The Star Trek project mentioned in the article is source-compatible, not binary-compatible, so every software manufacturer would have to recompile all of their code for the x86, and some would have problems, and many just wouldn't bother.
The thing is, there is already an x86 compiler on the MacOS that creates more efficient x86 code than most x86 native compilers. Furthermore, this compiler is already owned by nearly everyone who develops for the Mac, because it is included in CodeWarrior, the MacOS's leading development package. Given their history, Metrowerks (the makers of CodeWarrior) would likely make it very easy to recompile for MacOS on x86. The biggest stumbling block in doing this right now is that the Mac Toolbox is not availble on x86. Naturally, this problem would go away if MacOS was available on x86.
There are some cable companies who do forbid things like using NAT to support multiple machines from a single cable modem. I was an early adopter of one of the earliest cable modem systems in the U.S. (Cambridge, MA). At the time (and probably still, I don't live there anymore) MediaOne's contract specifically forbid using more than one computer through a single cable modem. On the other hand, MediaOne's own web site contained instructions on how to use NAT to support several machines over one cable modem, so it was clear to us (who set up just such a network the day we got the thing) that it was a restriction that MediaOne did not intend to enforce.
It seems likely to me that the reason it was there was to stop really flagrant abuses, such as giving net access to an entire apartment complex or something.
From Adbobe's web site, it appears that their suit may focus on a specific behavior of the tabbed palettes that Adobe uses, namely the ability to drag tabs in and out of the palette to make a new floating window out of it. This behavior strikes me as as being unique enough that adobe might win. They host a page with animated GIF examples of this behavior.
> I know these seem like all the wrong reasons to give money to charity,
Speaking of wrong reasons... let me say that I'm not really sure if I believe the following, but it occurred to me and won't get out of my head, so I thought I'd share it. It will probably offend some of you.
It seems to me that many (all? some? a few?) of the posters on Slashdot have more in common with the killers than the killed. Clearly, the killers had some severe problems and I find them repugnant, but at least some of the people killed were not random targets. Rather, they we perceived by the killers to have persecuted them in the past (true or not). It seems to me that a lot of Slashdotters know all to well the kind of persecution the killers went through.
With that in mind, it seems like this should have a bearing on the charity that is chosen. While it is tragic that these people were killed, it seems... I don't know... unfair (maybe?) to me that those who persecuted geeks should benefit. It seems like money from this book would be better off if used to help persecuted geeks, especially in those in high school.
I know this seems to fly in the face of thePsychotron's first point about making it "absolutely clear to everyone that we are in no way trying to justify the attackers actions". I would not perceive giving money to help those being persecuted as in any way justifying the attackers actions, but I concede that there would be those who would. I guess my question would be, should pleasing such people be a priority for Slashdot? Between helping geeks and pleasing the easily offended, I'll take geeks every time.
I wish I could suggest a charity which would help persecuted geeks, but I don't know of any.
Linux networking programs are conservatively constructed, without the high-level macro facilities that have enabled the recent Windows viruses to propagate so famously. This is not an inherent feature of Linux; it is simply a reflection of the differences between the two user bases and the resulting differences between the products that are successful in those markets.
This is a bit off the point, but the above snippet begs the question: is it possible to give a Linux application a "high-level macro facility" that approaches the power of those in Office, but in a secure way? How would the security aspects of such a system be handled?
Other unique cases (antiques and Legos)
on
Furry Cow Cases
·
· Score: 1
Pretty cool. There are two other cases made of unique materials that I know of, both Macs. One uses the case of an antique radio, the other uses Legos.
There are some proven technologies that Apple owns which could make their way to the iPalm. (Actually, when this product entered the rumormill months ago, it went by the name "iMate", if I remember correctly.)
1) The extremely good, on screen, handwriting recognition. The Newton allows you to write nearly anywhere on screen. By the time the 2100 came out, this recognition was really good. It even understood my hand-writing, which is no small task.
