alaorg/oif/ethics.html it's right up there in item II.
Remember, the ALA has a long history of fighting censorship. They were even the chief plaintiff in opposing the Communications Decency Act.
And if you work for a public library, there are some ALA prepared resources including a Q&A about why the ALA opposed filtering software at http://www.ala.org/pio/cyber/cando.html
That's not necessarily true. None of the schools I've attended (or applied to) actually teach any Microsoft as part of the CS curriculum. The departments did have "MS Word for Meatheads" classes, but they were introductory level, and don't count as CS credits for the CS programs.
All the upperclass CS classes are taught on Unix machines, with a few lower ones taught on PC's with the Borland (Inprise?) family of compilers, because they are cheap and the hardware is already in all of the IT labs.
It's my impression that Clinton is only funding actual CS students, not those taking word processing and spreadsheet courses.
What I meant by my question is: can you get to the BSD layer directly? Can you run BSD apps on it? Can you run X windows apps? Or are you stuck with just the top layer that MacOS has glued on top of the BSD?
IIRC, this format was used to compress images for the first CD version of MS Encarta, one of the few that could afford the licensing fees (IMHO).
The patent holder is Michael Fielding Barnsley, with Iterated Systems, and I believe it is a US patent (or several). He describes the IFS (iterated fractal system) algorithms for image compression and decompression in his book (with Lyman Hurd) Fractal Image Compression copyright 1993. Definitely an interesting read, and it has references to all the published papers where the in depth details of the algorithms are proven.
Fractal images also have no inherent dimension and can be decompressed to any size, like vector formats. This is true, but also a misleading statement... when images are decomressed to resolutions beyond that of their original creation, image data is "created" based on the fractal characteristics of the attractor generated from the surrounding data in the original image. It will have all the same characteristics as the surroundings, but if there are new, completely different details that would not have been present in the original image, they will not be present in a "higher resolution" image... instead you see what could be reconstructed from the general descriptions of the system.
This is not to say that the image compression is not amazing, because it is amazing for some applications, but you wouldn't want to use it for some things, like medical/hospital images.
Or even a few proof readers. And people to consolidate them with all the others out there.
I'm comfortable in UNIX, but I was setting up my first linux system recently, and I attempted to configure sendmail to work with PPP. What did I do first? Look for a HOW-TO on the web. I found at least 10 HOW-TOs for sendmail/ppp. None of them were even remotely the same.
I picked the most readable, with the clearest steps and examples and plunged in. Worked great until I got to the part about rebuilding my sendmail configs.... all it said was "now rebuild the config file." The author apparently forgot that a newbie might not know that m4 is the appropriate tool to use. I guessed m4 based on the file extensions (and my UNIX background) but the m4 man page is awful, and it took some experimenting to figure out what to do.
Granted, I won't forget that bit of knowledge for quite a while, but I can see where someone will less general UNIX experience would have given up and settled for never sending mail from linux.
no matter what happens, grocers and retails are cleaning up
Right up to the point when there's a riot (whether or not anything really happens at Y2K), and all those grocers and retailers are looted, losing all the profits they made preY2K to clean up the mess.
If you stocked up on the right stuff, save it for another emergency, or just save on January grcoery bills.
All you have to do is save the jugs of water for the next time your water pipes are frozen, or the water main is shut off for construction.
Canned goods can be eaten anytime... you could probably even skip a month's shopping if you are well stocked for Y2K.
But if you bought army surplus MRE's or other such "survival food" may the gods pity your stomach. Or just leave them to your descendants so they can have them for Y3K - they'll still be "good".
Early in the film, Damon's Mr. Ripley states that he has three talents: lying, impersonating anyone, and forgery. But where did he learn such things?
Hmmm. The film broke right at this point when I saw it, and I assumed they just lost the bit where they explained it. I agree, now that I know it really is lacking... they show him practicing signatures, but only a few times before it's perfect. I'd like to know a little more about the character's background.
And overall, it may have been a better fit to call him "The Confused Mr. Ripley", because for all his talents, he spends most of the film being very confused about who he is and what he wants in life. With no apparent history of such confusion before he is whisked away to Italy. Again, some more background would have been useful in understanding the character, instead of spending most of the film thinking "What?!?!? Why'd he do that?
It was a good movie, but I'd have to limit it to 4 stars out of 5.
And if you've bought a new car recently, you hear such things as "You have to have the oil changed here, because the local grease monkey can't do it right without our company training" and "don't be tempted to give anyone a jumpstart - a power surge could destroy all your onboard computers causing thousands of dollars of damage".
Soon you won't even be able to lift the hood without the special company certified and distributed key. Reminds me of the special "case-cracking tool" Apple/Mac repair shops had to use to open the one piece beige toasters of yesteryear.
