Just on a side note, while certain industries certainly use fancy hardware and CAD software, the bread and butter of CAD is done in AutoCAD on Windows, on Intel. While people certainly have a wide pallete of opinions on AutoCAD, you're much more likely to walk into a large CAD shop and find AutoCAD on an underpowered PC than anything else. Sad but true.
I worked for four years at TVA, a huge AutoCAD customer. You'd cry if you saw the PCs some power stations are designed on. You'd cry even more if you saw the people doing the designing. When I showed them that Windows can do more than 640x480, they'd look at me: who the hell can read that tiny type? No siree, they'll run AutoCAD at 640x480 with all tool bars and status bars on, and an effective design area of 100x100 pixels.
You're right on about computer illiteracy amongst CAD operators. Many of the guys were EEs and used CAD for the last 10 years, but they didn't even know rudimentary file management. One of the biggest software hits I wrote for them was a utility that completely shields them from the file system. All they need to know is the drawing name and what they want to do with it: plot it, delete it, or open it in AutoCAD. I'm afraid these kinds of users probably are the majority. While there certainly might be a market for CAD on Linux, these people ain't it. That's why I say that Linux hasn't got a paryer at becoming a leading CAD platform.
You're definitely right about the usefulness of legacy technologies, at least in the interim 10 years or so (:-) until even your wrist watch is USB. However, I believe that a better approach in the longer term would be to integrate legacy technologies via the new "scalable" technologies. I've played a bit with the new USB-to-serial converters, and they just plain work. Of course you need an OS that allows you to virtualize COM ports, but once you do, you're not limited in the number of legacy devices you can have. You could for example have 80 or so of these USB-to-serial devices in your system (well, 127 minus all the hubs required for all these ports). It would be interesting to do some stress testing to see how many 56k or so data streams you can REALLY run in parallel, since the theoretical USB bandwidth certainly would indicate A LOT.
Anyway, what I'm saying is move the serial ports away from IRQs 3/4 and virtualize them. Similarly, move the LPT ports from IRQs 5/7 and vitualize them. Heck, if somebody could come up with a 1394 device that simply contained 5 or so ISA slots and somehow virtualized those, that would be killer. Yeah, I realize that not all apps can use these kinds of "legacy" devices--in particular older DOS apps that write straight to the serial hardware, but I think that a huge section of the market could still be served.
I fact, with this move to legacy-free computers, I believe some of the biggest killer USB devices will be precisely serial and parallel converters. People simpy have too much money invested in hardware already to just throw it out and buy again. Of course, these converter devices will also have to become much cheaper to make it worthwhile to stick with the old stuff. Buying a $60 serial converter to keep a $70 external modem doesn't make much sense.
Professional designers don't switch CAD packages at a whim, because of new features or a new OS or what have you. People that don't use CAD professionally don't realize the skillset invested in a particular product--much more so than with most other software. In a way it's similar to how people use PhotoShop or some high-end 3D package. The proficiency in using a particular package is much more important to productivity than any potential benefit gained by switching, except in very rare cases, once a decade or less.
Besides, there's another aspect of CAD that precludes switching: scripts. Any serious AutoCAD user has a hard drive full of AutoLISP scripts (or even compiled C programs) that do most of the meat of production: insert entire subsections into a drawing, calculate complicated metrics, adjust scales etc. AutoCAD shops often hire people just to develop these scripts and have a lot of money invested in them. Switching CAD platforms requires you to pretty much throw out all this stuff.
No, in the real world people wait for their CAD vendor to offer all the new features of the competitors, even if it takes several versions. There's a product loyalty like in few other fields. The only people that switch CAD platforms at will are casual users like students or weekend designers who have nothing much invested in one product or another.
So far the most obvious examples are of exactly the opposite. Besides, US business used to be so isolationist earlier in the century, Europeans hardly had anything to fear.
This falls into the same category as "if you don't like this country/city/place why don't you leave?" The point is not the leaving, but the changing of a broken system.
By being born into this world--something I wasn't ask about, take note!--I have been endowed with certain rights that the constitution etc grant me. "Venturing outside my house" as you put it is one of those inalienable rights. The point people are trying to make is precisely that I should be able to "venture outside the house" without any ill effects.
You sound an awful lot like Scott McNealy, a person devoid of any notion of personal privacy.
Oh yeah. Sony is just a mountain of credibility in software. Their copious expertise in entertainment equipment has magically endowed them with the supernatural ability to write rock solid OS and end user software in no time at all. Lest we forget, Sony was a laughing stock in the computer business till recently, and still hasn't done much to exhonerate itself. Oh, yeah, the VAIO. And the PS1. Don't make me convulse in laughter. By the time Sony finally ships that damn brain buster, mainstream PC video cards will be dancing in the same ring...like a butterfly!
Idem, I also disagree. To make an analogy, look at other fields of science. Since WWII the US has won by far the most Nobel Prizes. These are all brownie points for US prestige, and nobody else. Yet when you check out the details, you'll find that a high percentage of the recipients weren't actually American by birth. Many immigrated late in life, for professional reasons. Others retained their respective citizenships and imparted some of the glory to their country of origin, and yet the US institution at which they worked--and indirectly the US--got much of the attention. Take this year's winner of the Nobel Price in Medicine: sure, he was born and educated in Gemany, but he researched for Stanford or whatever, and that's who in the end gets the fame. Who even knows which fine German institution imparted him his love of science?
