I'm not proposing it, it's already happening. Even Palm is going to dump the 68k for the ARM, and they are almost certainly going to go with a multitasking 32bit kernel for that. It's what the market demands, and Palm is, if anything, late to the game.
Ah yes. And that's exactly why most businessmen will buy this PDA.
If they want custom applications developed cheaply, that would be a reasonable choice.
Why go for simplicity and elegance in the Palm
The Palm is simple and elegant if you don't have to program it, if you stick to its standard applications, and if you are happy with exactly the kind of screen and UI it gives you. If you go beyond its basic four functions, it become increasingly messy and limited.
when you can run a process-check on something that shouldn't even have processes.
Multitasking makes a lot of sense to me on a device that may, in fact, have to interact with lots of different I/O channels (user, IR, network, modem, wireless, etc.).
There isn't much money in four function PDAs anymore; Palm now needs to go for vertical applications and portable media. For that, they need something more powerful, with a much better screen, and something that has standard APIs. Java+Linux and Java+EPOC are both reasonable choices.
PalmOS is pretty much obsolete; even Palm knows that. The next version may still be called "PalmOS", but it will be something entirely different.
I signed up for their developer program. I got back a message announcing the developer deal, so tried ordering at the developer price of $179. No luck: their site only offers retail pricing. Are these things still available at developer pricing?
If only PalmOS could take advantage of a 32bit processor, protected mode, and a larger display. Unfortunately, PalmOS can't, which is why people are looking at alternatives in the first place.
You can probably run PalmOS in emulation on various PDAs. In fact, I suspect that's what we are going to see in a year or two from Palm: an entirely new OS that runs the old stuff on an emulator.
The existence of "ps" on the Agenda is merely an indication that it runs standard software and standard utilities. That's a good thing. In day-to-day usage, of course, you shouldn't have to use it.
In contrast, the Palm Pilot requires you to write applications to its own limited and proprietary APIs, memory model, and threading model. The Palm Pilot runs its applications reasonably well, but someone had to invest a lot of time and effort in that; doing anything more with it is an uphill struggle.
The limitations of the PalmOS and the messy programming model are a consequence of trying to deliver a nice system on limited hardware. Those constraints don't apply anymore, and the Palm Pilot is now pretty much an anachronism.
I have a Palm Pilot. It's a reliable workhorse for its very limited set of applications. I'll gladly toss it out when I get a decent, reliable UNIX-based PDA. The Agenda could be it.
That is one of the many reasons why centralized web-based services are not such a good idea. In fact, we have a perfectly good service for distributing information widely without the ability for anyone to identify readers: USENET. For better or for worse (I think for worse, actually), it is even permanently archived and searchable now. And USENET offers a choice of jurisdiction of where the identity of a poster is protected.
What features would I like? I think that's pretty easy to answer by reference to existing languages. Java is pretty nice for many of the applications I have. For some applications, I would like to have the additional facilities found in Sather. I also like OCAML. And for interactive and GUI applications, I like Smalltalk.
The problem isn't lack of languages, it's lack of commitment to working on existing systems.
For example, Guido preferred going off and doing his own thing with Python rather than create a free Smalltalk implementation. The result is that we now have Python, a language that's a little bit nicer to use than Smalltalk but whose implementation is a whole lot worse than that of existing Smalltallk implementations. If people keep starting from scratch, no wonder that implementations hardly ever get really good.
It's kind of ironic that LX is tied to a system called "Mozart", which is also the name of another new, experimental language and environment.
I think the PC case is a pretty good place for a power supply. But many peripherals could be powered through their data cables; both USB and FireWire offer that choice. Unfortunately, making anything low power tends to make it more expensive, so the cheap stuff will likely continue to come with its own separate power supplies.
I don't see any information on that site about shielding. Computers create a lot of radio frequency emissions. That's why people stick them into these unsightly metal cases (or, in the case of the iMac, try to design around RF emissions in some other way). Computers also need FCC testing for radio frequency emissions, which is why computers you buy at a store usually have an FCC sticker on the back.
