You don't have to look quite that deep to find evidence of Bush's weak mind. How about his refences to the Japanians and YugoSLAVES? There were some equally laughable remarks just recently relating to geography, enough to evoke nothing but a snort at the mention of his name. Compared to this, Gore's claims of inventing the Internet are almost erudite.
> Same here. My facility at work is having 802.11 wireless installed
Ahhh, I'm trying to subtly plant the seed for such a need in the mind of managers here. My project manager brings in her laptop and plugs it in every morning, so I'm trying to convince her how much more convenient it would be to have a wireless NIC, allowing her to roam through the building. Besides, it would allow ME to roam through the building.
> I still like the Palm line for some things, but the color attempt needs to be redone. I wasn't impressed with the IIIc at all.
The IIIc is useless as far as I'm concerned. The pixels are too big and give the display a grainy appearance. I wish they would concentrate on getting beyond 160x160 and square aspect ratio instead, and getting rid of the dedicated Graffiti input area. All of the display opening in the front face should be display, and they should push the display more out towards the edges of the device. In other words, give me a bigger screen with more resolution!
One other thing I would LOVE in the Palm is a BACK button, that lets you switch to the last used app. Too often I need to pop up an app while in another one to look up something, and then I have to go to the App screen again, scroll around to find the previous app, and switch to it. With the BACK button I could simply switch to another app, do the lookup, then hit the BACK button and be back where I was. Oh, and please convert the calculator into a popup dialog, like the Find dialog, so it doesn't move away from the currently used app.
> Ever seen the iPaq Pocket PC outside? I'd have to say it's readibility actually beats my old Palm V.
No, I haven't seen it. But for every such glowing report I hear many more that complain about the washed-out screen under normal daylight conditions (not in the shade, or in a sunlit room, but under the clear sky).
> My iPaq is now a device that I charge nightly along with other things I own.
Well, my personal habits don't go for that. That's why I also don't carry other rechargable devices, such as a cell phone. I end up simply not doing it, and it's dead when I do need it. Besides, I simply don't hotsync often enough for the recharging to be transparent, maybe once a week or so. My wife hotsyncs her Visor maybe once a month. And I definitely don't leave it for long in the cradle. Hotsync and back in the pocket.
> With the new generation of flashy WinCE 3 devices, many of the 2.11 devices are cheaper then midrange Palms.
Yeah, but we're comparing latest generation Palms here, so I don't need an old and deficient WinCE 2.1. The latest V3 might be getting more usable, but the older versions were just crap.
> Overall, I moved from Palm to PocketPC because I wanted exactly that, a Pocket PC.
Fair enough. That's what I actually want as well, but for me the best compromise currently are the Palms. I'm simply too averted by short battery life, I just don't want something that I constantly have to baby. It's bad enough that I have to change the batteries every 2-3 weeks when I'm particularly heavy into Galaxian and Tetris, but worrying about power twice a week--forget it.
> But with a handheld that can play MPEG-4 movies
I'm sorry, but that's a joke. Besides the we-can-do-it factor, I see absolutely no purpose in that, and please spare me the I-stream-video-over-wireless thing. I know more people that have ditched their expensive Casio E-xxx because the video and MP3 novelty wore off than I know people that ditched Palms because they NEEDED video.
> access to PCMCIA wireless (modem or LAN)
Here I fully agree, I think wireless connectivity is the killer app that will push PDAs out of their niche. I'm pining for the day when I can carry my PDA around the entire company with constant connectivity, live email updates, web browsing during boring meetings, etc. All without killing the batteries in 2 hours or less.
Color screens I personally find quite unnecessary. I know many disagree, but for useful information display a good contrast monochrome screen like my M100 is perfectly acceptable. Only when color requires no compromises in readability and especially power usage will I change my opinion. I think a large screen size with good resolution is more important. I wish manufacturers would work harder on converting more of the frontal area of the PDAs into screen.
Actually, I do rather like the new iPaq. I think it's pretty close to an ideal PDA in terms of functionality and power. All it needs now is 1-2 weeks of good use per charge, and more expandability. And a lower price. Carrying something that cost over $500 carelessly in the pant pocket is not my idea of prudence. My cheap M100 on the other hand I don't worry too much about.
It's interesting that in the three lengthy paragraphs preceding your link you don't offer a single valid explanation for why even single-threaded apps would benefit from SMP, only strong assertions that such would be the case. In fact, the three paragraphs smack suspiciously of marketing-speak, conveniently sidestepping anything resembing an argument.
