No, no and no. You are assuming that any person can be replaced with anyone about as qualified, and the resulting accomplishments are the same. Further, you assume happines and motivation have nothing to do with productivity, related to motivation, experience and skills (and in rougly that order of importance).
This may be true for simplest of assembly line jobs, although even there I'd doubt that. But for any developer job (or similar, jobs that need skilled professionals) there's huge differences in productivity.
My estimation is that difference between minimally adequate (below average but still nominally skilled and/or motivated, that is, able to work to some degree) and average ("normal" skills and motivation) worker is in the order of 3 - 5 times, and between average and excellent further 2 - 3 times. So, roughly speaking, best of employees get as much done as 10 or more "sucky" employees. And that's ignoring the fundamental limitations of "bunch of cheap incompetent idiots" approach, which is that there are some demanding tasks where you just can NOT get things done without skilled, talented, motivated individuals (not all tasks, but some key tasks).
So, if an employer follows tactics you outlined, here's the likely scenario:
By treating employees badly, morale goes down, and productivity suffers significantly, easily to less than half of optimal sustainable productivity, in matter of weeks if not days.
By making it known everyone considers everyone to be replaceable, most employees (of any level) start looking for new job. Best ones (that are most productive and skilled) find new job more easily, thus there's significant brain leakage. Below-average people try to hang in there and do not leave involuntarily.
Replacing people lost to attrition and lay-offs is costly as well as risky; there's no reliable way to make sure person you hire is as good as you think. Only time will tell. Ones you had you had much better understanding of. So, chances are you won't be hiring better people than you had (assuming originally you very succesful in developing and keeping talent).
Training people for specific duties they have takes time (few people get a new job that's _exactly_ like they old one). Even highly skilled professionals take a while to adapt.
Newcomers have less reason to be loyal to begin with. Their morale is likely to be neutral, which depending on situation may be better or worse than average employee morale. But in case where it's higher, it _quickly_ lowers to average, as person learns what kind of a rathole job is. In case of existing employees having good morale, it takes longer for morale to raise... but it will happen if things stay good.
Another big mistake is misunderstanding the role of (monetary) compensation to motivation. Rewards are good, almost independent of size of reward (except of insulting small ones, like those Larry Ellison dolls given as annual bonus). But above and beyond that, money is not much of an incentive to otherwise content people. And for uncontent people, well, it's only temporary relief that soon is forgotten.
However, the opposite (ie. short-changing employees) does affect morale drastically and quickly. Whereas giving a raise helps a bit in short run, and stabilizes things in long run, salary reduction (or "too low" starting salary) is a quick and effective tool of demoralization. No matter what the situation, that's equivalent to middle finger salute.
I have always found this type of game to be rather odd. Isn't social interaction what you are supposed to be doing in real life?
No, not ONLY in real life. If you have ever played MUDs, the exotic combination of non-real world and 'real' characters (as in played by real people more or less acting as if they were not) is the killer substance that gets you hooked.
Also consider "but you can interact socially in real life" aspect a bit; wouldn't it be interesting if, unlike in Real World, you could actually tell your boss exactly what you think about him/her, act in totally different manner you otherwise do (without burning bridges in your real life)... Thus, "doing something you just can't do otherwise" is definitely not unique aspect of combat games or simulators. Rather, it (often referred to as role-playing) is one of main attractions of (massively) multi-player games.
And like many people have pointed out, social interaction need not just be talking and such; tactics and communication in otherwise boring 1st person shoot-em-ups also counts as form of social interaction.
that it took until 1990 for the first game capable of side-scrolling!
...
That is pretty bad and shows how far computer games were behind technology wise.
Well, games on _x86 PC_ platform were behind. All the other existing so-called "home computers" were able to do side-scrollers from day one (C64, Atari boxes, Apple II, Amiga, Atari ST). Of course, like you said, tables were turned with 1st person shoot-em-ups; there raw CPU power was needed, not (fairly simple) hardware graphics processor features for allowing pixel offsetting.
As to gaming world being dead, no, not really. Neither one person nor one new genre (that Doom, or rather Castle Wolfenstein started) really prevented any major catastrophe IMO.
There were so many different genres that I find it interesting that for a while FP shooters were so much dominating game scene. Why were gamers like lemmings?