This is important because fame brings money, and in the case of a university, lots of applicants. By flocking to the US, scientists are certainly bettering their careers, but they usually contribute little to their country of origin.
I'll agree in principle with most of what you say, except paragraph 4. And 2 and 3. I think impressions about earning potential in the US are colored by a few high profile Überearners. If you compare yearly average earning figures in IT I think you won't find crass differences to western Europe. It would be more accurate to compare disposable incomes, and I think you'll find differences are negligible. There must be a reason Germans travel so much...
Regarding the terms "cultural climate" and "America" in the same sentence, that's a bit of an oximoron, with a few notable exceptions. Don't confuse a social government with cultural conformism. Culturally the US are far less diverse than most countries, despite being a melting pot. Beyond a few exceptions such as NY and SF, American society is highly homogenized, indeed intolerant of cultural deviation. A foreigner travelling from St. Louis to Kansas City, to Detroit, to Dallas, to Houston etcetera will be hardpressed to even tell one from the other: people look and sound the same, buildings look the same, streets look the same, cities are laid out the same way. The same restaurants over and over again, the same shops, the same cinemas, the same ads on TV, the same news anchors. My opinion is that while a very large country has great economic potential through huge internal markets, it also leads to cultural homogenization and death. Frankly, the US can be pretty boring after you've seen the sights. There are simply no significant regional differences to make you come back again and again. That's what I love about Europe: you travel 1000 miles and you cross several countries, learn completely new histories, see vastly different architecture, meet people so different you wont't understand a word they say, eat highly varying foods, see different cars, hear different music, smell different air. Travel 1000 miles in the US, and you cross maybe two large cities. And you might not even be able to tell them apart.
> b. You are from outside the US, in which case you are being quite hypocritical
Hey, I'm from TN, Germany, just south of Munich. That's Tennissee, or Tennis Lake in English; it's a well known fact that tennis is very popular around here.
Seriously though, I'll give you that much, it does look very hypocritical indeed. But it is not. There are exculpatory circumstances. Frankly, I needed to make more money than I could back in Germany, and I really prefer dollars over those sissy Deutschmarks--or heaven forbid, Euros (what is that, a Greek food?)--so I simply HAD to come over here. Don't you see?
Even more seriously now, I was abducted by aliens (ok, my parents) and brought here at a time when I couldn't raise much legal resistance--i.e. I was a minor. You see, my parents, being the age that they are, grew up in a world in which the US were the promised land, the land of milk and honey (or whatever the metaphor of Paradise in your favorite religion may be). They did not understand that it also was the land of Bill Gates and the Antichrist, but then again, they're not much into Open Source. They're more of a patriarchal mindset, as befits their generation.
So as you see, I've been framed into living a life I can't really defend. I admit it, I'm a fraud. I'll cease posting on/. and I certainly will moderate no more! I have donned sack cloth and ashes and am on my way to a monastery as I type.
I fully agree about the financially ultra-conservative nature of European business. Risk-taking seems to have flourished in the US more than anywhere else. Pity Germany hasn't been as economically risk-taking as politically, twice this century, he, he. Still, it looks like things are improving. There's a great article on entrepreneurship in the New Germany in a recent Fortune magazine (August 2, 1999, p. 129). It seems that there is a new generation of entrepreneurs coming up--especially in the East, figures!--and big money in Europe is slowly waking up to them. We'll see.
Still, I'm ambivolent about VC. It seems to foster a quick-buck mentality and loose ethics. Lots of money up front makes twenty-somethings rush out and buy Ferraris, and lose focus before they've done anything truly worthwhile. If the business then fails, hey, it's not their money. VC seems to remove the incentive to succeed beyond that initial rush of fame and cash.
I find it very disturbing when people feel they need to be in the US to write real code, or to get a real job in IT. This trend just pisses me off. Particularly Europeans tend to be starry-eyed about the US, and of these, particularly Germans--who can count the legions of programmers who left for the supposed "greener pastures." Yes, there may be shitloads of VC in the US, but then, I'm no fan of VC either.
While it is said that countries don't matter anymore and that we live in a global economy, whom are they kidding? The US is sure to get the glory for any nugget of code they hatch, which certainly cannot be said for other places. Each "prominent" coder leaving his or her country for the US makes a little bit more cetain that IT in that country becomes that much less glamorous for the next generation.
> Nobody wants or needs to run multiple apps on a PDA at once.
We've heard that argument on the desktop in the '80s; it was wrong then, and it still is wrong today. While you might not feel you need multiple windowed apps visible at once, you very well need all sorts of background processes. Once you start differentiating between "real apps" and "background processes" you paint yourself into the Mac Desktop Accessory corner. After establishing that your device really MUST do more than one thing at once, do the right thing and multitask properly. It can be done very efficiently in little memory. Multitasking isn't what bloats OSs.
I think a lot of people confuse lots of onscreen windows--a busy interface--with multitasking. You can still have a simple, highly focused GUI on top of an agressively multitasking OS.
> Nobody want's or needs a "voice recorder".
Which "nobody" is that? Certainly not this one. That's the one feature I hate the Handspring Visor for not having. Just because WinCE devices have recording and are bloated doesn't mean recording implies bloat--except for the created data files, I guess. I often find myself in traffic trying to graffity in a reminder on the Pilot; a nicely integrated voice recorder would be SO much nicer. Depends all on how you integrate it into the whole.