Taking a PC designed for a shielded metal case and sticking it into an unshielded plastic case seems like a sure way of creating interference for your neighbors: radio, television, amateur radio, portable telephones, medical devices, wireless networking, cell phones, etc. It's kind of like turning up your stereo all the way, and I don't think that's particularly nice. If people track you down, FCC regulations let them demand that you shut down your computer because it creates harmful interference to licensed devices.
So, save yourself some time and money, be nice to your neighbors, and leave that motherboard in the shielded case it was designed for. If you want a clear PC, buy one that was designed that way from the start, like an iMac or a G4 Cube.
Intel is on the ropes and only by tuning small parts of the core OS to run at the "marketed speed" can they keep fooling the public into thinking it's a faster chip.
The P4 probably has many flaws, but for compute-intensive applications, the fastest P4 is significantly faster than the fastest Athlon. If you haven't benchmarked it yourself, check out the benchmarks at spec.org.
I agree that Intel is in trouble with their Itanium. It will probably run GCC and Sun JDK reasonably well (and VC++ as well as VC++ ever runs), but it will be nearly impossible for mere mortals to develop good compiler backends for other compilers.
Intel should pray that their transition is as smooth as Apple's 68k->PPC changeover; I don't think it will be. Apple really made that work as well as possible. In fact, Apple seems to be pretty good at that: the MacOS9->MacOSX transition looks like it's going OK as well.
Do you like to burn books, too? And perhaps we should destroy all the old microfilm archives of back-issues of newspapers?
No, actually, I generally just try to respect the intent of the author when feasible. Book authors and newspaper writers usually anticipate and intend for their writings to get archived for the long term, and they represent writings that people spent a long time creating for a public audience. But authors of USENET articles thought they were participating in an informal, short-lived discussion.
I participated for many years under my real name on USENET. The assumption was that, while people did archive parts of USENET from the beginning, nobody would republish the stuff widely: doing so would have been both impractical and legally questionable at the time.
Once it became clear that USENET was increasingly becoming permanently fixed and searchable, I stopped participating under my real name. I never flamed on it or participated in particularly controversial subjects, but I still didn't want to have to deal with the possibility of being quoted out of context years later.
While anonymity has many undesirable features, it is the second-best choice if you can't have informal, short-lived discussions (this is, incidentally, why I'm not using my real name on Slashdot). For me, what killed USENET was not anonymity but its permanent archiving.
I think something similar has happened in politics: since everything is getting recorded and republished and analyzed word-for-word, politicians can't engage in thoughtful debate anymore in public for fear of offending someone or getting attacked on out-of-context quotes. Instead, every political message has to be carefully crafted and rehearsed; no extraneous utterance or debate is possible.
It's questionable to me whether making old USENET postings is either legal or desirable.
From the legal point of view, the posters of the original messages hold the copyrights. While USENET always ended up on some backup tapes somewhere, arguably, the original posters didn't give consent to having their comments archived and made accessible forever by some big search engine company.
Legal issues aside, I think it is also a bad idea to make content from an informal discussion forum easily available for eternity. For participants in such forums to have to worry about whether something they say today may be used against them decades from now kills participation by thoughtful people. Of course, these days, many people try to protect themselves against that in part by using pseudonyms, but in 1995, that wasn't so common yet.
So, I don't applaud Google at all for putting this content back on-line. Google should not change the rules for content created years ago. And even today, I think it would be better to return USENET to an ephemeral, informal medium as much as possible.
I hope Google will go back to making only the last six months (or so) of content available. I believe that is what DejaNews ended up doing, and it's the right thing to do.
I took this as a hint that the real problem is that one of his coauthors is from Xerox and that a corporation is less willing than a university to expose itself to a lawsuit in the name of academic freedom.
I don't see that as an obstacle. The Xerox author could have removed himself from the paper (and instead been moved into the acknowledgement section) and Xerox could have formally protested the publication. Felten could then have gone ahead and published it anyway. Formally, he might have been guilty of copyright violation, but Xerox is under no obligation to pusue that.