What would have happened is that someone else would have sooner or later (sooner I'd guess) came up with same idea and decent enough implementation.
Really, the idea of using 1st person perspective and realistic-enough looking 3D display was nothing new. Implementation was innovative, and gave just enough advantage for ID to be the first one to get it done well enough; otherwise it would have taken bit longer for CPU speed to catch up.
First of all, cost of extending copyright need not be huge. I don't think only rich people can register domains, for example, as renewal cost while not negligible, is not huge.
But more to the point... you can't really even publish your copyrighted work, unless you are rich, or have produced something others (publishing company) thinks can make them rich (companies that are very likely to be "rich" compared to most individuals). Meaning that most publishable works can certainly earn enough money to allow them to be copyrighted for as long as they remain productive. So, even a poor (at least originally poor) author/composer can get copyright extended, as long as work in question is considered good enough (as measured by people buying copies of the work).
Plus, don't confuse copyright with authorship. Even if one doesn't hold a copyright on, say, a book, doesn't mean (s)he has lost authorship. That person just can not prevent others from reproducing (aka copying) the work in question.
Not really. "Limited" as used in legal context -- as opposed to in mathematics or so -- likely has stricter meaning. Long period of time (or enough successive increments) likely would be declared not limited. Not that it'd be impossible to find a lawyer to argue otherwise, but that would be difficult case to make.
Even more importantly, when interpreting constitution, judges do try to consider not just literal meaning, but also intention. Clearly intention wasn't to allow anything but whatever was considered reasonable limited period of time.
It's not your choice. And it's certainly not your right. Your sense of entitlement is sickening.
Sure it is his choice, if he chooses to make that. That doesn't guarantee it happens, but it's his choice. But where did he claim it's his right, or that he's entitled to it? He was barely saying he intends not to try to change the way things are, not that it's a god-given birth right, nor that it's the way things even should be.
I'm not saying you need to agree that things are just fine, but don't fight the strawmen, and read more into parent post than is there.
There is not only enough food in the world to feed every single living human being, but DRAMATICALLY MORE THAN NECESSARY
Note that while in general I mostly agree with what you were saying before this statement, I'm reasonably sure this statement is an urban legend. World's food output is not dramatically higher than need (currently), by any standards. It's barely keeping up with demand. Potential for food production is bigger, but that would require lots of changes, plus absolutely requires use of 'industrial' approach to food production (unlike many "green food production" proponents believe... industrial fertilizers at least are pretty much a must, just need to make sure usage is optimized so as not to cause problems with nutrient washout etc).
One factor behind myth of "having way too much food yet still hunger" is that in world market there is always "surplus" food on sale, and they are oftentimes hard to sell, too. Amounts, however, are reasonably small (compared to total consumption), and problem is more about poorest, most needy, countries not having means to purchase the goods than about actually having real excess food production capacity.
Yes, even then. Although US welfare system is amongst least comprehensive of western countries, it's still pretty darn good compared to, say, african countries.
Plus, at least here one can still go flipping burgers, and earn more than those 5 billion others (I might claim it's less than 5, perhaps 4, what with upcoming middle class of China and India, but what the hell).
And even if you were _really_ poor, for many foreigers stereotypic image of, say, american, would still be rich white anglosaxon middle-aged man. So his statement would be true.:-)
Note that when comparing foreign wages to ours, the author of the article specifically chooses to mention -programmer- and -project manager- salaries.
Thing I was wondering, however, was that salary rates for US workers seemed awfully low:
"fifth of those in the United States, with programmers earning $250 to $700 a month, compared with $1,600 to $3,600"
I didn't think I was hugely overpaid, but I earn twice as much as "high end" programmers are supposed to earn.
Perhaps that was just a typo or something, since they do mention hourly wages of 45$ later on (that'd translate to ~7200 instead of 3600).
Ugh. That's one of more ignorant "let's ship immigrants back to where they came from" propaganda pages, with unsubstantiated claims and falsified information.
Anyway; if you do read the page, try also talking to someone on H1B (and/or co-worker or manager of some such person) and see if reality matches the sensationalist claims of h1b."info", to get balanced view of the whole mess.
