I don't know how things are today, I gave up on Java a long time ago, but back when it got started NOBODY had compatible JVMs. "Write once, run anywhere" wasn't anything but a clever marketing slogan, which is why the sardonic "write once, test everywhere" saying became famous. You really did have to test on 10 or 12 different browser/JVM/JIT/OS combinations before you could say your code ran cleanly under most VMs, and VM-specific work-arounds were fairly common.
It is important to realize that Microsoft extended Java, they did not intentionally break it. Well, I guess some people define that as breaking it, but I mean they didn't intentionally build their JVMs and JITs so that they did standard Java things incorrectly. Even Sun admitted the MS JVM was the best available implementation, I heard Sun guys saying it all the time at the first couple Java One events.
I road race my 2001 Viper. Modifications and maintenance take up some time, but not much, so I've also started building a 34 Ford chopped 5-window coupe with a blown Hemi 426 for drag racing (and street use, heh heh). And when I say "build" I really mean "build" -- I'm fabricating as much as possible in my garage -- the frame, suspension, all kinds of things. The stuff I don't fabricate still requires quite a bit of work to make it usable with the rest of the car. For example, the rear brakes will be my old stock front brake system from my Viper, and the driveline uses parts from no less than six different major auto manufacturers. It's kind of like assembling a giant 400HP jigsaw puzzle that was never originally one single piece...
Actually, an intelligent image search engine like that would be really cool.
IBM has something similar to this, which you can read about here...
Of course, you have to start with an image database that potentially contains that thing you found at a garage sale (or the chick you ran into at the mall).
Microsoft didn't kill it, early JVM/JIT performance and the still-low-average-bandwidth in the early days of Java are what killed it. I know, I used to write web-delivered client-side Java applets in the early days.
Locally stored Java apps were a non-starter from day-one due to the total lack of consistency -- you simply couldn't put an app out there and tell the end-user how to start it. Either they knew (and most people won't know), or they used someone else's software.
It is a well-known and well-documented fact that Microsoft had the best-performing JVM throughout most of Java's early life. If anything, they did the most to prop it up. (Though not out of any noble purpose, of course.)
Increasing the number of reps would only make it easier for the reps to avoid issues and increase their ability to get away with selling out to the highest bidder. In effect, you would dilute the amount of attention the media could focus on any given rep (and unfortunately, the media is really the only way most people find out what a rep is up to). Most people don't know who their current reps are -- imagine having to keep track of 8300 of those guys... or trying to decide who to elect, or re-elect...
Indeed -- one factor among many (the rest of my post).
Unless your backup is so precious that you have it in a sealed nitrogen-filled Swiss vault somewhere, perhaps. Look, I have literally hundreds of backup CDs (used to be in the multimedia biz) and the great majority of them are just fine (I grab a few beers and sit down to test-read all of them once a year or so)... but the fact is, some just die for no good reason. It has happened to me and others I know often enough that your claim of 50 years is weak at best.
But do whatever floats your boat, I was just trying to help.
Burn it to a DVD, stick it in a safe-deposit box, and you're good for at least 50 years.
Don't be so sure. A HUGE number of factors play into the lifespan of burned media. I've had reasonably high quality CDs (e.g. high cyan Ricoh stuff, not spindled 100-bulk no-name crap) go bad after only 5 years or so.
Not that I don't use it myself for backup -- I just make two backups. It's certainly cheap enough.
If the government buys Microsoft software, it will have to pay for software and ALSO have to pay for support only available from a single source and outside the country.
<sarcasm> Yeah, it's pretty hard to find people to support Microsoft software. </sarcasm>
I liked Project Gotham Racing, too, but I didn't play it nearly as long because there weren't many cars and locations
29 cars is pretty good, especially when you consider the quality of the 3D models and the fact that they all handle and perform very differently (e.g. they aren't just skins over a small handful of performance characteristics). As for locations, there are about 200 tracks in the game -- granted, they're all set in 15 or 20 areas of only 4 cities, but that's still a lot of variety.
I would think increasing realism is detrimental to sales mainly because cars are somewhat difficult to drive properly at speed. I like Gotham because I can use real-world braking and cornering techniques to improve my time and speed. Most people, though, just wouldn't know what to do.
I'm on the fence about upgrades. Generally their impact is blown way out of proportion to their real-life counterparts, which to me makes it less fun to pick and choose among your options. If everything is flat-out awesome, then it's just a matter of trying to collect as many of them as you can.