Changing the language is not necessarily a bad thing. There's a lot of work going on that adds or changes features in existing languages, trying to make it a better language for a specific purpose. The main thing is that you don't try and market it as the original language.
Make it downwardly compatible, if you can, so that any specification of the new language is a subset of the original, and the only problems you have are when you try and port something to the new.
And if you add new features to the language, make sure people know your code won't run with anyone else's compiler.
And in the case of java, no matter what the input language looks like, or what compiler you use, if you get Sun compatible bytecode out of it, what real difference does it make?
No kidding... just seeing the headline started the recovering of memories. The review is great, I haven't laughed that hard in a while. I vaugely remember liking the show when I saw it (I was only 5!!), but seeing those images I began to think about just how bad it might have really been.
No wonder Lucas wanted to kill this. I bet someone in ILM is forwarding the address of this article around as we speak.
It is possible (not necessarily likely, but possible) that there are lower energy states that electrons could occupy surrounding the nucleus... A less energetic state equals a "smaller" orbit. But you would think we should have seen them in nature... at least once in a while. It could be just a very, very, improbable energy state, but statistically, it should have been observed. And we should have seen it because of the hard UV being kicked off if it is as great a change as this article implies.
This is high school chemistry! I agree that this is not very plausible in the context of how we understand the world today, but there's a lot that they don't tell you in high school chemistry. For that matter there's a lot they don't tell you in introductory college chemistry either. You wouldn't believe the number of times a chemistry major is told "Well, what we teach in the intro classes isn't necessarily 100% accurate..."
This is hardly a grand unified theory. But the existance of "hydrino" electron states is still an important theory.
After reading the article, I'm still pretty sceptical about all of it. As the article says, most of molecular/atomic physics and chemistry uses the ground state of the hydrogen electron as a fundamental law... I suppose it is possible that there are less enegertic stable electron states within the "probability cloud" that forms around the nucleus, but that would also mean that we have to reconsider all of the equations that support current electron wave mechanics and kinetics. (I think this is covered by a derivative of the Schroedinger wave equation?)
There's little/no independent confirmation of the results, even though it should be fairly easy to detect the hard UV emissions caused by the electron state changes as hydrinos are formed. Unless there's some other form of energy dispersion occuring (kinetic?), based on the quantum nature of electron states, there should be a discrete change in energy, resulting in a discrete band of photon emission.
This is exciting if it's for real, but I'm gonna wait until there's a good body of independent evidence to support it before I buy into it.
This morning during my commute to work, the manager for the observatory at the University of Colorado was interviewed on the KBCO morning show. The observatory manager said that the moon should be approximately 7% brighter than average this solstice. A little bit brighter, but you probably won't notice unless you regularly track the moon.
I agree that sometimes the OO reuse-nuts go too far. "Why did you use this complicated design?" "Because we could fit in a Factory pattern *and* a Composition pattern!" There are times when a simple answer is better than the most abstracted, most reuseable answer... and there are many times you just don't have the time to code in all the things you could.
On the other hand, even when you're hacking out code, you can identify reuse and more OO like fragments that can save you time and effort.
The seemingly more realistic approach is a blend of OO/reuse where it helps, and when you have the time, but also to leave room for "do what you have to to make it work" kinds of coding.
In either case, you still need to have some form of design, not necessarily a strict one, but somethign to keep you on track. I've seen some really ugly code "evolve" from poorly designed systems.
Sun has a page featuring free download links for the linux distros that support the ultrasparc. Currently it lists downloads for Redhat 6.0, Caldera 2.2, and Debian 2.1 - but it links back to the homepage for each of the distros.
http://www.sun.com/software/linux/ultralinux
Hopefully, they'll host a mirror of their own, or sell cheap CD's in addition.
Slashdot Mirror
If you're fighting a battle to use Linux in Kansas, I don't imagine many companies would send you to Las Vegas for a Linux trade show on their expense tab.
From what I've seen of this fiasco, the main problem was that they tried to scale a new trade show in a new area into something much bigger than the local Linux communities could support in its first year.
Start small, and build it into something greater, not the other way around.
When I took a microprocessors class several years ago, we used free C compilers and Assemblers for the 68HC11, but unfortunately they were for MS-DOS.
Now you just have to hope that people will find the sites.
When I got my G3, I opted for the cheapest SCSI card I could find, to drive my older scanner and an extra drive from my older PowerMac. (It's a Orange Micro connector). And I've had nothing but problems with it... It didn't support my drive even with the FWB drivers that came with the drive, and I've had intermittent problems with SCSI timeouts on the scanner. (Plus it's not compatible with LinuxPPC.)