Microkernels are great, not many people would dispute that, but what the hell do they have to do with SMP per se? In order for an efficient MP architecture (be it symmetrical, distributed etc) to be effective, applications have to be coded to take advantage of it. This includes multi-threading it in meaningful ways. Just putting QNX and Neutrino on your system won't make your linear app any more efficient than running it under DOS--except for the GUI being more responsive, possibly. Just because system resources are multi-threaded doesn't mean squat--your app is still sitting around waiting for disk-bound operations to complete, no matter how those are implemented.
Unless you know how to effectively multi-thread your app, neither QNX nor BeOS will do much to improve your code. You can make any assertions and include any links to the opposite you want, but that's a pretty fundamental tenet of current OSs.
I tend to disagree with your point about the WinCE interface being like Win9x and hence requiring no training. Especially Pocket PC deviates from the desktop interface quite a bit, I found myself hunting for stuff. With good reason, because they're trying to be more like Palm with instant access to certain features.
I think the important GUI paradigm is the WIMP (or WIP I guess), or the familiar window/dialog box, buttons, edit boxes, check boxes, combo boxes, menus. People are used to this paradigm by now, no matter what particular appearance these items might take. I maintain that these GUI widgets look only trivially different on Palm OS than in WinCE or Win9x. By far most people still recognize the (monochrome, with rounded corners) buttons as buttons, the edit boxes as edit boxes, the check boxes as check boxes. One possible exception is the drop down menu, which is not normally visible on Palm OS, so that's a small item to learn. But most people seeing a Palm form with some GUI widgets for the first time will intuitively know how to navigate it.
It does not take a verbatim copy of the Win9x desktop to instill familiarity, because even on WinCE the Start button menu doesn't really contain all the stuff from Win9x, and the differences increase rapidly from there.
My point is that the argument saying that WinCE is much more familiar to users than the Palm OS is only superficially true but really doesn't hold much water. I've seen too many non-Palm users pick up my Visor or Pilot and pretty quickly find their way around it.
+larger screens, better resolution, more fields per screen
+relative code portability from the desktop
-color screens almost unreadable in daylight, great at night
-code portability might depend on OS features used
-short battery life, which can be a problem depending on usage model
-expensive devices, a factor in a harsh environment
Palm:
+very readable screen in daylight, usable at night
+extremely long battery life, enabling all sorts of usage models
+cheap devices, almost disposable
-OS harder to program and more limitations than WinCE (except if you choose VC++ on WinCE, then it's a bitch regardless:-)
-smaller screen, fewer fields per screen
-Graffiti might make some people balk
All in all, I'm leaning towards Palm, because of the very readable screen, the long battery life, and the cheapness of the devices. However, you should evaluate WinCE and see how much of your legacy code would REALLY be portable; if a substantial amount needs a rewrite (e.g. because of GUI issues), write from scratch and go with Palm. If you can mostly port verbatim, go with WinCE.
On Palm you can use Satellite Forms to create very VB-ish apps, complete with event handler scripts etc. Check it out, for data acqusition it makes app development almost trivial, and all your data automatically ends up in database tables on the desktop. You could also use the Symbol devices with the barcode reader if you have that kind of need.
Before you go off on tirades, maybe analyse what you're about to say first. First, there's nothing magical about QNX or BeOS. Sure, BeOS apps may be automatically multithreaded, but only in a relatively superficial way (by putting the message loop into a separate thread if I'm not mistaken). BeOS certainly doesn't take your linear code and somehow magically extract multiple threads of execuion--you still have to do the leg work. Same with QNX. Same with WinNT/2K, if you multi-thread you apps, they will spread quite nicely across CPUs (though HOW nicely is a matter of debate for the religious). Don't get me wrong, I think what BeOS is doing is perfectly fine and laudable, but it's not exactly what you're implying.
The fact is, in you model the unit of execution is the thread. What you yourself don't know how to de-serialise and pull into separate threads (and properly synchronise), the compiler or OS certainly won't do for you. So even if your app is multithreaded, if your threads are big fat chains of serial code, the app won't benefit any from multiple CPUs. The holy grail of MP is a compiler that could, for a trivial example, look at your loop and be able to unroll it into x smaller loops working on subunits of the data.