2) A better UI metaphor. The Newton defined its own interface guidelines, instead of trying to be like a desktop machine. Its core UI concepts are tons better than any other handheld I've seen. I'd be surprised if these are actually used on the iMate, but if they are, I would guess that Palm will (eventually) start using them on their own devices as well.
3) Better UI widgets. Even if they don't change the whole metaphor, the iMate could learn a lot from the Newton about tiny wigdets, tricks and interface tools which make using a Newton way easier than other palmtops.
4) StrongARM support. Again, this probably won't happen, but the Newton 2x00 series used the StrongARM processor, running at something like 160MHz. For a handheld of several years ago, that is quite a bit of processing power, especially since the batteries lasted for several weeks of average use (I think they were rated at 24hrs continuous use).
5) Speech. Apple ported MacInTalk to the Newton, so the Newton can read notes back to you. Apple has also been doing quite a bit with speech recognition lately, but that probably takes more juice than the iMate would have.
6) QuickTime. Eventually (probably not in the near future), palmtops will be powerful enough to playback video. Actually, the Newton could sort of do this, but not using a standard video format. (Anyone who saw the "Eat Millhouse" Simpsons clip on a Newton knows what I mean.) It would be in Apple's best interest to get QuickTime onto palmtops as soon as possible. Since the core of QuickTime is basically a compression engine, there may be some peices of it being used in the iMate.
7) TCP/IP. The Newton (eventually) got a fairly decent, small TCP/IP stack. I'm not sure how good Palm's is, but it is possible that Apple's is better.
8) Color. This is a bit of a stretch, but Apple has had years of experience in shifting black & white OS/hardware/APIs to color. As a Mac programmer during the transition from the Plus to the Mac II, I can tell you this was not very pleasant. Apple eventually got the hang of it, and I'm sure could give Palm some (much needed) advise about the best way to transition to full color.
One thing to keep in mind about all this is that Steve Jobs absolutely hated the Newton. This might have been because it was the pet project of the guy who fired him, but I remember seeing interview with people who said things like "Steve Jobs just doesn't get the Newton". So, looking to the Newton for inspiration may be the totally opposite of what Apple did.
Modern day humans are currently selected for resistance to disease. This is one big reason why Europeans conquered the Western Hemisphere so easily.
Biological Reality and Nonreality
on
The Regulon
·
· Score: 1
"Therefore," Gopnik quotes the economist as saying, "there must exist in the biological sphere a principle, which I will call the Regulon, which prevents this from happening."
This may be, but to say that "preditors consume most of them" is very wrong for a lot of populations. A culture of microbes in a test tube does not have preditors, yet will become exinct long before they fill the test tube (i.e. "cover the earth"). They will either run out of food or choke on their own waste.
Little stops the ebola virus, for example. They only reason it stops spreading is that it kills everyone in an area and can't transmit into other bodies (i.e. it runs out of "food").
This can be extended to the media, but this is foolish because the media has a very non-biological property: to succeed, it must be seen and/or heard. Thus, media will never be infinite because a) there are a limited number of eyeballs with limited time available to watch media and b) there is not infinite money to fund media which is not viewed.
In Triumph of the Nerds, you were referred to as "the Mozart of computer design". First, what to you think of this comparison? Second, who (if anyone) do you consider the "Mozart of software design"?
Through all his venting, there are two basic points he made which I can actually agree with.
1) The questions given to him were not stellar.
2) Harvard is definitely not afraid of being sued by anybody. If anyone thinks that the way to make Harvard cave in is to threaten them with a law suit, think again. Harvard has more money than God (Harvard could buy several stealth bombers) and some of the best lawyers on the planet. Any individuals which sued them would either be bought off or ground into the dirt.
This, in my opinion, is what it would take to force Microsoft to be a force of good (or, at least, a force of better) instead of a force of evil. I'm not much of a lawyer, so much of what I suggest probably is not justifiable given the facts of the case. Still, there is talk of a settlement, and if it looked like this, I'd be happy. I don't know the legal-eze of this, so I'm just talking intent here.