Everything I've heard from other users is that the Adaptec card is worth the extra cash.
On the other hand, the recent incarnations of that class seem to be using Modern Compiler Implementation in {C/Java} by Andrew Appel, but I can't speak for it firsthand.
With a bit of hindsight, and the approaches I learned in my second Compilers class, I would say that real compiler frameworks are built using tools like lex, yacc, bison, or Eli. But still, you need the basics from handwritting a parser to really appreciate how these tools work.
The ACPL already had a policy that stated that children should be under responsible adult supervision. This includes their use of the internet terminals. Handy posters next to every terminal reiterated this fact.
As a member of the American Library Association, the ACPL had a responsibility to prevent any kind of censorship of library resources. See the ALA Code of ethics at http://www.ala.org/alaorg/oif/ethics.html it's right up there in item II.
Remember, the ALA has a long history of fighting censorship. They were even the chief plaintiff in opposing the Communications Decency Act.
And if you work for a public library, there are some ALA prepared resources including a Q&A about why the ALA opposed filtering software at http://www.ala.org/pio/cyber/cando.html
That's not necessarily true. None of the schools I've attended (or applied to) actually teach any Microsoft as part of the CS curriculum. The departments did have "MS Word for Meatheads" classes, but they were introductory level, and don't count as CS credits for the CS programs.
All the upperclass CS classes are taught on Unix machines, with a few lower ones taught on PC's with the Borland (Inprise?) family of compilers, because they are cheap and the hardware is already in all of the IT labs.
It's my impression that Clinton is only funding actual CS students, not those taking word processing and spreadsheet courses.
Huh? Mac OS X is BSD.
What I meant by my question is: can you get to the BSD layer directly? Can you run BSD apps on it? Can you run X windows apps? Or are you stuck with just the top layer that MacOS has glued on top of the BSD?
I want to know what this will mean for LinuxPPC.
If MacOSX has access to a shell, and if it's possible to port *nix and X apps to it, then I no longer need to dual boot MacOS/LinuxPPC.
If MacOSX isn't that close to a real *nix, then will I still be able to dual boot into Linux?
Anybody actually played with it that cna answer my questions?
The patent holder is Michael Fielding Barnsley, with Iterated Systems, and I believe it is a US patent (or several). He describes the IFS (iterated fractal system) algorithms for image compression and decompression in his book (with Lyman Hurd) Fractal Image Compression copyright 1993. Definitely an interesting read, and it has references to all the published papers where the in depth details of the algorithms are proven.
Fractal images also have no inherent dimension and can be decompressed to any size, like vector formats.
This is true, but also a misleading statement... when images are decomressed to resolutions beyond that of their original creation, image data is "created" based on the fractal characteristics of the attractor generated from the surrounding data in the original image. It will have all the same characteristics as the surroundings, but if there are new, completely different details that would not have been present in the original image, they will not be present in a "higher resolution" image... instead you see what could be reconstructed from the general descriptions of the system.
This is not to say that the image compression is not amazing, because it is amazing for some applications, but you wouldn't want to use it for some things, like medical/hospital images.
Or even a few proof readers. And people to consolidate them with all the others out there.
I'm comfortable in UNIX, but I was setting up my first linux system recently, and I attempted to configure sendmail to work with PPP. What did I do first? Look for a HOW-TO on the web. I found at least 10 HOW-TOs for sendmail/ppp. None of them were even remotely the same.
I picked the most readable, with the clearest steps and examples and plunged in. Worked great until I got to the part about rebuilding my sendmail configs.... all it said was "now rebuild the config file." The author apparently forgot that a newbie might not know that m4 is the appropriate tool to use. I guessed m4 based on the file extensions (and my UNIX background) but the m4 man page is awful, and it took some experimenting to figure out what to do.
Granted, I won't forget that bit of knowledge for quite a while, but I can see where someone will less general UNIX experience would have given up and settled for never sending mail from linux.
Right up to the point when there's a riot (whether or not anything really happens at Y2K), and all those grocers and retailers are looted, losing all the profits they made preY2K to clean up the mess.
If you stocked up on the right stuff, save it for another emergency, or just save on January grcoery bills.
All you have to do is save the jugs of water for the next time your water pipes are frozen, or the water main is shut off for construction.
Canned goods can be eaten anytime... you could probably even skip a month's shopping if you are well stocked for Y2K.
But if you bought army surplus MRE's or other such "survival food" may the gods pity your stomach. Or just leave them to your descendants so they can have them for Y3K - they'll still be "good".
Provided you aren't smacked with a jet liner that has lost all instrumentation and is falling out of the sky in your general vicinity.