> With DSL, though, I think everything is built for 24/7 connectivity...
Well, there's nothing to prevent the telcos from still metering you artificially. Deutsche Telekom in Berlin offers METERED ADSL access, where they concocted some sort of packet metering system. Purely artificial, purely greed! Companies used to a metered charge model simply can't let go--it's like giving the goose that lay the golden eggs to the bum at the corner.
I beg to differ there, if you check carefully I think you'll find that IE5 implements more of the useful stuff in the latest HTML4 and DHTML specs. Especially the new text control in DHTML often works more correctly and predictably under IE. Of course, neither browser deserves too much credit, since full support is still far off.
I'm mainly using IE5 simply because it feels faster on my machines. I've used both Netscape and IE since versions 3, and IE has always been snappier as far as I'm concerned. True, IE5 has the extremely irritating disappearing BACK button feature, which sends me up the wall. Still, not enough reason to switch to Netscape. Besides, OE5 is head and shoulders above Messenger, which can't even check multiple POP3 accounts. So unless I feel that I'm truly missing out by not using Netscape, I'll poop on it.
>> Does Linux keep track of sockets opened by an >> application and clean them up if it croacks?
> Of course it does. Managing resources is one of > the fundamental points of an Operating System. > If you can't do that right you might as well go > home and give up.
Of course it's a theoretical duty of an OS to keep track of such stuff. I guess my question should have been "does Linux do so successfully", at least more successfully than NT? Because if you ask MS, I'm sure they'll tell you that NT does so as well. Just not very well, I guess...
I'm running NT4 SP4 at work, and I never switch off the machine. Depending on what I currently work on, I can go for weeks without rebooting. In such cases I typically only have to reboot for moronic software installs that demand so.
On the other hand, I'm doing a lot of sockets programming lately. Boy, have I discovered NT4 weaknesses. There's nothing like debugging TCP/IP stuff to bring NT4 down. If you don't release sockets properly, pretty soon you'll get the "Insufficient buffer space..." error and no TCP/IP app will work anymore. Hello reboot...
To be fair, I don't know how other OSs handle this scenario. Does Linux keep track of sockets opened by an application and clean them up if it croacks? I don't know.
Incidentally, when I have to debug code that requires frequent reboots, I very much prefer WinLite (95/98) over NT. While they crash even easier, they also boot much quicker. I can do about 5 Edit/Run/Crash/Reboot cycles on 95/98 for every two such cycles on NT.
Along with the big words regarding Linux support, how about some real action? Like opening up the specs of some of their hardware so people can write Linux drivers--or at least write the drivers themselves. I have a bunch of their Wireless LAN Entry cards which work well enough under Windows (yes, even 98), but are no go under Linux. They were discontinued something like 2 years ago, but they still won't release the specs. And that's just one example, there are many more.
You might be entirely right. I was simply offering some pure conjecture, nothing more. I was hoping for something more truly revolutionary than what it might end up being. After all, if all they're doing is software emulation, that's been done many times before, sometimes rather well. For example, Digital's x86-to-Alpha translation system (I forgot the name, something with "W" and "X" and "32" in it I believe) which saves translated code and uses that one the next time around is quite brilliant and works very well from what I've heard.
I guess I'm simply hoping for someone to provide that decisive push to get more parallel computing into the mainstream. I used to root for the BeOS, until they went the OTHER way, supporting single CPU architectures. The current CPU upgrade path is just some lame and tired; every new Pentium generation gives another few percentage points of improvement in doing the same old stuff. I can hardly wait for the Merced--NOT.
> [...] a design to emulate a set of gates generally > uses more gates than the design it's emulating
That's precisely the point I was making.
> Not to mention the cost!!! Per gate, FPGA's > are more expensive to manufacture [...]
Cost is a relative issue. Once a technology is proven viable and desirable, it can undergo iterations of streamlining, simplification, optimization, cheapification etc to reach a price point. Don't forget economies of scale. Still, a first generation product COULD be costly.
There are many ways performance and cost could be optimized. By its very nature, an FPGA does not have to implement an entire CPU at once. It can selectively model the target CPU based on the instruction stream. Subsystems on a CPU often lie idle while others are performing work. An FPGA basically allows you to "swap in" CPU subsystems as needed. Of course, this interacts with pipelining etc in strange and wondrous ways, and is by no means an easy problem to solve. If it were, it would have been done a long time ago. Basically, a lot of the same issues of parallel processing have to be addressed.
Still, Transmeta might just be the company that solved all the problems. Of course, I could also be way off, and they might be working on something entirely different. Like a Super-Aibo.
That emulation of the x86 instructions is involved was pretty obvious from the patent document itself. The question is HOW is it done, and no more light has been shed on that.
Personally, I believe they're working on a meta-CPU which doesn't have a native general purpose instruction set of its own, but rather can be programmed to impersonate any number of processors on the fly. This makes sense particularly in light of the rising popularity of FPGA-type approaches to computing. The recent announcement by a company (I forgot the name) of a "mainframe on the desktop" (no, it wasn't Apple!) which is based on 256 or so FPGAs is typical of the trend. Scientific American in its previous issue had several articles about research at MIT into future computing devices, particularly handheld units containing FPGA-type hardware that is reconfigured dynamically to perform whatever functionality the current task requires.