One of the big promises of MP is to avoid the end of Moore's law through parallel computing, IN GENERAL PURPOSE COMPUTERS, not just arcane research machines. We already know how to tickle esoteric MP hardware today into doing our bidding, but it's no trivial task and takes a lot of skill. If MP is to give us a mainstream migration path from single processing, it can't expect more from programmers than they can give today. In other words, MP machines will have to deliver even with mediocre programmers, because they form the bulk of the work force. You can't stipulate as a condition for effective MP an overall higher quality work force, because it ain't happening.
As already mentioned, as Moore's law runs out of steam, something else will be needed. Besides, it really doesn't matter how long Moore's law still holds; for any extremely powerful single processor it still holds true that for certain problems X of them will be X times more powerful still. While the theoretical processing power of some as-yet-unimplemented molecuar or quantum computing devices could be quite high, it will still be very finite and fall short of the requirements of some types of processing--such as nuclear reaction simulations, for that next generation of smart atomic weapons that can single you out and kill you based on certain profiles.
Besides, this 128 processor devices isn't really any different from most other multi-processing systems. The holy grail is still the development of smart compilers and algorithms that can allow even dumb programmers to write effective multi-processing code.
"Menschheit" is gender-neutral and literally means humanity. "Mensch" means human, not man, though is often translated that way. Usually phrases like "Der Mensch ist ein seltsames Tier" get translated as "Man is a strange animal" rather than "The human is a strange animal", I guess because it sounds less awkward.
> a windows application developer going to do when faced with the task of
> porting his windows app to unix.
He ought to use Kylix, which hopefully will take care of the KDE/Gnome differences fairly transparently. Not to speak of the Windows/Linux differences.
> People who view these screen shots [...] enjoy having sex with children.
Since you know what's on those screenshots, you must have viewed them as well, ergo you also enjoy having sex with children. Correct me if I'm wrong:-)
I was actually thinking something along those lines when I saw the screenshots, and it strikes me that a lot of geeks seem to be into the repressed sexual phantasies thing, be it the pedophile or just plain nudie pics angle. Also check out the latest screenshots of KDE 2, where the guy might think that he covered up his desktop wallpaper strategically. But the desktop thumbnail gave him away. Dirty young man!!!
I don't think the math for your proposal works out quite like that. There are indeed rail gun efforts under way, but none of them aim to eliminate rocket boosters altogether. The best you can hope for is a good push that reduces the size of the booster, but I doubt you can completely eliminate it. It's a long way through thick atmosphere into space, and once the vehicle leaves the rail, it is no longer accelerating. So it has to reach its maximum speed at the end of the rail, after which it will only (strongly) decelerate. I'd love to be proved wrong on that, really, but I don't think it'll quite work out that way.
I think NASA's glory days are over. For the last three decades they've pretty much been coasting on the Apollo program success, the shuttles notwithstanding (which were conceived at the height of their success anyway). They've done little remotely innovative since then.
The ESA operates on very different principles. From day one they were more of a commercial venture rather than a prestige program. Of course, part of the reason for that was that they knew it was futile to try to outspend the Americans. Europeans are notoriously reluctant to spend money on big technology without the prospect of ROI. I'm sure exceptions exist, but as a rule that's pretty true. So the ESA has slowly chipped away at the commercial satellite market, and unbeknownst to most, they're now the largest satellite launcher in the world. The fact is, until we start mining asteroids or what have you, satellites are the only lucrative space business.
My feeling is that--like Airbus--the ESA will fairly quietly work away at it, until one day they'll just happen to be the number one player (ok, so Airbus hasn't been that quiet really, but the ESA certainly has). I think one key element of that success will be the development of SSTO technology. I've been reading the ESA web pages (in particular DASA), and more than NASA almost they seem to be under the impression that SSTO will be the only way to REALLY make money in the long run. They view the Ariane launchers as strictly a short-term money making technology, but nothing to base the future on.
Basically, until we can literally take off into space from a runway and return with the exact same vehicle, space travel will still be in its infancy and considered brute force technology. The longer we fool around with concepts like the Venture Star, which fights the atmosphere every step of the way into orbit, instead of taking advantage of it, the longer it will take to REALLY get into space.
> Where did you go to school, Bob Jones University?
He, he, yeah, they probably offered their famous Creationist Quantum Mechanics course that summer. The lab was conducted by Charlton Heston, demonstrating moving molecules with the tip of a bullet.
> Wasn't Bill G. quoted once as saying something like, "At least their pirating MY software." ?