;-)
Despite the Y2K panic, I bet you'll have a few airborne traffic hazards to avoid.
Hmmm. The film broke right at this point when I saw it, and I assumed they just lost the bit where they explained it. I agree, now that I know it really is lacking... they show him practicing signatures, but only a few times before it's perfect. I'd like to know a little more about the character's background.
And overall, it may have been a better fit to call him "The Confused Mr. Ripley", because for all his talents, he spends most of the film being very confused about who he is and what he wants in life. With no apparent history of such confusion before he is whisked away to Italy. Again, some more background would have been useful in understanding the character, instead of spending most of the film thinking "What?!?!? Why'd he do that?
It was a good movie, but I'd have to limit it to 4 stars out of 5.
And if you've bought a new car recently, you hear such things as "You have to have the oil changed here, because the local grease monkey can't do it right without our company training" and "don't be tempted to give anyone a jumpstart - a power surge could destroy all your onboard computers causing thousands of dollars of damage".
Soon you won't even be able to lift the hood without the special company certified and distributed key. Reminds me of the special "case-cracking tool" Apple/Mac repair shops had to use to open the one piece beige toasters of yesteryear.
Changing the language is not necessarily a bad thing. There's a lot of work going on that adds or changes features in existing languages, trying to make it a better language for a specific purpose. The main thing is that you don't try and market it as the original language.
Make it downwardly compatible, if you can, so that any specification of the new language is a subset of the original, and the only problems you have are when you try and port something to the new.
And if you add new features to the language, make sure people know your code won't run with anyone else's compiler.
And in the case of java, no matter what the input language looks like, or what compiler you use, if you get Sun compatible bytecode out of it, what real difference does it make?
We'd definitely need to go get some stronger sunscreen.
Maybe SPF2000 UV blocker with lead shielding?
No kidding... just seeing the headline started the recovering of memories. The review is great, I haven't laughed that hard in a while. I vaugely remember liking the show when I saw it (I was only 5!!), but seeing those images I began to think about just how bad it might have really been.
No wonder Lucas wanted to kill this. I bet someone in ILM is forwarding the address of this article around as we speak.
This is high school chemistry!
I agree that this is not very plausible in the context of how we understand the world today, but there's a lot that they don't tell you in high school chemistry. For that matter there's a lot they don't tell you in introductory college chemistry either. You wouldn't believe the number of times a chemistry major is told "Well, what we teach in the intro classes isn't necessarily 100% accurate..."
This is hardly a grand unified theory. But the existance of "hydrino" electron states is still an important theory.
After reading the article, I'm still pretty sceptical about all of it. As the article says, most of molecular/atomic physics and chemistry uses the ground state of the hydrogen electron as a fundamental law... I suppose it is possible that there are less enegertic stable electron states within the "probability cloud" that forms around the nucleus, but that would also mean that we have to reconsider all of the equations that support current electron wave mechanics and kinetics. (I think this is covered by a derivative of the Schroedinger wave equation?)
There's little/no independent confirmation of the results, even though it should be fairly easy to detect the hard UV emissions caused by the electron state changes as hydrinos are formed. Unless there's some other form of energy dispersion occuring (kinetic?), based on the quantum nature of electron states, there should be a discrete change in energy, resulting in a discrete band of photon emission.
This is exciting if it's for real, but I'm gonna wait until there's a good body of independent evidence to support it before I buy into it.
This morning during my commute to work, the manager for the observatory at the University of Colorado was interviewed on the KBCO morning show. The observatory manager said that the moon should be approximately 7% brighter than average this solstice. A little bit brighter, but you probably won't notice unless you regularly track the moon.
I agree that sometimes the OO reuse-nuts go too far. "Why did you use this complicated design?" "Because we could fit in a Factory pattern *and* a Composition pattern!" There are times when a simple answer is better than the most abstracted, most reuseable answer... and there are many times you just don't have the time to code in all the things you could.
On the other hand, even when you're hacking out code, you can identify reuse and more OO like fragments that can save you time and effort.
The seemingly more realistic approach is a blend of OO/reuse where it helps, and when you have the time, but also to leave room for "do what you have to to make it work" kinds of coding.
In either case, you still need to have some form of design, not necessarily a strict one, but somethign to keep you on track. I've seen some really ugly code "evolve" from poorly designed systems.
Sun has a page featuring free download links for the linux distros that support the ultrasparc. Currently it lists downloads for Redhat 6.0, Caldera 2.2, and Debian 2.1 - but it links back to the homepage for each of the distros.
http://www.sun.com/software/linux/ultralinux
Hopefully, they'll host a mirror of their own, or sell cheap CD's in addition.