Eh? A computer that could pass the Turing Test would surely be able to respond to commands, so that once you have useful voice recognition, just pipe its output to the Turing Test passer...no need to wait another fifty years!
I'm hoping that we'll advance much faster than you think (see discussions of Vinge's Singularity). (Heck, I just cut the time down by a third, just by using the Unix tools philosophy.:-)
That thanks to Okuda's refusal to define or use a consistent mapping between Klingon characters and sounds, which makes no sense whatsoever, we're stuck with piqaD for I/O.
..do I want a medium that has fair use prevent--er, "rights management"--built in, and has all the features of videocassettes (no random access, drop-outs, stretches, Rube Goldberg transport and heads...)?
There's already some of that...looking at the current version of the software on my LinkSys box, it has an option that says "tell the outside world that your MAC address is [fill in blank]," which I would presumably set to the MAC address of the Ethernet card I first connected to the cable modem with.
Maybe Lucas is doing it backwards this time. After all, it's Episode I that has the smarmy character thrown in to appeal to children, blatant "let's throw this in so we can base a video game on it" stuff, etc. So...we can expects a "cracking second film" and then a cheesy third. (After all, it will in the SW chronology be closest to the cheesy first film, right?)
Agreed about C--but OTOH, Intel's compiler suffers from the same source language limitations, so if there is a difference in performance, one can't attribute it to conservative optimization forced by possible aliasing.
They'd better make pencils and paper illegal...I might write down copyrighted material with them, and piracy is the main factor for consideration above all other, right?
The reason the usual techniques are band limited is the problem of aliasing (as we all can remember from watching the wagon wheels go backwards in old Westerns). The limitation of our ears makes uniform sampling techniques feasible for digital audio, but that doesn't mean that the new theory isn't applicable to digital audio.
Feh. The standard is the current HTML specification, not IE. Do let me know when you can distribute to me a version of IE that will run on OS/2 or Linux, though.
Well...I personally have three CoCo 3s, and I wouldn't mind a flash card to move data among them. Mark Marlette of Cloud-9 is working on the ultimate CoCo 3 add-on card, to support 2 Mbytes of RAM, flash, ethernet, two good serial ports, clock, an AT keyboard interface, SCSI, IDE, and MIDI. (I may have overlooked some things.) He's already done SCSI, the 2 Mbyte RAM hack, the clock, and the AT keyboard interface on other addons in the past, so I'm confident he can do it. IDE's already been done for the CoCo at least a couple of times--once as a Glenside Color Computer Club project, and once by a fellow who has done an amazing number of CoCo hardware projects on his own.
I have deposits down on three of the cards, one per CoCo of course. Now, if only someone would put one of the 6809, or better still, 6309, reimplementations on a FPGA or ASIC...
...they can always shake down the citizens for more money or pile on some more deficits rather than fold. If people would come and throw you in jail and take all your money if you didn't buy a BeBox, Be would have done a lot better. Wouldn't have been right, but they'd have done a lot better.
Check out a recent thread on/. about a Borland EULA that includes a clause claiming the right to look at your stuff. How many people, I wonder, actually read EULAs?
Not a dumb question at all. As others have pointed out, if you user your own (software) tools, then you're that much more motivated to fix them when they break. Also, a compiler is a Big Program, and likely to exercise itself in ways that small sample programs from C# for Dummies and the like won't. The wider the variety of programs you can test a compiler with, the better--another canonical stress test for compilers is the output of program generators, which will do things humans typically won't. (Canonical example: parser generators like yacc, which churns out basically a for loop around a HUGE switch statement and a really big initialized array.)
Fine, they'll consider OSS for new stuff. Remember the UK's existing "e-government" that is only accessible via Windows? The policy should mandate the revision of all existing facilities to use open standards.
I miss the TNT Godzillathons (and the clips they'd run over BOC before commercials), and especially the wretched "Hercathons" they would run around New Years--Edith Hamilton never told us about Hercules's trip to Peru (Hercules Versus the Sons of the Sun), now did she?:-)
I agree with you about the 6502, but not about the Z-80. IMHO the best 8/16 bit CPU is the 6809, or perhaps the Hitachi 6309 (which extended the 6809 architecture nicely...too bad folks in the US didn't know about it for years). The 63C09 has been run at 5 MHz, and there's a company called INICORE that counts among its products iniCPU, a logic design burnable into FPGA or ASIC that implements the 6809 instruction set--the web page cites its performance running at 40 MHz.
Sounds like Linus is in accord with Dick Gabriel's "Worse is Better" essay, or rather the "New Jersey" school of design. The "worse" package gets out quickly to a place where "natural [user/hacker] selection" can work on it.
Well, almost...@home tells you to use just "mail" for your mail server, so unless you have found out what your mail server's name really is, another DNS won't map that appropriately and you won't be able to get your email. (OTOH, I'm in Iowa, and still can't get my email anyway, even now that Mediacom has the DNS going again. Supposedly that will be back this morning; I hope it is, because I've been laid off, and this is the email address I put on the resumés I've submitted at the various job search sites...)
I agree, it will eventually get cheap enough--but there may be a point before that when someone will decide that there are sufficient other advantages to outweigh some additional cost. If nothing else, CG characters are unconstrained by the laws of physics and biology. (Probably better to let the readers' imagination think of the possibilities...)
The "rights" you're referring to that the FSF wants to increase are bogus--they are so-called "positive rights," which are really demands on others' resources. I limit the rights of many individuals by insisting that they not steal my posessions--does that make me evil?