1) Microsoft can no longer create both operating systems and applications (including web browsers). The intent here is to force Microsoft to spin either its application or OS divisions into a separate company. I don't care if they both use the name Microsoft, as long as they are separate legally.
2) Microsoft's applications (including web browsers) cannot use _any_ knowledge about Microsoft's operating systems that is not publicly available. This would allow MS to, say, build a browser that can plug into the OS, but it also would allow someone else to plug in their own browser instead. To make this clear, I don't care if the app company and the OS company collaborate, as long as results of such collaboration are accessible to everyone. It could be argued that this would not be enforceable, but I'd wager that attempts to dodge this would be detected by competitors or anti-MS watchdogs, and they would either alert authorities or sue.
Preferably, this point would contain language to make violations of this point a criminal act, and thus settled in criminal court.
3) Any non-public APIs used by existing Microsoft applications (including web browsers) must be publicly disclosed immediately. This would probably be very damaging to MS (at least in the development community), because they do a lot of unpublished, internal stuff in their applications, especially Office. (An example to those of you who know Win32 programming: most of the controls used in Office are not HWNDs, which means they don't even use their own published APIs to, say, create a button. Try using Spy on a Word dialog to see what I mean.)
Microsoft would probably whine about these, but I don't think any of them would really cause too much trouble for them in the long run. It would, however, open up their system slightly, and I think would generate cleaner products, for both them and other Win32 developers.
None of these address Microsoft's strategy of buying competition and so on, so there is probably a need to address that kind of thing. I don't care as much about that practice, though, so I'll leave that as an exercise to others.
As a long time Apple watcher, I'm baffled that during this whole debacle (a period of about 4 working days), Apple's stock has jumped about 15%. I guess people just see earnings statements and figure they don't need to read anything else.
I know that Metrowerk's Win32 compiler on the Mac builds code that is, on average, about 20% faster than the same source compiled with DevStudio. How does the Linux version's output compare to mainstream Linux compilers?
Slashdot Mirror
The thing is, there is already an x86 compiler on the MacOS that creates more efficient x86 code than most x86 native compilers. Furthermore, this compiler is already owned by nearly everyone who develops for the Mac, because it is included in CodeWarrior, the MacOS's leading development package. Given their history, Metrowerks (the makers of CodeWarrior) would likely make it very easy to recompile for MacOS on x86. The biggest stumbling block in doing this right now is that the Mac Toolbox is not availble on x86. Naturally, this problem would go away if MacOS was available on x86.
There are some cable companies who do forbid things like using NAT to support multiple machines from a single cable modem. I was an early adopter of one of the earliest cable modem systems in the U.S. (Cambridge, MA). At the time (and probably still, I don't live there anymore) MediaOne's contract specifically forbid using more than one computer through a single cable modem. On the other hand, MediaOne's own web site contained instructions on how to use NAT to support several machines over one cable modem, so it was clear to us (who set up just such a network the day we got the thing) that it was a restriction that MediaOne did not intend to enforce.
It seems likely to me that the reason it was there was to stop really flagrant abuses, such as giving net access to an entire apartment complex or something.
From Adbobe's web site, it appears that their suit may focus on a specific behavior of the tabbed palettes that Adobe uses, namely the ability to drag tabs in and out of the palette to make a new floating window out of it. This behavior strikes me as as being unique enough that adobe might win. They host a page with animated GIF examples of this behavior.
Hey, no fair. The link worked when I previewed it.
Also, I noticed later later that ZDNet had the same story with better formatting.
> I know these seem like all the wrong reasons to give money to charity,
Speaking of wrong reasons... let me say that I'm not really sure if I believe the following, but it occurred to me and won't get out of my head, so I thought I'd share it. It will probably offend some of you.