Technically, it was almost certain from the beginning that all these schemes could be broken. And once the preprint was published on their web site, the cat was out of the bag.
The whole participation of the Princeton group in the SDMI effort was a political statement from the beginning. The decision to withdraw the paper is likewise a political and strategic decision with no technical significance. Let's just hope those guys know what they are doing when it comes to politics and strategy, because, so far, it isn't clear to me where they are going with this.
Load balancing via IP routing tricks is kind of nice. Mosix goes one step further and allows live processes to migrate across a cluster. Experimental add-ons also will do socket migration and havedistributed file system support. I think that's the kind of approach to clustering you are going to see in the long run.
Why bother dealing with these people? They have trouble understanding that the price of content goes down as distribution costs go down. They are fighting a losing battle against companies that do understand economics. The content that gets distributed with fewer restrictions at lower production costs will win out over high-priced content from a bunch of media czars and over-the-hill writers, composers, and pop stars.
They also seem to have trouble understanding that watermarking is not technically feasible. It won't take some really smart guys from Princeton to break this or future systems. Given Chiariglione's inelegant and messy technical track record, I doubt they are going to get a technical clue any time soon either.
Let them add poor watermarks to poor content and create players with all sorts of limitations. In the long run, it's only going to hurt their business. Dealing with these people is a waste of time in my opinion.
But instead, he insulted TrollTech and KDE yet again,
Insulted? In what way? Can you provide details? How can you "insult" a company anyway?
He should have praised [TrollTech] diplomatically for their decision, as this was a victory for the FSF and the GPL.
You mean he should have praised TrollTech for their decision not to go out of business? TrollTech's hand was forced: without KDE, Qt would be just one of a dozen obscure commercial toolkits. And without changing Qt to a GPL license, KDE would have had to come to an end.
What this whole thing shows is that RMS's care and stubbornness in designing the GPL has paid off. Without it, TrollTech would have been able to hijack KDE, a major high-quality open-source project, for their commercial purposes.
As for TrollTech, I'd continue to view them with suspicion. Their vision seems to be that any commercial user of open source software should pay some commercial company, and they seem eager to want to do whatever they can to make Qt an integral part of whereever Linux goes. That is not the kind of vision many supporters of open source have, and pragmatically, it is also very bad for the long term viability of open source software.
Eazel already got the benefit from making their software free: the chance of having it adopted. Sure, their business model may not have worked out, so that wasn't enough. But without it, how much chance of widespread use do you think yet another closed source C-based application suite would have had?
Unfortunately, Guido wasn't asked, and didn't provide answers, about efficient compilation for Python into standalone executables. There are actually efforts underway, but it would be nice to find out more about his thinking on them.
In principle, Python could be compiled like C, C++, Java, Fortran, Eiffel, etc. into standalone executables. That would both simplify deployment and (if done well) improve speed. Previous attempts at this have not been particularly popular or successful, however.
The problem with JavaScript is not lack of expressive power; it has plenty of that. It is poor scoping rules, poor error handling, serious limitations on available APIs, and subtle differences among implementations. If TIBET manages to work around some of those problems, that would be great. But I have my doubts. I guess we'll have to see.
But a prototype-based scripting language with JavaScript semantics isn't very Smalltalk-like. Smalltalk has much better development tools, better APIs, and better error handling. It can also be compiled more efficiently.
The extreme programming (XP) methodology grew out of a Smalltalk project 3-4 years ago. The founder of extreme programming, Kent Beck, advocates Smalltalk as the most productive language for XP, as does Ron Jeffries, the author of the Extreme Programming Installed book.
Like so many things that recently have made a splash (extreme programming, refactoring, patterns, etc.), this is really just giving a name to programming practices and ideas that have been around for two decades. Smalltalk was one of the languages where such ideas originated (Lisp was another major one), they just didn't originate recently.
(There have also been some refactoring tools written for Java, but the nature of the Java language will make it difficult to create a tool as powerful as the Refactoring Browser for Smalltalk.)