Big deal. All that old hardware still runs Windows just fine, too.
One P1-90 running NT4 (very usable, a dev SQL Server box in fact)
Two P1-100s running NT4 (very usable)
Two P1-166s running Win98 (webcams)
Two P2-266s running Win98 (webcams, MP3s)
One P3-500 running Win2K (alarm & home control)
One P3-700 running Win2K (MP3s, basement Quake3, DVD, web)
One 1GHz Althon running WinXP (MP3s, living-room Quake3, DVD, web)
Two P4-2.6s running XP (main PCs)
Mainly I use so many because of physical separation -- some upstairs, some downstairs at various places in the garages, etc.
And don't give me crap about stability. These machines do not crash, ever. The ONLY BSOD I've seen in YEARS was (coincidentally) yesterday, after trying the much-heralded BitTorrent for the first time. Apparently it doesn't like something about one of my LinkSys NIC drivers. Prior to that, I haven't had a non-hardware failure in at least two years. These machines are powered up and running 24x7, all on battery backup.
Could I do the same with Linux? Sure. Do I want to? Nope. I already owned all those copies of Windows (yeah, I actually paid for it, go figure), and in many cases it would be a royal bitch to do what I want under Linux (old webcams, alarm system & home control interfacing, DVD playback).
The surprising one, of course, is NT4 and SQL Server on a P90. That machine has 96MB of RAM and an old throw-away 6GB HDD, and has been in continuous usage since the mid-90's. I had a CD-ROM drive die in it, but other than that, it's just hummed along without a hitch.
I also have a 286 I've been thinking about powering up just for laughs. Not sure what I'd do with it yet, though.
Not only that, but I'm sure most people would spend the 50$ or so to reprogram the vehicles computer, since putting wrong size tires on your vehicle and NOT doing this will lead to horrible fuel mileage, acceleration, etc on any modern vehicle that relies on accurate info to forecast what will happen next
Most vehicles only support a very small range of settings. As an example, I put 33" tires on one of my trucks, but the engine computer can't be set for anything larger than 32" tires. Thus, my speedometer is always incorrect with that set of tires on it. (I do happen to also run a 31" set for towing.)
Also, current engine computers almost exclusively rely upon O2 sensors to control anything that would affect your mileage or acceleration or other performance characteristics. About the only major exception I can think of is the GM system that shuts down half of the engine during steady cruising, and that only requires the ability to recognize a steady speed.
Funny, I thought TMC did the comparison and slashdot just linked to it.
(No wonder they're so far behind on submissions, they're all busy in Slasdot Labs (TM) running the latest round of amazing new benchmarks. I can't wait to see the results!)
When people were building a software community of openness and sharing, he came along to poison the well
-1, Poster Smoking Crack
Your lollipops-and-sunshine view of the software world before Gates is one of the most ridiculous things I've ever read on slashdot, and that's saying a lot. Gates was not some nine-headed demon who rose from the pits of hell and tricked everyone into selling software. I was there long before and during this period of time, and everybody was looking for a way to make a buck. Shitty little flat-file database programs cost $695. I know, I wrote one of them. Word processors that weren't much more than today's for-free Windows Wordpad would set you back $499. Games were priced the same ($30-$50).
Obviously, there was a wide array of free software out there -- just as there is today. Go figure.
I actually race Vipers on an almost-monthly basis and I have to disagree with you. I love the tiny, tiny handful of well-done racing games (of which I think Project Gotham is probably the best by leaps and bounds, and yes I've played the hell out of the Grand Turisimo series). It's possible to create a fairly realistic driving game, but it's difficult.
Plain and simple: Under real racing conditions, if you're driving by feel, you're going to crash. Feel does play into it (an F1 driver recently said the two most important instruments were his eyes and his ass), but probably 99% of your most important input is visual cues. Hitting turns correctly requires a lot of setup which has to be done quite a long time before you get anywhere close to the turn itself.
Based on his subsequent explanation, much of what the F1 driver had in mind was using feel to monitor the car's "health," which is mostly ignored in current video game racing titles (other than vague performance degradation from accident damage in certain games).
I can see how you'd feel that way from regular street driving, but regular street driving isn't what the games are trying to simulate...
Also, I'm sure exotic car manufactures such as Farrari would be interested in the stuff.
That's a very interesting thought, and worth expanding upon for anyone not familiar with the state of the art, or possibly unfamiliar with cars in general.