It seems to me that many (all? some? a few?) of the posters on Slashdot have more in common with the killers than the killed. Clearly, the killers had some severe problems and I find them repugnant, but at least some of the people killed were not random targets. Rather, they we perceived by the killers to have persecuted them in the past (true or not). It seems to me that a lot of Slashdotters know all to well the kind of persecution the killers went through.
With that in mind, it seems like this should have a bearing on the charity that is chosen. While it is tragic that these people were killed, it seems... I don't know... unfair (maybe?) to me that those who persecuted geeks should benefit. It seems like money from this book would be better off if used to help persecuted geeks, especially in those in high school.
I know this seems to fly in the face of thePsychotron's first point about making it "absolutely clear to everyone that we are in no way trying to justify the attackers actions". I would not perceive giving money to help those being persecuted as in any way justifying the attackers actions, but I concede that there would be those who would. I guess my question would be, should pleasing such people be a priority for Slashdot? Between helping geeks and pleasing the easily offended, I'll take geeks every time.
I wish I could suggest a charity which would help persecuted geeks, but I don't know of any.
Linux networking programs are conservatively constructed, without the high-level macro facilities that have enabled the recent Windows viruses to propagate so famously. This is not an inherent feature of Linux; it is simply a reflection of the differences between the two user bases and the resulting differences between the products that are successful in those markets.
This is a bit off the point, but the above snippet begs the question: is it possible to give a Linux application a "high-level macro facility" that approaches the power of those in Office, but in a secure way? How would the security aspects of such a system be handled?
Pretty cool. There are two other cases made of unique materials that I know of, both Macs. One uses the case of an antique radio, the other uses Legos.
One of the best Mac games ever should be on every platform: Spaceward Ho! A definite classic.
There are some proven technologies that Apple owns which could make their way to the iPalm. (Actually, when this product entered the rumormill months ago, it went by the name "iMate", if I remember correctly.)
1) The extremely good, on screen, handwriting recognition. The Newton allows you to write nearly anywhere on screen. By the time the 2100 came out, this recognition was really good. It even understood my hand-writing, which is no small task.
2) A better UI metaphor. The Newton defined its own interface guidelines, instead of trying to be like a desktop machine. Its core UI concepts are tons better than any other handheld I've seen. I'd be surprised if these are actually used on the iMate, but if they are, I would guess that Palm will (eventually) start using them on their own devices as well.
3) Better UI widgets. Even if they don't change the whole metaphor, the iMate could learn a lot from the Newton about tiny wigdets, tricks and interface tools which make using a Newton way easier than other palmtops.
4) StrongARM support. Again, this probably won't happen, but the Newton 2x00 series used the StrongARM processor, running at something like 160MHz. For a handheld of several years ago, that is quite a bit of processing power, especially since the batteries lasted for several weeks of average use (I think they were rated at 24hrs continuous use).
5) Speech. Apple ported MacInTalk to the Newton, so the Newton can read notes back to you. Apple has also been doing quite a bit with speech recognition lately, but that probably takes more juice than the iMate would have.
6) QuickTime. Eventually (probably not in the near future), palmtops will be powerful enough to playback video. Actually, the Newton could sort of do this, but not using a standard video format. (Anyone who saw the "Eat Millhouse" Simpsons clip on a Newton knows what I mean.) It would be in Apple's best interest to get QuickTime onto palmtops as soon as possible. Since the core of QuickTime is basically a compression engine, there may be some peices of it being used in the iMate.
7) TCP/IP. The Newton (eventually) got a fairly decent, small TCP/IP stack. I'm not sure how good Palm's is, but it is possible that Apple's is better.
8) Color. This is a bit of a stretch, but Apple has had years of experience in shifting black & white OS/hardware/APIs to color. As a Mac programmer during the transition from the Plus to the Mac II, I can tell you this was not very pleasant. Apple eventually got the hang of it, and I'm sure could give Palm some (much needed) advise about the best way to transition to full color.