That's exactly right. I think MS is prepared to (and does) tacitly accept a lot of piracy in the home arena for the sake of brainshare. While some people might blindly disagree that home software piracy is rampant, that is one of the principal reasons for MS's desktop dominance. Unfortunately it is also something extremely hard to get concrete figures for, since few people will openly admit to piracy when questioned (even when promised anonymity). You pretty much have to rely on empirical data and on personal observations.
a. I'm not talking about corporate piracy, that would be pretty damn stupid. Large corporations are easy and juicy targets to sue if they're stupid enough to pirate. I'm strictly talking about individuals in a home environment.
b. You might have found the one exception to the rule there: low-end MS OS upgrades. For some reasons, these upgrades fly off the shelves. I think a large part of the reason is the price point: they're at or below the $100 magic mark. You'll most likely find the same isn't happening with Win2K or MS Office. Go ask your Electronic Boutique, Best Buys, CompuUSA etc, they'll most likely tell you that those sales completely pale in comparison to Win9x/Me upgrades.
Speaking of MSDN subscriptions, that's one of the HUGE copy-for-home-use candidates in the developer community. Considering the price point, that's not really surprising.
The fact is, most people probably DON'T buy MS software, they swipe it one way or another. If my last few jobs at LARGE companies are any indication of the general state of piracy, most people get their software either from work or friends. Or installed on new machines, but most people refresh their software versions way more frequently than their machines. Considering that the software on many people's machines would amount to over a thousand dollars if purchased legally, there's no way people would actually shell out that kind of money. If people actually had to plonk down the $300 or so for MS Office, there's no doubt in my mind that MS Works or Star Office would be the market leaders.
So in a perverted sense, while MS fights piracy with all their might, their brainshare and desktop share is due in large part to soft piracy. In that light, it doesn't matter how nasty Microsoft's image gets, since a lot of people don't pay for the software anyway, they don't mind swiping it regardless of MS's bad PR.
I see absolutely no disadvantage to being able to layout the application GIU visually, rather than creating widgets in code and constantly having to fool with the metrics and go through endless fiddle/compile/run cycles until it looks right. In particular a nice RAD IDE like Delphi makes life very easy, with decent layout control like Align and Anchors, though they could add a few more options such as percent of window.
I used to be a C++ zealot until about 1997 when I got the first taste of Delphi 2, and I never went back. Sure, I had to switch languages, but I've made bigger sacrifices before. Besides, OP leads to a clarity and readability of code rarely achieved nowadays with C++, expecially when dealing with COM.
RAD from the perspective of Delphi gives you the best of both worlds: easy GUI design, as well as a great class framework and a powerful language. It ties the two together in such a slick and unobtrusive way; it's the way I'd probably do things even if Delphi didn't exist. Sure it adds some overhead, but then every framework does--even MFC, if you can call it a framework. If you want leaner apps, you pretty much have to drop to the API/message loop level, and we've all been there and didn't like it. Compared to something like VC++ that gives you RAD Light and forces you to keep your GUI and event handlers synchronized manually and always keeps you worrying about resource IDs, Delphi takes you to a whole other level of productivity.
The project manager at the company I used to work for was a great MS zealot, and his big mission in life was to convert the shop from Delphi to VC++. We were five developers churning out the code of ten and still were constantly behind schedule (talk about over-commitment of sales). We pleaded with him repeately to reconsider, explaining what a great work multiplier Delphi is and how such things are important to a small company. His retort was always that VC++ is the tool of the future and that tons of very smart guys work at MS and they can't all be wrong. He constantly denied that there were any productivity differences between Delphi and VC++ and accused us of just being irrational Delphi zealots (which probably was true, but with good reason we felt).
Anyway, these kinds of attitudes--influenced more by the great MS marketing machine than reality--are what is keeping Delphi down. I guess all of us Delphi supporters are hoping that Kylix will be the second wind for Delphi. With the momentum Linux has at the moment, and its lack of a serious RAD tool, Delphi should stand a decent second chance. And please don't anyone mention.NET or I'll have to hunt you down and decompose you into little bits.
Slashdot Mirror
Japanians, Yugoslaves.
If I only have a choice between two fools, I take the articulate one, thank you very much.
What the hell did you just say?
You don't have to look quite that deep to find evidence of Bush's weak mind. How about his refences to the Japanians and YugoSLAVES? There were some equally laughable remarks just recently relating to geography, enough to evoke nothing but a snort at the mention of his name. Compared to this, Gore's claims of inventing the Internet are almost erudite.