At first I was tempted to dismiss your statement because carbon fiber cloth is easy to get, is relatively cheap (it's the autoclave that makes it so expensive to use/make), is well understood, and works very well -- and has been in widespread use for exotic automotive applications for about a decade. A friend of mine recently had a minor crash in his F50, as an example, and the repairs involved $150,000 worth of new carbon fiber -- 8 layers for most body panels, with each layer being completely different than the others, with each layer put there for specific reasons. Some are straight weaves, some are cross-weaves, and some are honeycombs -- all in a car that is 8 years old. So carbon fiber in general is definitely well-understood.
However, in thinking about the properties of this specific application, I realized you might be on to something. Where this new strand-format CF might be interesting is places where steel tension cables are used today -- shifter cables, parking brake cables, wing adjustments, support structures, and so on. After all, in certain circles, any weight reduction is worth the money. I would expect to see it first in Formula-1 -- assuming the newly tech-averse rule-making idiots at the FIA don't make carbon fiber illegal, too.
Exactly. I believe a more accurate explanation is probably:
"As Japanese teens are so used to pecking messages on their cellphones, they are now too lazy to actually learn to type on a two-handed QWERY keyboard."
I can type 130 WPM with very, very few errors. Closer to 150 WPM on a real, full-size Microsoft Natural keyboard. It's simply physically impossible to move one or both thumbs that quickly.
The other question is... is this "typing" some form of Japanese? In that case you probably have to redefine "WPM" -- although the article does refer to a QWERTY keyboard. If it isn't Japanese, is it that abbreviated SMS lingo? That hardly qualifies as typing. At my real speed of 130 WPM, imagine how my WPM rate would skyrocket if I dropped 50% of the letters and didn't correct errors, yet claimed full speed for the sake of a sensational article...
I had only one interview and it basically went like this:
President (of the company): "Have you ever used a Mac?"
Me: "Yes I have, for about 15 years."
President: "Can you start on Monday?"
So in other words, you walked into exactly the same situation, but it just happened to be Macs instead of PCs, and that makes it OK?
As a side question, does 15 years experience on Macs really carry over to the current platform? I mean, 15 years ago puts you in the Mac IIx (SE/30) days. I don't use Macs, I'm asking because I don't know and I'm curious.
I know your point is a PC-only school policy would have prevented most people from landing that job, but if merely using a Mac is the only qualification, that either says a lot about that company, or possibly the state of Mac-oriented employment in general...
The US is BOTH a republic (because Dubya is an American), and a democracy (because the head of state is elected by the people, loosely speaking)
American stupidity, indeed.
The US is a representative republic. This means the government operates through the decisions made by elected individuals who are expected to represent the interests of the citizenry. It does not make any statement about how those representatives are elected.
The US is not a democracy. In a democracy, the "will of the people" outweighs the rights of any individual, and that is not how our government operates (not yet, anyway). Since you said "loosely speaking", I assume you're aware that our President is chosen by the Electoral College, and is absolutely not elected by the people, so you are apparently aware that we are not a democracy. The obvious example is John Quincy Adams being chosen as President in spite of Andrew Jackson having won the popular vote (notwithstanding conspiracy theories and messy facts about whether "the popular vote" even existed prior to Jackson rasing hell about it).
A large portion of shipped executables are blocks of standard code from the compiler. If you're using a Microsoft compiler, you can strip out the standard chunks and pull those chunks in from the binaries that are already in Windows. If you're using another compiler, you can still probably do the same kind of thing: some intelligent block compression with the included code that'll do better than the "dumb" compression from "zip" or other algorithms.
Actually, this is what the "sliding dictionary" is in LZW compression (and I assume in other types of comrpession, but I've only written LZW codecs in my life, and never studied compression formally). You scan through the target files looking for large chunks of redundant content. Through some process of weighting number-of-repeated-bytes versus number-of-occurrances, you assign progressively longer "codes" to each chunk of repeated bytes. Viola, compression. The various flags to a compression application tell it how much scanning to do and whether to scan across files (both are speed/compression tradeoffs).
The only thing novel in your suggestion (novel compared to standard compression) is trying to recover (decode) those chunks of content from locations other than the archive file itself. That would probably be very, very risky -- certainly not the type of risk a company like Adobe, for example, would be willing to take on an installer for one of the most popular pieces of software in the world, and more importantly (to them), their bread & butter.
A lot of compilers do tricks to make code faster. Not many do things to explicitly make code smaller.
Actually, it's common to have compiler switches to optimize for either space or performance, and even feature-specific switches...