If they strike the pot of gold and do it right--if indeed it can be successfully done--the implications are sublime. Imagine having a PC that can "multitask" being a Pentium, a MIPS, an Alpha, a PPC, a Z80, etc. The processing unit constantly time slices between the different CPU "emulations", each of which in turn is running some operating system. This wouldn't be real emulation in the sense of translating foreign instructions to a native set, incurring the necessay slowdown, but rather instruction decoding in hardware.
The main problem I see is the relatively low logic density of FPGAs as compared to dedicated hardware. Implementing an ALU on an FPGA would require considerably more logic gates and real estate, simply because those gates are general purpose and can also do lots of other things besides being part of an ALU. In addition, having more gates implies more propagation delays, longer signal paths, etc etc. The real gem would be to come up with a strategy or technology which can resolve these issues while still providing all the benefits of a truly generic "CPU".
Uh, correction. You're the only superpower left because the other one shot itself in the head. Sure, winning by default is still winning, but it's not something to gloat about.
Besides, you're only a superpower because most of the rest of the world doesn't mind it. In fact, they're quite happy for the US to shell out the money for their fleets of carriers and subs to police the world. Particularly Europe is fond of their trans-Atlantic "partner" to do the things it ought to take care of itself.
The US could only fight about 2-3 major wars at once by their own reckoning. This means that they can't seriously piss off more than about 2-3 influential entities. As long as they behave in ways that Europe, Russia and Asia (the main other military powers) don't mind too much, things are fine. Within these confines, the US can do "anything they want", which isn't really saying too much.
If they REALLY wanted to take advantage of their supremacy in a spit-in-your-face way that was meant to show THE WHOLE WORLD that they can't do anything about it, the 2-3 war limit would quickly be reached. And isn't the proof of the pudding in the eating?
Say the US decided to annex Mexico and Canada, so that they own the whole North American continent. This would piss off Russia and China majorly, and even prissy Europe wouldn't just take it. Tri-lateral hostilities against the US would almost certainly ensue. Contending that the US could win this conflict borders on the grotesque.
Of course, contending that such a thing could ever take place is equally grotesque. However, such incidents have taken place many times in history: the Babylonians, the Romans, the Ottomans, the French under the Napoleons, Germany, who all conquered very substantial parts of their respective world. They also (save maybe for the Romans) showed that it's impossible to really win in this game. However, they PROVED that they were the biggest kid on the block for a while. In order for the US to claim the same--as they do--they would have to engage in similar activities.
Which of course they won't, much to their credit. However, as long as they don't, their claims of supremacy are nothing more than dick waving. So maybe they should stop.
Well, the Telia web site isn't terribly useful, since my grasp of Swedish is tenuous at best--basically anything that is close to German or English words. Heh, some sentences can sound like German dialect almost ("Läs mer" for I guess Read More).
Anyway, went to www.gsmworld.com and they have links to most global providers. According to them, Telia's coverage is as follows: http://www.gsmworld.com/gsminfo/cov_sete.htm. Looks like maybe 80%, but still, I'm impressed. And no, I never said such coverage would be silly, just economically unlikely. But if they see a market, hey, more power to them. Just to prove my point that coverage != market penetration, here's a map for D2 coverage in Germany: http://www.gsmworld.com/gsminfo/cov_dema.htm. Basically only my aunt's outhouse in Baden Wuerttemberg isn't covered, and they're working on that. Still, probably more Swedes use cell phones.
Here in the US, GTE Wireless I guess doesn't even see a market outside Chattanooga. My digital (and analog, for that matter) coverage stops halfway on the way home, which is in an outer "suburb". Far from the Middle Of Nowhere(TM), but obviously not important enough. Heck, they don't even provide SMS or paging, even though the phones they sell certainly are equipped to. So Much for the US being market driven.
Yes I have, but I don't have to. Since 80% or more of the population lives in a few urban hives, it doesn't matter how bloody big the country is. It makes no economic sense to transform Lapland into a veritable porcupine of towers.
The GSM 900 MHz fiasco is this: it happens to be a very nasty frequency to be using around people with--say--pacemakers. Which is why few hospitals in Europe permit their use indoors--certainly true in Germany. Natch to all those snobbish StarTac-carrying doctors. 1800+ MHz doesn't seem to have this problem, hence the move away from 900 MHz, regardless of your impressions. There's also the matter of bandwith--the higher the frequency, the more simultaneous conversations, the more users per cell, the more money for Mr. cell phone service provider.
Well, "fiasco" might have been a strong word. Overall, Europe still did the right thing, but it would have been handy to forego the 900 MHz step. The new UMTS standard works above 2 GHz, so that's the way it's going.
I meant that as a joke--observe the "he, he". However, I have a VERY hard time believing the almost 100% coverage in Sweden. It's a big country by European standards, with a population of less than 10M. Most of them are concentrated in several large cities. The population density in the backwoods approaches a total vacuum. On the other hand, Germany has a pop. density not unlike a can of sardines, yet they still don't boast 100% coverage. So I do have a hard time believing the Sweden figures. Keep in mind here, we're talking COVERAGE, not market penetration. Yes, more Swedes might use cell phones, but that doesn't imply wider coverage. Why would they put a tower up somewhere in Lapland, so that Olaf can talk to his sweetiepie on the phone while ice fishing on the john?