Quite to the contrary. While Java isn't as convenient to program in as Smalltalk, the additional type information that Java source code contains lets you do more kinds of refactoring safely. In fact, one of the reasons refactoring has not been such a big deal in the C++/Java world is because static type checking in those languages already helps you with a lot of refactorings. For example, in Java, if you change an interface or move instance variables around, the type checker will point out every single place that you need to change. If you use a good IDE, you can breeze through those places and a refactoring only takes a single edit, compile, edit cycle. You still need to make the changes by hand, but that's pretty quick, actually faster than any of the refactoring browsers I have used in many cases.
(contained objects don't know about their containers unless you create an explicit link back to the container. That makes instantiation in context more complicated.)
That's what weak dictionaries are for: they allow you to create such bidirectional associations easily.
Note that Blogger output is on your personal site unless you're using blogspot.com.
That doesn't change things much (yes, I had actually looked at the site). If the service goes away and you don't have any alternatives, you are still stuck. And those alternatives will only exist if there is demand for people to run web applications at their own hosting services. (It also raises all sorts of concerns about security and your terms of service with your web hosting company.)
The kind of freedom you're talking about comes at a price. Not everybody wants to be a programmer.
You don't have to be a programmer to install a web page editor on your machine or to install a web application on a well-run hosting service.
Note that by posting to Slashdot, you're creating content that allows a business you don't own to make money
I have no problem with people making money, Blogger, Trellix, or otherwise.
Slashdot Mirror
I'm not proposing it, it's already happening. Even Palm is going to dump the 68k for the ARM, and they are almost certainly going to go with a multitasking 32bit kernel for that. It's what the market demands, and Palm is, if anything, late to the game.
If they want custom applications developed cheaply, that would be a reasonable choice.
Why go for simplicity and elegance in the Palm
The Palm is simple and elegant if you don't have to program it, if you stick to its standard applications, and if you are happy with exactly the kind of screen and UI it gives you. If you go beyond its basic four functions, it become increasingly messy and limited.
when you can run a process-check on something that shouldn't even have processes.
Multitasking makes a lot of sense to me on a device that may, in fact, have to interact with lots of different I/O channels (user, IR, network, modem, wireless, etc.).
There isn't much money in four function PDAs anymore; Palm now needs to go for vertical applications and portable media. For that, they need something more powerful, with a much better screen, and something that has standard APIs. Java+Linux and Java+EPOC are both reasonable choices.
PalmOS is pretty much obsolete; even Palm knows that. The next version may still be called "PalmOS", but it will be something entirely different.
I signed up for their developer program. I got back a message announcing the developer deal, so tried ordering at the developer price of $179. No luck: their site only offers retail pricing. Are these things still available at developer pricing?
You can probably run PalmOS in emulation on various PDAs. In fact, I suspect that's what we are going to see in a year or two from Palm: an entirely new OS that runs the old stuff on an emulator.
In contrast, the Palm Pilot requires you to write applications to its own limited and proprietary APIs, memory model, and threading model. The Palm Pilot runs its applications reasonably well, but someone had to invest a lot of time and effort in that; doing anything more with it is an uphill struggle.
The limitations of the PalmOS and the messy programming model are a consequence of trying to deliver a nice system on limited hardware. Those constraints don't apply anymore, and the Palm Pilot is now pretty much an anachronism.
I have a Palm Pilot. It's a reliable workhorse for its very limited set of applications. I'll gladly toss it out when I get a decent, reliable UNIX-based PDA. The Agenda could be it.
That is one of the many reasons why centralized web-based services are not such a good idea. In fact, we have a perfectly good service for distributing information widely without the ability for anyone to identify readers: USENET. For better or for worse (I think for worse, actually), it is even permanently archived and searchable now. And USENET offers a choice of jurisdiction of where the identity of a poster is protected.