Exactly. Non-OEM version. Unless you're building machines for re-sale, you are NOT entitled to buy the OEM version of the software. You have obtained your software through fraudulent means, and are in violation of the Microsoft EULA, and probably several other laws, as the AC mentioned above, the Theft Act of 1968 in the U.K.
The question is: Are you lying, or do you really believe this? The OEM license means you are not entitled to support from Microsoft. There are no additional implications. I have even personally checked with Microsoft about this.
The only Microsoft license I can think which could get a legitimate licensee into legal trouble is that which comes with MSDN subscriptions. That license is explicitly intended to allow for testing and development usage only, not permanent personal or business use, and most MSDN subscribers do not seem to know or understand this (or perhaps they don't care).
It's too bad your reply is buried 9 pages out from the original article, you deserve to be modded up. That was exactly the point I was prepared to make. Sure the loss of handwriting skills is worth discussing, but it's not remotely as disturbing as the sharp decline in most people's ability to construct complete, correct sentences, to comprehend what they read, or to simply spell words correctly.
Slashdot Mirror
It is important to realize that Microsoft extended Java, they did not intentionally break it. Well, I guess some people define that as breaking it, but I mean they didn't intentionally build their JVMs and JITs so that they did standard Java things incorrectly. Even Sun admitted the MS JVM was the best available implementation, I heard Sun guys saying it all the time at the first couple Java One events.
I road race my 2001 Viper. Modifications and maintenance take up some time, but not much, so I've also started building a 34 Ford chopped 5-window coupe with a blown Hemi 426 for drag racing (and street use, heh heh). And when I say "build" I really mean "build" -- I'm fabricating as much as possible in my garage -- the frame, suspension, all kinds of things. The stuff I don't fabricate still requires quite a bit of work to make it usable with the rest of the car. For example, the rear brakes will be my old stock front brake system from my Viper, and the driveline uses parts from no less than six different major auto manufacturers. It's kind of like assembling a giant 400HP jigsaw puzzle that was never originally one single piece...
IBM has something similar to this, which you can read about here...
Of course, you have to start with an image database that potentially contains that thing you found at a garage sale (or the chick you ran into at the mall).
Microsoft didn't kill it, early JVM/JIT performance and the still-low-average-bandwidth in the early days of Java are what killed it. I know, I used to write web-delivered client-side Java applets in the early days.
Locally stored Java apps were a non-starter from day-one due to the total lack of consistency -- you simply couldn't put an app out there and tell the end-user how to start it. Either they knew (and most people won't know), or they used someone else's software.
It is a well-known and well-documented fact that Microsoft had the best-performing JVM throughout most of Java's early life. If anything, they did the most to prop it up. (Though not out of any noble purpose, of course.)
Increasing the number of reps would only make it easier for the reps to avoid issues and increase their ability to get away with selling out to the highest bidder. In effect, you would dilute the amount of attention the media could focus on any given rep (and unfortunately, the media is really the only way most people find out what a rep is up to). Most people don't know who their current reps are -- imagine having to keep track of 8300 of those guys... or trying to decide who to elect, or re-elect...
Indeed -- one factor among many (the rest of my post).
Unless your backup is so precious that you have it in a sealed nitrogen-filled Swiss vault somewhere, perhaps. Look, I have literally hundreds of backup CDs (used to be in the multimedia biz) and the great majority of them are just fine (I grab a few beers and sit down to test-read all of them once a year or so)... but the fact is, some just die for no good reason. It has happened to me and others I know often enough that your claim of 50 years is weak at best.
But do whatever floats your boat, I was just trying to help.
You mean American industries like the recording industry, where Sony is a major player? Yeah, thought so.
Don't be so sure. A HUGE number of factors play into the lifespan of burned media. I've had reasonably high quality CDs (e.g. high cyan Ricoh stuff, not spindled 100-bulk no-name crap) go bad after only 5 years or so.
Not that I don't use it myself for backup -- I just make two backups. It's certainly cheap enough.
<sarcasm> Yeah, it's pretty hard to find people to support Microsoft software. </sarcasm>
29 cars is pretty good, especially when you consider the quality of the 3D models and the fact that they all handle and perform very differently (e.g. they aren't just skins over a small handful of performance characteristics). As for locations, there are about 200 tracks in the game -- granted, they're all set in 15 or 20 areas of only 4 cities, but that's still a lot of variety.
I would think increasing realism is detrimental to sales mainly because cars are somewhat difficult to drive properly at speed. I like Gotham because I can use real-world braking and cornering techniques to improve my time and speed. Most people, though, just wouldn't know what to do.