Slashdot Mirror
Just on a side note, while certain industries certainly use fancy hardware and CAD software, the bread and butter of CAD is done in AutoCAD on Windows, on Intel. While people certainly have a wide pallete of opinions on AutoCAD, you're much more likely to walk into a large CAD shop and find AutoCAD on an underpowered PC than anything else. Sad but true.
I worked for four years at TVA, a huge AutoCAD customer. You'd cry if you saw the PCs some power stations are designed on. You'd cry even more if you saw the people doing the designing. When I showed them that Windows can do more than 640x480, they'd look at me: who the hell can read that tiny type? No siree, they'll run AutoCAD at 640x480 with all tool bars and status bars on, and an effective design area of 100x100 pixels.
You're right on about computer illiteracy amongst CAD operators. Many of the guys were EEs and used CAD for the last 10 years, but they didn't even know rudimentary file management. One of the biggest software hits I wrote for them was a utility that completely shields them from the file system. All they need to know is the drawing name and what they want to do with it: plot it, delete it, or open it in AutoCAD. I'm afraid these kinds of users probably are the majority. While there certainly might be a market for CAD on Linux, these people ain't it. That's why I say that Linux hasn't got a paryer at becoming a leading CAD platform.
You're definitely right about the usefulness of legacy technologies, at least in the interim 10 years or so (:-) until even your wrist watch is USB. However, I believe that a better approach in the longer term would be to integrate legacy technologies via the new "scalable" technologies. I've played a bit with the new USB-to-serial converters, and they just plain work. Of course you need an OS that allows you to virtualize COM ports, but once you do, you're not limited in the number of legacy devices you can have. You could for example have 80 or so of these USB-to-serial devices in your system (well, 127 minus all the hubs required for all these ports). It would be interesting to do some stress testing to see how many 56k or so data streams you can REALLY run in parallel, since the theoretical USB bandwidth certainly would indicate A LOT.
Anyway, what I'm saying is move the serial ports away from IRQs 3/4 and virtualize them. Similarly, move the LPT ports from IRQs 5/7 and vitualize them. Heck, if somebody could come up with a 1394 device that simply contained 5 or so ISA slots and somehow virtualized those, that would be killer. Yeah, I realize that not all apps can use these kinds of "legacy" devices--in particular older DOS apps that write straight to the serial hardware, but I think that a huge section of the market could still be served.
I fact, with this move to legacy-free computers, I believe some of the biggest killer USB devices will be precisely serial and parallel converters. People simpy have too much money invested in hardware already to just throw it out and buy again. Of course, these converter devices will also have to become much cheaper to make it worthwhile to stick with the old stuff. Buying a $60 serial converter to keep a $70 external modem doesn't make much sense.
Professional designers don't switch CAD packages at a whim, because of new features or a new OS or what have you. People that don't use CAD professionally don't realize the skillset invested in a particular product--much more so than with most other software. In a way it's similar to how people use PhotoShop or some high-end 3D package. The proficiency in using a particular package is much more important to productivity than any potential benefit gained by switching, except in very rare cases, once a decade or less.
Besides, there's another aspect of CAD that precludes switching: scripts. Any serious AutoCAD user has a hard drive full of AutoLISP scripts (or even compiled C programs) that do most of the meat of production: insert entire subsections into a drawing, calculate complicated metrics, adjust scales etc. AutoCAD shops often hire people just to develop these scripts and have a lot of money invested in them. Switching CAD platforms requires you to pretty much throw out all this stuff.
No, in the real world people wait for their CAD vendor to offer all the new features of the competitors, even if it takes several versions. There's a product loyalty like in few other fields. The only people that switch CAD platforms at will are casual users like students or weekend designers who have nothing much invested in one product or another.
So far the most obvious examples are of exactly the opposite. Besides, US business used to be so isolationist earlier in the century, Europeans hardly had anything to fear.
Point taken. Your argument however in no way lessens mine, but rather strengthens it.
This falls into the same category as "if you don't like this country/city/place why don't you leave?" The point is not the leaving, but the changing of a broken system.
By being born into this world--something I wasn't ask about, take note!--I have been endowed with certain rights that the constitution etc grant me. "Venturing outside my house" as you put it is one of those inalienable rights. The point people are trying to make is precisely that I should be able to "venture outside the house" without any ill effects.
You sound an awful lot like Scott McNealy, a person devoid of any notion of personal privacy.
Oh yeah. Sony is just a mountain of credibility in software. Their copious expertise in entertainment equipment has magically endowed them with the supernatural ability to write rock solid OS and end user software in no time at all. Lest we forget, Sony was a laughing stock in the computer business till recently, and still hasn't done much to exhonerate itself. Oh, yeah, the VAIO. And the PS1. Don't make me convulse in laughter. By the time Sony finally ships that damn brain buster, mainstream PC video cards will be dancing in the same ring...like a butterfly!
Idem, I also disagree. To make an analogy, look at other fields of science. Since WWII the US has won by far the most Nobel Prizes. These are all brownie points for US prestige, and nobody else. Yet when you check out the details, you'll find that a high percentage of the recipients weren't actually American by birth. Many immigrated late in life, for professional reasons. Others retained their respective citizenships and imparted some of the glory to their country of origin, and yet the US institution at which they worked--and indirectly the US--got much of the attention. Take this year's winner of the Nobel Price in Medicine: sure, he was born and educated in Gemany, but he researched for Stanford or whatever, and that's who in the end gets the fame. Who even knows which fine German institution imparted him his love of science?