> Same here. My facility at work is having 802.11 wireless installed
Ahhh, I'm trying to subtly plant the seed for such a need in the mind of managers here. My project manager brings in her laptop and plugs it in every morning, so I'm trying to convince her how much more convenient it would be to have a wireless NIC, allowing her to roam through the building. Besides, it would allow ME to roam through the building.
> I still like the Palm line for some things, but the color attempt needs to be redone. I wasn't impressed with the IIIc at all.
The IIIc is useless as far as I'm concerned. The pixels are too big and give the display a grainy appearance. I wish they would concentrate on getting beyond 160x160 and square aspect ratio instead, and getting rid of the dedicated Graffiti input area. All of the display opening in the front face should be display, and they should push the display more out towards the edges of the device. In other words, give me a bigger screen with more resolution!
One other thing I would LOVE in the Palm is a BACK button, that lets you switch to the last used app. Too often I need to pop up an app while in another one to look up something, and then I have to go to the App screen again, scroll around to find the previous app, and switch to it. With the BACK button I could simply switch to another app, do the lookup, then hit the BACK button and be back where I was. Oh, and please convert the calculator into a popup dialog, like the Find dialog, so it doesn't move away from the currently used app.
> Ever seen the iPaq Pocket PC outside? I'd have to say it's readibility actually beats my old Palm V.
No, I haven't seen it. But for every such glowing report I hear many more that complain about the washed-out screen under normal daylight conditions (not in the shade, or in a sunlit room, but under the clear sky).
> My iPaq is now a device that I charge nightly along with other things I own.
Well, my personal habits don't go for that. That's why I also don't carry other rechargable devices, such as a cell phone. I end up simply not doing it, and it's dead when I do need it. Besides, I simply don't hotsync often enough for the recharging to be transparent, maybe once a week or so. My wife hotsyncs her Visor maybe once a month. And I definitely don't leave it for long in the cradle. Hotsync and back in the pocket.
> With the new generation of flashy WinCE 3 devices, many of the 2.11 devices are cheaper then midrange Palms.
Yeah, but we're comparing latest generation Palms here, so I don't need an old and deficient WinCE 2.1. The latest V3 might be getting more usable, but the older versions were just crap.
> Overall, I moved from Palm to PocketPC because I wanted exactly that, a Pocket PC.
Fair enough. That's what I actually want as well, but for me the best compromise currently are the Palms. I'm simply too averted by short battery life, I just don't want something that I constantly have to baby. It's bad enough that I have to change the batteries every 2-3 weeks when I'm particularly heavy into Galaxian and Tetris, but worrying about power twice a week--forget it.
> But with a handheld that can play MPEG-4 movies
I'm sorry, but that's a joke. Besides the we-can-do-it factor, I see absolutely no purpose in that, and please spare me the I-stream-video-over-wireless thing. I know more people that have ditched their expensive Casio E-xxx because the video and MP3 novelty wore off than I know people that ditched Palms because they NEEDED video.
> access to PCMCIA wireless (modem or LAN)
Here I fully agree, I think wireless connectivity is the killer app that will push PDAs out of their niche. I'm pining for the day when I can carry my PDA around the entire company with constant connectivity, live email updates, web browsing during boring meetings, etc. All without killing the batteries in 2 hours or less.
Color screens I personally find quite unnecessary. I know many disagree, but for useful information display a good contrast monochrome screen like my M100 is perfectly acceptable. Only when color requires no compromises in readability and especially power usage will I change my opinion. I think a large screen size with good resolution is more important. I wish manufacturers would work harder on converting more of the frontal area of the PDAs into screen.
Actually, I do rather like the new iPaq. I think it's pretty close to an ideal PDA in terms of functionality and power. All it needs now is 1-2 weeks of good use per charge, and more expandability. And a lower price. Carrying something that cost over $500 carelessly in the pant pocket is not my idea of prudence. My cheap M100 on the other hand I don't worry too much about.
> This would refute EVERYTHING that you said.
It's interesting that in the three lengthy paragraphs preceding your link you don't offer a single valid explanation for why even single-threaded apps would benefit from SMP, only strong assertions that such would be the case. In fact, the three paragraphs smack suspiciously of marketing-speak, conveniently sidestepping anything resembing an argument.