I'm on the fence about upgrades. Generally their impact is blown way out of proportion to their real-life counterparts, which to me makes it less fun to pick and choose among your options. If everything is flat-out awesome, then it's just a matter of trying to collect as many of them as you can.
One P1-90 running NT4 (very usable, a dev SQL Server box in fact)
Two P1-100s running NT4 (very usable)
Two P1-166s running Win98 (webcams)
Two P2-266s running Win98 (webcams, MP3s)
One P3-500 running Win2K (alarm & home control) One P3-700 running Win2K (MP3s, basement Quake3, DVD, web) One 1GHz Althon running WinXP (MP3s, living-room Quake3, DVD, web) Two P4-2.6s running XP (main PCs)
Mainly I use so many because of physical separation -- some upstairs, some downstairs at various places in the garages, etc.
And don't give me crap about stability. These machines do not crash, ever. The ONLY BSOD I've seen in YEARS was (coincidentally) yesterday, after trying the much-heralded BitTorrent for the first time. Apparently it doesn't like something about one of my LinkSys NIC drivers. Prior to that, I haven't had a non-hardware failure in at least two years. These machines are powered up and running 24x7, all on battery backup.
Could I do the same with Linux? Sure. Do I want to? Nope. I already owned all those copies of Windows (yeah, I actually paid for it, go figure), and in many cases it would be a royal bitch to do what I want under Linux (old webcams, alarm system & home control interfacing, DVD playback).
The surprising one, of course, is NT4 and SQL Server on a P90. That machine has 96MB of RAM and an old throw-away 6GB HDD, and has been in continuous usage since the mid-90's. I had a CD-ROM drive die in it, but other than that, it's just hummed along without a hitch.
I also have a 286 I've been thinking about powering up just for laughs. Not sure what I'd do with it yet, though.
Most vehicles only support a very small range of settings. As an example, I put 33" tires on one of my trucks, but the engine computer can't be set for anything larger than 32" tires. Thus, my speedometer is always incorrect with that set of tires on it. (I do happen to also run a 31" set for towing.)
Also, current engine computers almost exclusively rely upon O2 sensors to control anything that would affect your mileage or acceleration or other performance characteristics. About the only major exception I can think of is the GM system that shuts down half of the engine during steady cruising, and that only requires the ability to recognize a steady speed.
(No wonder they're so far behind on submissions, they're all busy in Slasdot Labs (TM) running the latest round of amazing new benchmarks. I can't wait to see the results!)
Of course. Nobody stores anything interesting on Linux boxes...
Well, except for that "moon" thing.
-1, Poster Smoking Crack
Your lollipops-and-sunshine view of the software world before Gates is one of the most ridiculous things I've ever read on slashdot, and that's saying a lot. Gates was not some nine-headed demon who rose from the pits of hell and tricked everyone into selling software. I was there long before and during this period of time, and everybody was looking for a way to make a buck. Shitty little flat-file database programs cost $695. I know, I wrote one of them. Word processors that weren't much more than today's for-free Windows Wordpad would set you back $499. Games were priced the same ($30-$50).
Obviously, there was a wide array of free software out there -- just as there is today. Go figure.
Plain and simple: Under real racing conditions, if you're driving by feel, you're going to crash. Feel does play into it (an F1 driver recently said the two most important instruments were his eyes and his ass), but probably 99% of your most important input is visual cues. Hitting turns correctly requires a lot of setup which has to be done quite a long time before you get anywhere close to the turn itself.
Based on his subsequent explanation, much of what the F1 driver had in mind was using feel to monitor the car's "health," which is mostly ignored in current video game racing titles (other than vague performance degradation from accident damage in certain games).
I can see how you'd feel that way from regular street driving, but regular street driving isn't what the games are trying to simulate...
That's a very interesting thought, and worth expanding upon for anyone not familiar with the state of the art, or possibly unfamiliar with cars in general.
At first I was tempted to dismiss your statement because carbon fiber cloth is easy to get, is relatively cheap (it's the autoclave that makes it so expensive to use/make), is well understood, and works very well -- and has been in widespread use for exotic automotive applications for about a decade. A friend of mine recently had a minor crash in his F50, as an example, and the repairs involved $150,000 worth of new carbon fiber -- 8 layers for most body panels, with each layer being completely different than the others, with each layer put there for specific reasons. Some are straight weaves, some are cross-weaves, and some are honeycombs -- all in a car that is 8 years old. So carbon fiber in general is definitely well-understood.