This is important because fame brings money, and in the case of a university, lots of applicants. By flocking to the US, scientists are certainly bettering their careers, but they usually contribute little to their country of origin.
I'll agree in principle with most of what you say, except paragraph 4. And 2 and 3. I think impressions about earning potential in the US are colored by a few high profile Überearners. If you compare yearly average earning figures in IT I think you won't find crass differences to western Europe. It would be more accurate to compare disposable incomes, and I think you'll find differences are negligible. There must be a reason Germans travel so much...
Regarding the terms "cultural climate" and "America" in the same sentence, that's a bit of an oximoron, with a few notable exceptions. Don't confuse a social government with cultural conformism. Culturally the US are far less diverse than most countries, despite being a melting pot. Beyond a few exceptions such as NY and SF, American society is highly homogenized, indeed intolerant of cultural deviation. A foreigner travelling from St. Louis to Kansas City, to Detroit, to Dallas, to Houston etcetera will be hardpressed to even tell one from the other: people look and sound the same, buildings look the same, streets look the same, cities are laid out the same way. The same restaurants over and over again, the same shops, the same cinemas, the same ads on TV, the same news anchors. My opinion is that while a very large country has great economic potential through huge internal markets, it also leads to cultural homogenization and death. Frankly, the US can be pretty boring after you've seen the sights. There are simply no significant regional differences to make you come back again and again. That's what I love about Europe: you travel 1000 miles and you cross several countries, learn completely new histories, see vastly different architecture, meet people so different you wont't understand a word they say, eat highly varying foods, see different cars, hear different music, smell different air. Travel 1000 miles in the US, and you cross maybe two large cities. And you might not even be able to tell them apart.
> b. You are from outside the US, in which case you are being quite hypocritical
/. and I certainly will moderate no more! I have donned sack cloth and ashes and am on my way to a monastery as I type.
Hey, I'm from TN, Germany, just south of Munich. That's Tennissee, or Tennis Lake in English; it's a well known fact that tennis is very popular around here.
Seriously though, I'll give you that much, it does look very hypocritical indeed. But it is not. There are exculpatory circumstances. Frankly, I needed to make more money than I could back in Germany, and I really prefer dollars over those sissy Deutschmarks--or heaven forbid, Euros (what is that, a Greek food?)--so I simply HAD to come over here. Don't you see?
Even more seriously now, I was abducted by aliens (ok, my parents) and brought here at a time when I couldn't raise much legal resistance--i.e. I was a minor. You see, my parents, being the age that they are, grew up in a world in which the US were the promised land, the land of milk and honey (or whatever the metaphor of Paradise in your favorite religion may be). They did not understand that it also was the land of Bill Gates and the Antichrist, but then again, they're not much into Open Source. They're more of a patriarchal mindset, as befits their generation.
So as you see, I've been framed into living a life I can't really defend. I admit it, I'm a fraud. I'll cease posting on
I fully agree about the financially ultra-conservative nature of European business. Risk-taking seems to have flourished in the US more than anywhere else. Pity Germany hasn't been as economically risk-taking as politically, twice this century, he, he. Still, it looks like things are improving. There's a great article on entrepreneurship in the New Germany in a recent Fortune magazine (August 2, 1999, p. 129). It seems that there is a new generation of entrepreneurs coming up--especially in the East, figures!--and big money in Europe is slowly waking up to them. We'll see.
Still, I'm ambivolent about VC. It seems to foster a quick-buck mentality and loose ethics. Lots of money up front makes twenty-somethings rush out and buy Ferraris, and lose focus before they've done anything truly worthwhile. If the business then fails, hey, it's not their money. VC seems to remove the incentive to succeed beyond that initial rush of fame and cash.
I find it very disturbing when people feel they need to be in the US to write real code, or to get a real job in IT. This trend just pisses me off. Particularly Europeans tend to be starry-eyed about the US, and of these, particularly Germans--who can count the legions of programmers who left for the supposed "greener pastures." Yes, there may be shitloads of VC in the US, but then, I'm no fan of VC either.
While it is said that countries don't matter anymore and that we live in a global economy, whom are they kidding? The US is sure to get the glory for any nugget of code they hatch, which certainly cannot be said for other places. Each "prominent" coder leaving his or her country for the US makes a little bit more cetain that IT in that country becomes that much less glamorous for the next generation.
Ok, just some venting here.
I agree with several of your points, but please:
> Nobody wants or needs to run multiple apps on a PDA at once.
We've heard that argument on the desktop in the '80s; it was wrong then, and it still is wrong today. While you might not feel you need multiple windowed apps visible at once, you very well need all sorts of background processes. Once you start differentiating between "real apps" and "background processes" you paint yourself into the Mac Desktop Accessory corner. After establishing that your device really MUST do more than one thing at once, do the right thing and multitask properly. It can be done very efficiently in little memory. Multitasking isn't what bloats OSs.
I think a lot of people confuse lots of onscreen windows--a busy interface--with multitasking. You can still have a simple, highly focused GUI on top of an agressively multitasking OS.
> Nobody want's or needs a "voice recorder".
Which "nobody" is that? Certainly not this one. That's the one feature I hate the Handspring Visor for not having. Just because WinCE devices have recording and are bloated doesn't mean recording implies bloat--except for the created data files, I guess. I often find myself in traffic trying to graffity in a reminder on the Pilot; a nicely integrated voice recorder would be SO much nicer. Depends all on how you integrate it into the whole.