The problem isn't lack of languages, it's lack of commitment to working on existing systems. For example, Guido preferred going off and doing his own thing with Python rather than create a free Smalltalk implementation. The result is that we now have Python, a language that's a little bit nicer to use than Smalltalk but whose implementation is a whole lot worse than that of existing Smalltallk implementations. If people keep starting from scratch, no wonder that implementations hardly ever get really good.
It's kind of ironic that LX is tied to a system called "Mozart", which is also the name of another new, experimental language and environment.
I think the PC case is a pretty good place for a power supply. But many peripherals could be powered through their data cables; both USB and FireWire offer that choice. Unfortunately, making anything low power tends to make it more expensive, so the cheap stuff will likely continue to come with its own separate power supplies.
Taking a PC designed for a shielded metal case and sticking it into an unshielded plastic case seems like a sure way of creating interference for your neighbors: radio, television, amateur radio, portable telephones, medical devices, wireless networking, cell phones, etc. It's kind of like turning up your stereo all the way, and I don't think that's particularly nice. If people track you down, FCC regulations let them demand that you shut down your computer because it creates harmful interference to licensed devices.
So, save yourself some time and money, be nice to your neighbors, and leave that motherboard in the shielded case it was designed for. If you want a clear PC, buy one that was designed that way from the start, like an iMac or a G4 Cube.
The P4 probably has many flaws, but for compute-intensive applications, the fastest P4 is significantly faster than the fastest Athlon. If you haven't benchmarked it yourself, check out the benchmarks at spec.org.
Intel should pray that their transition is as smooth as Apple's 68k->PPC changeover; I don't think it will be. Apple really made that work as well as possible. In fact, Apple seems to be pretty good at that: the MacOS9->MacOSX transition looks like it's going OK as well.
No, actually, I generally just try to respect the intent of the author when feasible. Book authors and newspaper writers usually anticipate and intend for their writings to get archived for the long term, and they represent writings that people spent a long time creating for a public audience. But authors of USENET articles thought they were participating in an informal, short-lived discussion.
So, why is your response anonymous? Where are your real name and your E-mail address?
Once it became clear that USENET was increasingly becoming permanently fixed and searchable, I stopped participating under my real name. I never flamed on it or participated in particularly controversial subjects, but I still didn't want to have to deal with the possibility of being quoted out of context years later.
While anonymity has many undesirable features, it is the second-best choice if you can't have informal, short-lived discussions (this is, incidentally, why I'm not using my real name on Slashdot). For me, what killed USENET was not anonymity but its permanent archiving.
I think something similar has happened in politics: since everything is getting recorded and republished and analyzed word-for-word, politicians can't engage in thoughtful debate anymore in public for fear of offending someone or getting attacked on out-of-context quotes. Instead, every political message has to be carefully crafted and rehearsed; no extraneous utterance or debate is possible.
From the legal point of view, the posters of the original messages hold the copyrights. While USENET always ended up on some backup tapes somewhere, arguably, the original posters didn't give consent to having their comments archived and made accessible forever by some big search engine company.
Legal issues aside, I think it is also a bad idea to make content from an informal discussion forum easily available for eternity. For participants in such forums to have to worry about whether something they say today may be used against them decades from now kills participation by thoughtful people. Of course, these days, many people try to protect themselves against that in part by using pseudonyms, but in 1995, that wasn't so common yet.
So, I don't applaud Google at all for putting this content back on-line. Google should not change the rules for content created years ago. And even today, I think it would be better to return USENET to an ephemeral, informal medium as much as possible.
I hope Google will go back to making only the last six months (or so) of content available. I believe that is what DejaNews ended up doing, and it's the right thing to do.
I don't see that as an obstacle. The Xerox author could have removed himself from the paper (and instead been moved into the acknowledgement section) and Xerox could have formally protested the publication. Felten could then have gone ahead and published it anyway. Formally, he might have been guilty of copyright violation, but Xerox is under no obligation to pusue that.
Technically, it was almost certain from the beginning that all these schemes could be broken. And once the preprint was published on their web site, the cat was out of the bag.