Microkernels are great, not many people would dispute that, but what the hell do they have to do with SMP per se? In order for an efficient MP architecture (be it symmetrical, distributed etc) to be effective, applications have to be coded to take advantage of it. This includes multi-threading it in meaningful ways. Just putting QNX and Neutrino on your system won't make your linear app any more efficient than running it under DOS--except for the GUI being more responsive, possibly. Just because system resources are multi-threaded doesn't mean squat--your app is still sitting around waiting for disk-bound operations to complete, no matter how those are implemented.
Unless you know how to effectively multi-thread your app, neither QNX nor BeOS will do much to improve your code. You can make any assertions and include any links to the opposite you want, but that's a pretty fundamental tenet of current OSs.
I tend to disagree with your point about the WinCE interface being like Win9x and hence requiring no training. Especially Pocket PC deviates from the desktop interface quite a bit, I found myself hunting for stuff. With good reason, because they're trying to be more like Palm with instant access to certain features.
I think the important GUI paradigm is the WIMP (or WIP I guess), or the familiar window/dialog box, buttons, edit boxes, check boxes, combo boxes, menus. People are used to this paradigm by now, no matter what particular appearance these items might take. I maintain that these GUI widgets look only trivially different on Palm OS than in WinCE or Win9x. By far most people still recognize the (monochrome, with rounded corners) buttons as buttons, the edit boxes as edit boxes, the check boxes as check boxes. One possible exception is the drop down menu, which is not normally visible on Palm OS, so that's a small item to learn. But most people seeing a Palm form with some GUI widgets for the first time will intuitively know how to navigate it.
It does not take a verbatim copy of the Win9x desktop to instill familiarity, because even on WinCE the Start button menu doesn't really contain all the stuff from Win9x, and the differences increase rapidly from there.
My point is that the argument saying that WinCE is much more familiar to users than the Palm OS is only superficially true but really doesn't hold much water. I've seen too many non-Palm users pick up my Visor or Pilot and pretty quickly find their way around it.
WinCE:
:-)
+larger screens, better resolution, more fields per screen
+relative code portability from the desktop
-color screens almost unreadable in daylight, great at night
-code portability might depend on OS features used
-short battery life, which can be a problem depending on usage model
-expensive devices, a factor in a harsh environment
Palm:
+very readable screen in daylight, usable at night
+extremely long battery life, enabling all sorts of usage models
+cheap devices, almost disposable
-OS harder to program and more limitations than WinCE (except if you choose VC++ on WinCE, then it's a bitch regardless
-smaller screen, fewer fields per screen
-Graffiti might make some people balk
All in all, I'm leaning towards Palm, because of the very readable screen, the long battery life, and the cheapness of the devices. However, you should evaluate WinCE and see how much of your legacy code would REALLY be portable; if a substantial amount needs a rewrite (e.g. because of GUI issues), write from scratch and go with Palm. If you can mostly port verbatim, go with WinCE.
On Palm you can use Satellite Forms to create very VB-ish apps, complete with event handler scripts etc. Check it out, for data acqusition it makes app development almost trivial, and all your data automatically ends up in database tables on the desktop. You could also use the Symbol devices with the barcode reader if you have that kind of need.
Before you go off on tirades, maybe analyse what you're about to say first. First, there's nothing magical about QNX or BeOS. Sure, BeOS apps may be automatically multithreaded, but only in a relatively superficial way (by putting the message loop into a separate thread if I'm not mistaken). BeOS certainly doesn't take your linear code and somehow magically extract multiple threads of execuion--you still have to do the leg work. Same with QNX. Same with WinNT/2K, if you multi-thread you apps, they will spread quite nicely across CPUs (though HOW nicely is a matter of debate for the religious). Don't get me wrong, I think what BeOS is doing is perfectly fine and laudable, but it's not exactly what you're implying.
The fact is, in you model the unit of execution is the thread. What you yourself don't know how to de-serialise and pull into separate threads (and properly synchronise), the compiler or OS certainly won't do for you. So even if your app is multithreaded, if your threads are big fat chains of serial code, the app won't benefit any from multiple CPUs. The holy grail of MP is a compiler that could, for a trivial example, look at your loop and be able to unroll it into x smaller loops working on subunits of the data.
One of the big promises of MP is to avoid the end of Moore's law through parallel computing, IN GENERAL PURPOSE COMPUTERS, not just arcane research machines. We already know how to tickle esoteric MP hardware today into doing our bidding, but it's no trivial task and takes a lot of skill. If MP is to give us a mainstream migration path from single processing, it can't expect more from programmers than they can give today. In other words, MP machines will have to deliver even with mediocre programmers, because they form the bulk of the work force. You can't stipulate as a condition for effective MP an overall higher quality work force, because it ain't happening.