> With DSL, though, I think everything is built for 24/7 connectivity...
Well, there's nothing to prevent the telcos from still metering you artificially. Deutsche Telekom in Berlin offers METERED ADSL access, where they concocted some sort of packet metering system. Purely artificial, purely greed! Companies used to a metered charge model simply can't let go--it's like giving the goose that lay the golden eggs to the bum at the corner.
I beg to differ there, if you check carefully I think you'll find that IE5 implements more of the useful stuff in the latest HTML4 and DHTML specs. Especially the new text control in DHTML often works more correctly and predictably under IE. Of course, neither browser deserves too much credit, since full support is still far off.
I'm mainly using IE5 simply because it feels faster on my machines. I've used both Netscape and IE since versions 3, and IE has always been snappier as far as I'm concerned. True, IE5 has the extremely irritating disappearing BACK button feature, which sends me up the wall. Still, not enough reason to switch to Netscape. Besides, OE5 is head and shoulders above Messenger, which can't even check multiple POP3 accounts. So unless I feel that I'm truly missing out by not using Netscape, I'll poop on it.
>> Does Linux keep track of sockets opened by an
>> application and clean them up if it croacks?
> Of course it does. Managing resources is one of
> the fundamental points of an Operating System.
> If you can't do that right you might as well go
> home and give up.
Of course it's a theoretical duty of an OS to keep track of such stuff. I guess my question should have been "does Linux do so successfully", at least more successfully than NT? Because if you ask MS, I'm sure they'll tell you that NT does so as well. Just not very well, I guess...
I'm running NT4 SP4 at work, and I never switch off the machine. Depending on what I currently work on, I can go for weeks without rebooting. In such cases I typically only have to reboot for moronic software installs that demand so.
On the other hand, I'm doing a lot of sockets programming lately. Boy, have I discovered NT4 weaknesses. There's nothing like debugging TCP/IP stuff to bring NT4 down. If you don't release sockets properly, pretty soon you'll get the "Insufficient buffer space..." error and no TCP/IP app will work anymore. Hello reboot...
To be fair, I don't know how other OSs handle this scenario. Does Linux keep track of sockets opened by an application and clean them up if it croacks? I don't know.
Incidentally, when I have to debug code that requires frequent reboots, I very much prefer WinLite (95/98) over NT. While they crash even easier, they also boot much quicker. I can do about 5 Edit/Run/Crash/Reboot cycles on 95/98 for every two such cycles on NT.
Along with the big words regarding Linux support, how about some real action? Like opening up the specs of some of their hardware so people can write Linux drivers--or at least write the drivers themselves. I have a bunch of their Wireless LAN Entry cards which work well enough under Windows (yes, even 98), but are no go under Linux. They were discontinued something like 2 years ago, but they still won't release the specs. And that's just one example, there are many more.
You might be entirely right. I was simply offering some pure conjecture, nothing more. I was hoping for something more truly revolutionary than what it might end up being. After all, if all they're doing is software emulation, that's been done many times before, sometimes rather well. For example, Digital's x86-to-Alpha translation system (I forgot the name, something with "W" and "X" and "32" in it I believe) which saves translated code and uses that one the next time around is quite brilliant and works very well from what I've heard.
I guess I'm simply hoping for someone to provide that decisive push to get more parallel computing into the mainstream. I used to root for the BeOS, until they went the OTHER way, supporting single CPU architectures. The current CPU upgrade path is just some lame and tired; every new Pentium generation gives another few percentage points of improvement in doing the same old stuff. I can hardly wait for the Merced--NOT.
> [...] a design to emulate a set of gates generally
> uses more gates than the design it's emulating
That's precisely the point I was making.
> Not to mention the cost!!! Per gate, FPGA's
> are more expensive to manufacture [...]
Cost is a relative issue. Once a technology is proven viable and desirable, it can undergo iterations of streamlining, simplification, optimization, cheapification etc to reach a price point. Don't forget economies of scale. Still, a first generation product COULD be costly.
There are many ways performance and cost could be optimized. By its very nature, an FPGA does not have to implement an entire CPU at once. It can selectively model the target CPU based on the instruction stream. Subsystems on a CPU often lie idle while others are performing work. An FPGA basically allows you to "swap in" CPU subsystems as needed. Of course, this interacts with pipelining etc in strange and wondrous ways, and is by no means an easy problem to solve. If it were, it would have been done a long time ago. Basically, a lot of the same issues of parallel processing have to be addressed.
Still, Transmeta might just be the company that solved all the problems. Of course, I could also be way off, and they might be working on something entirely different. Like a Super-Aibo.
That emulation of the x86 instructions is involved was pretty obvious from the patent document itself. The question is HOW is it done, and no more light has been shed on that.
Personally, I believe they're working on a meta-CPU which doesn't have a native general purpose instruction set of its own, but rather can be programmed to impersonate any number of processors on the fly. This makes sense particularly in light of the rising popularity of FPGA-type approaches to computing. The recent announcement by a company (I forgot the name) of a "mainframe on the desktop" (no, it wasn't Apple!) which is based on 256 or so FPGAs is typical of the trend. Scientific American in its previous issue had several articles about research at MIT into future computing devices, particularly handheld units containing FPGA-type hardware that is reconfigured dynamically to perform whatever functionality the current task requires.