The whole participation of the Princeton group in the SDMI effort was a political statement from the beginning. The decision to withdraw the paper is likewise a political and strategic decision with no technical significance. Let's just hope those guys know what they are doing when it comes to politics and strategy, because, so far, it isn't clear to me where they are going with this.
Load balancing via IP routing tricks is kind of nice. Mosix goes one step further and allows live processes to migrate across a cluster. Experimental add-ons also will do socket migration and havedistributed file system support. I think that's the kind of approach to clustering you are going to see in the long run.
They also seem to have trouble understanding that watermarking is not technically feasible. It won't take some really smart guys from Princeton to break this or future systems. Given Chiariglione's inelegant and messy technical track record, I doubt they are going to get a technical clue any time soon either.
Let them add poor watermarks to poor content and create players with all sorts of limitations. In the long run, it's only going to hurt their business. Dealing with these people is a waste of time in my opinion.
Insulted? In what way? Can you provide details? How can you "insult" a company anyway?
He should have praised [TrollTech] diplomatically for their decision, as this was a victory for the FSF and the GPL.
You mean he should have praised TrollTech for their decision not to go out of business? TrollTech's hand was forced: without KDE, Qt would be just one of a dozen obscure commercial toolkits. And without changing Qt to a GPL license, KDE would have had to come to an end.
What this whole thing shows is that RMS's care and stubbornness in designing the GPL has paid off. Without it, TrollTech would have been able to hijack KDE, a major high-quality open-source project, for their commercial purposes.
As for TrollTech, I'd continue to view them with suspicion. Their vision seems to be that any commercial user of open source software should pay some commercial company, and they seem eager to want to do whatever they can to make Qt an integral part of whereever Linux goes. That is not the kind of vision many supporters of open source have, and pragmatically, it is also very bad for the long term viability of open source software.
Eazel already got the benefit from making their software free: the chance of having it adopted. Sure, their business model may not have worked out, so that wasn't enough. But without it, how much chance of widespread use do you think yet another closed source C-based application suite would have had?
In principle, Python could be compiled like C, C++, Java, Fortran, Eiffel, etc. into standalone executables. That would both simplify deployment and (if done well) improve speed. Previous attempts at this have not been particularly popular or successful, however.
But a prototype-based scripting language with JavaScript semantics isn't very Smalltalk-like. Smalltalk has much better development tools, better APIs, and better error handling. It can also be compiled more efficiently.
Like so many things that recently have made a splash (extreme programming, refactoring, patterns, etc.), this is really just giving a name to programming practices and ideas that have been around for two decades. Smalltalk was one of the languages where such ideas originated (Lisp was another major one), they just didn't originate recently.
(There have also been some refactoring tools written for Java, but the nature of the Java language will make it difficult to create a tool as powerful as the Refactoring Browser for Smalltalk.)
Quite to the contrary. While Java isn't as convenient to program in as Smalltalk, the additional type information that Java source code contains lets you do more kinds of refactoring safely. In fact, one of the reasons refactoring has not been such a big deal in the C++/Java world is because static type checking in those languages already helps you with a lot of refactorings. For example, in Java, if you change an interface or move instance variables around, the type checker will point out every single place that you need to change. If you use a good IDE, you can breeze through those places and a refactoring only takes a single edit, compile, edit cycle. You still need to make the changes by hand, but that's pretty quick, actually faster than any of the refactoring browsers I have used in many cases.
That's what weak dictionaries are for: they allow you to create such bidirectional associations easily.
That doesn't change things much (yes, I had actually looked at the site). If the service goes away and you don't have any alternatives, you are still stuck. And those alternatives will only exist if there is demand for people to run web applications at their own hosting services. (It also raises all sorts of concerns about security and your terms of service with your web hosting company.)
The kind of freedom you're talking about comes at a price. Not everybody wants to be a programmer.
You don't have to be a programmer to install a web page editor on your machine or to install a web application on a well-run hosting service.
Note that by posting to Slashdot, you're creating content that allows a business you don't own to make money
I have no problem with people making money, Blogger, Trellix, or otherwise.