As already mentioned, as Moore's law runs out of steam, something else will be needed. Besides, it really doesn't matter how long Moore's law still holds; for any extremely powerful single processor it still holds true that for certain problems X of them will be X times more powerful still. While the theoretical processing power of some as-yet-unimplemented molecuar or quantum computing devices could be quite high, it will still be very finite and fall short of the requirements of some types of processing--such as nuclear reaction simulations, for that next generation of smart atomic weapons that can single you out and kill you based on certain profiles.
Besides, this 128 processor devices isn't really any different from most other multi-processing systems. The holy grail is still the development of smart compilers and algorithms that can allow even dumb programmers to write effective multi-processing code.
Explain!
> If more americans had been to school...
If more of them had been to school, indeed, they might actually capitalize "Americans".
"Menschheit" is gender-neutral and literally means humanity. "Mensch" means human, not man, though is often translated that way. Usually phrases like "Der Mensch ist ein seltsames Tier" get translated as "Man is a strange animal" rather than "The human is a strange animal", I guess because it sounds less awkward.
> a windows application developer going to do when faced with the task of
> porting his windows app to unix.
He ought to use Kylix, which hopefully will take care of the KDE/Gnome differences fairly transparently. Not to speak of the Windows/Linux differences.
> People who view these screen shots [...] enjoy having sex with children.
:-)
Since you know what's on those screenshots, you must have viewed them as well, ergo you also enjoy having sex with children. Correct me if I'm wrong
I was actually thinking something along those lines when I saw the screenshots, and it strikes me that a lot of geeks seem to be into the repressed sexual phantasies thing, be it the pedophile or just plain nudie pics angle. Also check out the latest screenshots of KDE 2, where the guy might think that he covered up his desktop wallpaper strategically. But the desktop thumbnail gave him away. Dirty young man!!!
Precisely, that's why the maglev rail guns are under investigation. But eliminating the boosters altogether would be quite hard.
I don't think the math for your proposal works out quite like that. There are indeed rail gun efforts under way, but none of them aim to eliminate rocket boosters altogether. The best you can hope for is a good push that reduces the size of the booster, but I doubt you can completely eliminate it. It's a long way through thick atmosphere into space, and once the vehicle leaves the rail, it is no longer accelerating. So it has to reach its maximum speed at the end of the rail, after which it will only (strongly) decelerate. I'd love to be proved wrong on that, really, but I don't think it'll quite work out that way.
I think NASA's glory days are over. For the last three decades they've pretty much been coasting on the Apollo program success, the shuttles notwithstanding (which were conceived at the height of their success anyway). They've done little remotely innovative since then.
The ESA operates on very different principles. From day one they were more of a commercial venture rather than a prestige program. Of course, part of the reason for that was that they knew it was futile to try to outspend the Americans. Europeans are notoriously reluctant to spend money on big technology without the prospect of ROI. I'm sure exceptions exist, but as a rule that's pretty true. So the ESA has slowly chipped away at the commercial satellite market, and unbeknownst to most, they're now the largest satellite launcher in the world. The fact is, until we start mining asteroids or what have you, satellites are the only lucrative space business.
My feeling is that--like Airbus--the ESA will fairly quietly work away at it, until one day they'll just happen to be the number one player (ok, so Airbus hasn't been that quiet really, but the ESA certainly has). I think one key element of that success will be the development of SSTO technology. I've been reading the ESA web pages (in particular DASA), and more than NASA almost they seem to be under the impression that SSTO will be the only way to REALLY make money in the long run. They view the Ariane launchers as strictly a short-term money making technology, but nothing to base the future on.
Basically, until we can literally take off into space from a runway and return with the exact same vehicle, space travel will still be in its infancy and considered brute force technology. The longer we fool around with concepts like the Venture Star, which fights the atmosphere every step of the way into orbit, instead of taking advantage of it, the longer it will take to REALLY get into space.
> Where did you go to school, Bob Jones University?
He, he, yeah, they probably offered their famous Creationist Quantum Mechanics course that summer. The lab was conducted by Charlton Heston, demonstrating moving molecules with the tip of a bullet.
> Wasn't Bill G. quoted once as saying something like, "At least their pirating MY software." ?