If they strike the pot of gold and do it right--if indeed it can be successfully done--the implications are sublime. Imagine having a PC that can "multitask" being a Pentium, a MIPS, an Alpha, a PPC, a Z80, etc. The processing unit constantly time slices between the different CPU "emulations", each of which in turn is running some operating system. This wouldn't be real emulation in the sense of translating foreign instructions to a native set, incurring the necessay slowdown, but rather instruction decoding in hardware.
The main problem I see is the relatively low logic density of FPGAs as compared to dedicated hardware. Implementing an ALU on an FPGA would require considerably more logic gates and real estate, simply because those gates are general purpose and can also do lots of other things besides being part of an ALU. In addition, having more gates implies more propagation delays, longer signal paths, etc etc. The real gem would be to come up with a strategy or technology which can resolve these issues while still providing all the benefits of a truly generic "CPU".
Uh, correction. You're the only superpower left because the other one shot itself in the head. Sure, winning by default is still winning, but it's not something to gloat about.
Besides, you're only a superpower because most of the rest of the world doesn't mind it. In fact, they're quite happy for the US to shell out the money for their fleets of carriers and subs to police the world. Particularly Europe is fond of their trans-Atlantic "partner" to do the things it ought to take care of itself.
The US could only fight about 2-3 major wars at once by their own reckoning. This means that they can't seriously piss off more than about 2-3 influential entities. As long as they behave in ways that Europe, Russia and Asia (the main other military powers) don't mind too much, things are fine. Within these confines, the US can do "anything they want", which isn't really saying too much.
If they REALLY wanted to take advantage of their supremacy in a spit-in-your-face way that was meant to show THE WHOLE WORLD that they can't do anything about it, the 2-3 war limit would quickly be reached. And isn't the proof of the pudding in the eating?
Say the US decided to annex Mexico and Canada, so that they own the whole North American continent. This would piss off Russia and China majorly, and even prissy Europe wouldn't just take it. Tri-lateral hostilities against the US would almost certainly ensue. Contending that the US could win this conflict borders on the grotesque.
Of course, contending that such a thing could ever take place is equally grotesque. However, such incidents have taken place many times in history: the Babylonians, the Romans, the Ottomans, the French under the Napoleons, Germany, who all conquered very substantial parts of their respective world. They also (save maybe for the Romans) showed that it's impossible to really win in this game. However, they PROVED that they were the biggest kid on the block for a while. In order for the US to claim the same--as they do--they would have to engage in similar activities.
Which of course they won't, much to their credit. However, as long as they don't, their claims of supremacy are nothing more than dick waving. So maybe they should stop.
Well, the Telia web site isn't terribly useful, since my grasp of Swedish is tenuous at best--basically anything that is close to German or English words. Heh, some sentences can sound like German dialect almost ("Läs mer" for I guess Read More).
Anyway, went to www.gsmworld.com and they have links to most global providers. According to them, Telia's coverage is as follows: http://www.gsmworld.com/gsminfo/cov_sete.htm. Looks like maybe 80%, but still, I'm impressed. And no, I never said such coverage would be silly, just economically unlikely. But if they see a market, hey, more power to them. Just to prove my point that coverage != market penetration, here's a map for D2 coverage in Germany: http://www.gsmworld.com/gsminfo/cov_dema.htm. Basically only my aunt's outhouse in Baden Wuerttemberg isn't covered, and they're working on that. Still, probably more Swedes use cell phones.
Here in the US, GTE Wireless I guess doesn't even see a market outside Chattanooga. My digital (and analog, for that matter) coverage stops halfway on the way home, which is in an outer "suburb". Far from the Middle Of Nowhere(TM), but obviously not important enough. Heck, they don't even provide SMS or paging, even though the phones they sell certainly are equipped to. So Much for the US being market driven.
Yes I have, but I don't have to. Since 80% or more of the population lives in a few urban hives, it doesn't matter how bloody big the country is. It makes no economic sense to transform Lapland into a veritable porcupine of towers.
The GSM 900 MHz fiasco is this: it happens to be a very nasty frequency to be using around people with--say--pacemakers. Which is why few hospitals in Europe permit their use indoors--certainly true in Germany. Natch to all those snobbish StarTac-carrying doctors. 1800+ MHz doesn't seem to have this problem, hence the move away from 900 MHz, regardless of your impressions. There's also the matter of bandwith--the higher the frequency, the more simultaneous conversations, the more users per cell, the more money for Mr. cell phone service provider.
Well, "fiasco" might have been a strong word. Overall, Europe still did the right thing, but it would have been handy to forego the 900 MHz step. The new UMTS standard works above 2 GHz, so that's the way it's going.
I meant that as a joke--observe the "he, he". However, I have a VERY hard time believing the almost 100% coverage in Sweden. It's a big country by European standards, with a population of less than 10M. Most of them are concentrated in several large cities. The population density in the backwoods approaches a total vacuum. On the other hand, Germany has a pop. density not unlike a can of sardines, yet they still don't boast 100% coverage. So I do have a hard time believing the Sweden figures. Keep in mind here, we're talking COVERAGE, not market penetration. Yes, more Swedes might use cell phones, but that doesn't imply wider coverage. Why would they put a tower up somewhere in Lapland, so that Olaf can talk to his sweetiepie on the phone while ice fishing on the john?