One thing to keep in mind about all this is that Steve Jobs absolutely hated the Newton. This might have been because it was the pet project of the guy who fired him, but I remember seeing interview with people who said things like "Steve Jobs just doesn't get the Newton". So, looking to the Newton for inspiration may be the totally opposite of what Apple did.
Modern day humans are currently selected for resistance to disease. This is one big reason why Europeans conquered the Western Hemisphere so easily.
This may be, but to say that "preditors consume most of them" is very wrong for a lot of populations. A culture of microbes in a test tube does not have preditors, yet will become exinct long before they fill the test tube (i.e. "cover the earth"). They will either run out of food or choke on their own waste.
Little stops the ebola virus, for example. They only reason it stops spreading is that it kills everyone in an area and can't transmit into other bodies (i.e. it runs out of "food").
This can be extended to the media, but this is foolish because the media has a very non-biological property: to succeed, it must be seen and/or heard. Thus, media will never be infinite because a) there are a limited number of eyeballs with limited time available to watch media and b) there is not infinite money to fund media which is not viewed.
In Triumph of the Nerds, you were referred to as "the Mozart of computer design". First, what to you think of this comparison? Second, who (if anyone) do you consider the "Mozart of software design"?
Through all his venting, there are two basic points he made which I can actually agree with.
1) The questions given to him were not stellar.
2) Harvard is definitely not afraid of being sued by anybody. If anyone thinks that the way to make Harvard cave in is to threaten them with a law suit, think again. Harvard has more money than God (Harvard could buy several stealth bombers) and some of the best lawyers on the planet. Any individuals which sued them would either be bought off or ground into the dirt.
This, in my opinion, is what it would take to force Microsoft to be a force of good (or, at least, a force of better) instead of a force of evil. I'm not much of a lawyer, so much of what I suggest probably is not justifiable given the facts of the case. Still, there is talk of a settlement, and if it looked like this, I'd be happy. I don't know the legal-eze of this, so I'm just talking intent here.
1) Microsoft can no longer create both operating systems and applications (including web browsers). The intent here is to force Microsoft to spin either its application or OS divisions into a separate company. I don't care if they both use the name Microsoft, as long as they are separate legally.
2) Microsoft's applications (including web browsers) cannot use _any_ knowledge about Microsoft's operating systems that is not publicly available. This would allow MS to, say, build a browser that can plug into the OS, but it also would allow someone else to plug in their own browser instead. To make this clear, I don't care if the app company and the OS company collaborate, as long as results of such collaboration are accessible to everyone. It could be argued that this would not be enforceable, but I'd wager that attempts to dodge this would be detected by competitors or anti-MS watchdogs, and they would either alert authorities or sue.
Preferably, this point would contain language to make violations of this point a criminal act, and thus settled in criminal court.
3) Any non-public APIs used by existing Microsoft applications (including web browsers) must be publicly disclosed immediately. This would probably be very damaging to MS (at least in the development community), because they do a lot of unpublished, internal stuff in their applications, especially Office. (An example to those of you who know Win32 programming: most of the controls used in Office are not HWNDs, which means they don't even use their own published APIs to, say, create a button. Try using Spy on a Word dialog to see what I mean.)
Microsoft would probably whine about these, but I don't think any of them would really cause too much trouble for them in the long run. It would, however, open up their system slightly, and I think would generate cleaner products, for both them and other Win32 developers.
None of these address Microsoft's strategy of buying competition and so on, so there is probably a need to address that kind of thing. I don't care as much about that practice, though, so I'll leave that as an exercise to others.
As a long time Apple watcher, I'm baffled that during this whole debacle (a period of about 4 working days), Apple's stock has jumped about 15%. I guess people just see earnings statements and figure they don't need to read anything else.
I know that Metrowerk's Win32 compiler on the Mac builds code that is, on average, about 20% faster than the same source compiled with DevStudio. How does the Linux version's output compare to mainstream Linux compilers?