That's exactly right. I think MS is prepared to (and does) tacitly accept a lot of piracy in the home arena for the sake of brainshare. While some people might blindly disagree that home software piracy is rampant, that is one of the principal reasons for MS's desktop dominance. Unfortunately it is also something extremely hard to get concrete figures for, since few people will openly admit to piracy when questioned (even when promised anonymity). You pretty much have to rely on empirical data and on personal observations.
a. I'm not talking about corporate piracy, that would be pretty damn stupid. Large corporations are easy and juicy targets to sue if they're stupid enough to pirate. I'm strictly talking about individuals in a home environment.
b. You might have found the one exception to the rule there: low-end MS OS upgrades. For some reasons, these upgrades fly off the shelves. I think a large part of the reason is the price point: they're at or below the $100 magic mark. You'll most likely find the same isn't happening with Win2K or MS Office. Go ask your Electronic Boutique, Best Buys, CompuUSA etc, they'll most likely tell you that those sales completely pale in comparison to Win9x/Me upgrades.
Speaking of MSDN subscriptions, that's one of the HUGE copy-for-home-use candidates in the developer community. Considering the price point, that's not really surprising.
The fact is, most people probably DON'T buy MS software, they swipe it one way or another. If my last few jobs at LARGE companies are any indication of the general state of piracy, most people get their software either from work or friends. Or installed on new machines, but most people refresh their software versions way more frequently than their machines. Considering that the software on many people's machines would amount to over a thousand dollars if purchased legally, there's no way people would actually shell out that kind of money. If people actually had to plonk down the $300 or so for MS Office, there's no doubt in my mind that MS Works or Star Office would be the market leaders.
So in a perverted sense, while MS fights piracy with all their might, their brainshare and desktop share is due in large part to soft piracy. In that light, it doesn't matter how nasty Microsoft's image gets, since a lot of people don't pay for the software anyway, they don't mind swiping it regardless of MS's bad PR.
I see absolutely no disadvantage to being able to layout the application GIU visually, rather than creating widgets in code and constantly having to fool with the metrics and go through endless fiddle/compile/run cycles until it looks right. In particular a nice RAD IDE like Delphi makes life very easy, with decent layout control like Align and Anchors, though they could add a few more options such as percent of window.
.NET or I'll have to hunt you down and decompose you into little bits.
I used to be a C++ zealot until about 1997 when I got the first taste of Delphi 2, and I never went back. Sure, I had to switch languages, but I've made bigger sacrifices before. Besides, OP leads to a clarity and readability of code rarely achieved nowadays with C++, expecially when dealing with COM.
RAD from the perspective of Delphi gives you the best of both worlds: easy GUI design, as well as a great class framework and a powerful language. It ties the two together in such a slick and unobtrusive way; it's the way I'd probably do things even if Delphi didn't exist. Sure it adds some overhead, but then every framework does--even MFC, if you can call it a framework. If you want leaner apps, you pretty much have to drop to the API/message loop level, and we've all been there and didn't like it. Compared to something like VC++ that gives you RAD Light and forces you to keep your GUI and event handlers synchronized manually and always keeps you worrying about resource IDs, Delphi takes you to a whole other level of productivity.
The project manager at the company I used to work for was a great MS zealot, and his big mission in life was to convert the shop from Delphi to VC++. We were five developers churning out the code of ten and still were constantly behind schedule (talk about over-commitment of sales). We pleaded with him repeately to reconsider, explaining what a great work multiplier Delphi is and how such things are important to a small company. His retort was always that VC++ is the tool of the future and that tons of very smart guys work at MS and they can't all be wrong. He constantly denied that there were any productivity differences between Delphi and VC++ and accused us of just being irrational Delphi zealots (which probably was true, but with good reason we felt).
Anyway, these kinds of attitudes--influenced more by the great MS marketing machine than reality--are what is keeping Delphi down. I guess all of us Delphi supporters are hoping that Kylix will be the second wind for Delphi. With the momentum Linux has at the moment, and its lack of a serious RAD tool, Delphi should stand a decent second chance. And please don't anyone mention
If they keep bungling things up like they do, they'll probably be closing their doors by year end. Bye bye USB CueCat.
> Be wary of placing too much faith on the spelling of Slashdot stories.
Well, I took advantage of the opportunity and ran with it. It would be a very fitting name anyway.
> [...] and I think Oregonians talk funny, too.
He, he, and that coming from a Southerner. Go figger...