Slashdot Mirror
I'm hoping that we'll advance much faster than you think (see discussions of Vinge's Singularity). (Heck, I just cut the time down by a third, just by using the Unix tools philosophy. :-)
That thanks to Okuda's refusal to define or use a consistent mapping between Klingon characters and sounds, which makes no sense whatsoever, we're stuck with piqaD for I/O.
No.
There's already some of that...looking at the current version of the software on my LinkSys box, it has an option that says "tell the outside world that your MAC address is [fill in blank]," which I would presumably set to the MAC address of the Ethernet card I first connected to the cable modem with.
Maybe Lucas is doing it backwards this time. After all, it's Episode I that has the smarmy character thrown in to appeal to children, blatant "let's throw this in so we can base a video game on it" stuff, etc. So...we can expects a "cracking second film" and then a cheesy third. (After all, it will in the SW chronology be closest to the cheesy first film, right?)
Agreed about C--but OTOH, Intel's compiler suffers from the same source language limitations, so if there is a difference in performance, one can't attribute it to conservative optimization forced by possible aliasing.
They'd better make pencils and paper illegal...I might write down copyrighted material with them, and piracy is the main factor for consideration above all other, right?
...that's about the size of the typical introductory C++ book these days, isn't it?
Hmmm. Microsoft's so-called "Freedom to Innovate Network" page has a "How You Can Help" link (Passport registration required)...maybe that's the web page in question?
The reason the usual techniques are band limited is the problem of aliasing (as we all can remember from watching the wagon wheels go backwards in old Westerns). The limitation of our ears makes uniform sampling techniques feasible for digital audio, but that doesn't mean that the new theory isn't applicable to digital audio.
Feh. The standard is the current HTML specification, not IE. Do let me know when you can distribute to me a version of IE that will run on OS/2 or Linux, though.
Well...I personally have three CoCo 3s, and I wouldn't mind a flash card to move data among them. Mark Marlette of Cloud-9 is working on the ultimate CoCo 3 add-on card, to support 2 Mbytes of RAM, flash, ethernet, two good serial ports, clock, an AT keyboard interface, SCSI, IDE, and MIDI. (I may have overlooked some things.) He's already done SCSI, the 2 Mbyte RAM hack, the clock, and the AT keyboard interface on other addons in the past, so I'm confident he can do it. IDE's already been done for the CoCo at least a couple of times--once as a Glenside Color Computer Club project, and once by a fellow who has done an amazing number of CoCo hardware projects on his own.
I have deposits down on three of the cards, one per CoCo of course. Now, if only someone would put one of the 6809, or better still, 6309, reimplementations on a FPGA or ASIC...
...they can always shake down the citizens for more money or pile on some more deficits rather than fold. If people would come and throw you in jail and take all your money if you didn't buy a BeBox, Be would have done a lot better. Wouldn't have been right, but they'd have done a lot better.
Check out a recent thread on /. about a Borland EULA that includes a clause claiming the right to look at your stuff. How many people, I wonder, actually read EULAs?
For that matter, I can't help wondering whether MS would fund the development of Linux viruses.
Not a dumb question at all. As others have pointed out, if you user your own (software) tools, then you're that much more motivated to fix them when they break. Also, a compiler is a Big Program, and likely to exercise itself in ways that small sample programs from C# for Dummies and the like won't. The wider the variety of programs you can test a compiler with, the better--another canonical stress test for compilers is the output of program generators, which will do things humans typically won't. (Canonical example: parser generators like yacc, which churns out basically a for loop around a HUGE switch statement and a really big initialized array.)
"Gamera will save us--he's the children's friend!" -- from Gamera vs. Guiron, in which Gamera fights an Exacto knife with legs from outer space
Fine, they'll consider OSS for new stuff. Remember the UK's existing "e-government" that is only accessible via Windows? The policy should mandate the revision of all existing facilities to use open standards.
I miss the TNT Godzillathons (and the clips they'd run over BOC before commercials), and especially the wretched "Hercathons" they would run around New Years--Edith Hamilton never told us about Hercules's trip to Peru (Hercules Versus the Sons of the Sun), now did she? :-)
I agree with you about the 6502, but not about the Z-80. IMHO the best 8/16 bit CPU is the 6809, or perhaps the Hitachi 6309 (which extended the 6809 architecture nicely...too bad folks in the US didn't know about it for years). The 63C09 has been run at 5 MHz, and there's a company called INICORE that counts among its products iniCPU, a logic design burnable into FPGA or ASIC that implements the 6809 instruction set--the web page cites its performance running at 40 MHz.
And you believe Gates's statement because...?
Sounds like Linus is in accord with Dick Gabriel's "Worse is Better" essay, or rather the "New Jersey" school of design. The "worse" package gets out quickly to a place where "natural [user/hacker] selection" can work on it.
Well, almost...@home tells you to use just "mail" for your mail server, so unless you have found out what your mail server's name really is, another DNS won't map that appropriately and you won't be able to get your email. (OTOH, I'm in Iowa, and still can't get my email anyway, even now that Mediacom has the DNS going again. Supposedly that will be back this morning; I hope it is, because I've been laid off, and this is the email address I put on the resumés I've submitted at the various job search sites...)
I agree, it will eventually get cheap enough--but there may be a point before that when someone will decide that there are sufficient other advantages to outweigh some additional cost. If nothing else, CG characters are unconstrained by the laws of physics and biology. (Probably better to let the readers' imagination think of the possibilities...)
The "rights" you're referring to that the FSF wants to increase are bogus--they are so-called "positive rights," which are really demands on others' resources. I limit the rights of many individuals by insisting that they not steal my posessions--does that make me evil?