However, in thinking about the properties of this specific application, I realized you might be on to something. Where this new strand-format CF might be interesting is places where steel tension cables are used today -- shifter cables, parking brake cables, wing adjustments, support structures, and so on. After all, in certain circles, any weight reduction is worth the money. I would expect to see it first in Formula-1 -- assuming the newly tech-averse rule-making idiots at the FIA don't make carbon fiber illegal, too.
Definitely an interesting thought.
"As Japanese teens are so used to pecking messages on their cellphones, they are now too lazy to actually learn to type on a two-handed QWERY keyboard."
I can type 130 WPM with very, very few errors. Closer to 150 WPM on a real, full-size Microsoft Natural keyboard. It's simply physically impossible to move one or both thumbs that quickly.
The other question is... is this "typing" some form of Japanese? In that case you probably have to redefine "WPM" -- although the article does refer to a QWERTY keyboard. If it isn't Japanese, is it that abbreviated SMS lingo? That hardly qualifies as typing. At my real speed of 130 WPM, imagine how my WPM rate would skyrocket if I dropped 50% of the letters and didn't correct errors, yet claimed full speed for the sake of a sensational article...
Either way... pure, unadulterated BS.
President (of the company): "Have you ever used a Mac?"
Me: "Yes I have, for about 15 years."
President: "Can you start on Monday?" So in other words, you walked into exactly the same situation, but it just happened to be Macs instead of PCs, and that makes it OK?
As a side question, does 15 years experience on Macs really carry over to the current platform? I mean, 15 years ago puts you in the Mac IIx (SE/30) days. I don't use Macs, I'm asking because I don't know and I'm curious.
I know your point is a PC-only school policy would have prevented most people from landing that job, but if merely using a Mac is the only qualification, that either says a lot about that company, or possibly the state of Mac-oriented employment in general...
The US is BOTH a republic (because Dubya is an American), and a democracy (because the head of state is elected by the people, loosely speaking)
American stupidity, indeed.
The US is a representative republic. This means the government operates through the decisions made by elected individuals who are expected to represent the interests of the citizenry. It does not make any statement about how those representatives are elected.
The US is not a democracy. In a democracy, the "will of the people" outweighs the rights of any individual, and that is not how our government operates (not yet, anyway). Since you said "loosely speaking", I assume you're aware that our President is chosen by the Electoral College, and is absolutely not elected by the people, so you are apparently aware that we are not a democracy. The obvious example is John Quincy Adams being chosen as President in spite of Andrew Jackson having won the popular vote (notwithstanding conspiracy theories and messy facts about whether "the popular vote" even existed prior to Jackson rasing hell about it).
Actually, this is what the "sliding dictionary" is in LZW compression (and I assume in other types of comrpession, but I've only written LZW codecs in my life, and never studied compression formally). You scan through the target files looking for large chunks of redundant content. Through some process of weighting number-of-repeated-bytes versus number-of-occurrances, you assign progressively longer "codes" to each chunk of repeated bytes. Viola, compression. The various flags to a compression application tell it how much scanning to do and whether to scan across files (both are speed/compression tradeoffs).
The only thing novel in your suggestion (novel compared to standard compression) is trying to recover (decode) those chunks of content from locations other than the archive file itself. That would probably be very, very risky -- certainly not the type of risk a company like Adobe, for example, would be willing to take on an installer for one of the most popular pieces of software in the world, and more importantly (to them), their bread & butter.
A lot of compilers do tricks to make code faster. Not many do things to explicitly make code smaller.
Actually, it's common to have compiler switches to optimize for either space or performance, and even feature-specific switches...
Um, the article AGREES with you.
Reading for context: try it.
The question is: Are you lying, or do you really believe this? The OEM license means you are not entitled to support from Microsoft. There are no additional implications. I have even personally checked with Microsoft about this.
The only Microsoft license I can think which could get a legitimate licensee into legal trouble is that which comes with MSDN subscriptions. That license is explicitly intended to allow for testing and development usage only, not permanent personal or business use, and most MSDN subscribers do not seem to know or understand this (or perhaps they don't care).
It's too bad your reply is buried 9 pages out from the original article, you deserve to be modded up. That was exactly the point I was prepared to make. Sure the loss of handwriting skills is worth discussing, but it's not remotely as disturbing as the sharp decline in most people's ability to construct complete, correct sentences, to comprehend what they read, or to simply spell words correctly.