For what it's worth my understanding is that it's not worth envying H1B peon's position, and claiming they are here just leeching here stealing jobs. Job security with H1B is really really bad; visas really are temporary, and if you do lose your job, finding a new one is next to impossible (currently, didn't use to be, but that was same for 'natives')
Plus, right now managing to get a NEW H1B visa is next to impossible for most programmers. Extending existing ones (or transferring) is ok, but proving there is shortage of qualified american programmers is tricky. Like it should be, I might add. All this ranting about "increasing H1B visa quota" is so clueless... they didn't even have to use all 200K they had for last year, and even getting past 65K might be tough call (that quota is only used against new H1B visas, not extensions).
Finally, there's one more thing many people even here are unaware of. Since it's possible to apply for a green card after getting a H1B visa, majority of all current or previous H1B holders are or are becoming permanent residents (and eventually citizens if they choose to). So, "closing the gates" wouldn't make that much of a difference either way. Influx of foreign work force has already happened in high tech. Many of those indian DBAs or chinese programmers (etc. etc.) are actually NOT on H1B any more.
Re:Separating Content from Presentation a Good Thi
on
Office 2003 and XML
·
· Score: 1
A well-formed SGML or XML document should have absolutely NO formatting information contained
within the content.
But that's not what he was saying. Try to avoid knee-jerk reactions. What he (and most everyone else) is saying is that presentation information should be available somehow. No one's asking it to necessarily be embedded in content, just be available, either on different file, or be defined as static specification (like HTML default rendering suggestions or de facto 'standards').
Without presentation information there just is no way to know how to adequately render content for human usage. And MS certainly stores and uses such information, but apparently they just don't want to share that... since that would make the somewhat open format much much more useful for competitors.
MS should just look at how OpenOffice folks did their format. It's not perfect (far from it), but it adequately both stores presentation information and separates actual structured content.
Perhaps because they could cut costs and earn more money? AMD chips have traditionally been less expensive (comparing similarly performing CPUs). Companies are known to be interested in getting bigger profit margins you know, and getting price of one of costlier components should help.
I'm willing to make the choice for linux, but forcing Linux onto 7000 students, who might just want to use hotmail in the library, or catch a quicktime CNN news clip, is extreme enough to merit contention.
Although I do agree with your points in general, it's worth nothing that for "non power users" pretty much any OS does ok nowadays (any as in a Unix/Linux, MacOs, Windows). All have decent browsers, enough plugins for multimedia content, basic word proc. and spreadsheet apps, email clients etc. etc. So, general attitude from "mainstream" people really should just be "whatever".
Read parent post again, and make a note of the fact he wasn't claiming there has to be choice. He was just pointing out the lack of choice that has often existed, even before offers like the one article describes. This because parent poster implied this would limit the choice that existed.
So, what you have here is essentially a reply to strawman argument.
2) US restrictions on exporting high powered chips and other computer parts are easily diluted by open standards.
Well I don't know if I'd call a watered down 500mhz MIPS based chip "high powered". Maybe once you lash 128 of them together you'll have a decently powered box, but individually, it's way less than yesterdays tech.
You are ignoring his point, though; export restrictions aren't aimed at preventing "power users" in China from getting fastest running single-processor system, but from building "super computers", that are without an exception (massively) multi-processor systems.
And for such systems, distinction between 3 GHz vs. 500Mhz (assuming CPU power was directly proportional to clock rate) is not all that critical. Especially when you combine SMP with distributed systems.
Then again, export controls don't really protect against this even without this new CPU; it'd be possible to use freely traded Intel/AMD chips as well. However, it may be easier to design the big system using these CPUs if/when they are properly designed for SMP systems (since x86 procs in general are not designed for bigger than, say, 4-way SMP systems).
and how could it be competitive running at 500mhz?
It is entirely possible to do totally different design trade-offs if raw computing power is not THE goal. So, although
this is just speculation, here are some potential improvements it could offer:
Lower power consumption, meaning easier cooling. Quiet systems without CPU fans, embedded devices.
Due to lower power consumption, longer life-span for CPU (plus higher reliability).
Simpler interface design, to allow simpler (and cheaper) motherboard design (compare to development of more and more complicated sockets for modern CPUs, required by higher clockspeeds)
Basically it all comes down to this: for most things, any modern CPU (even one that runs at 500mhz) is enough. Thus, 3Ghz Octium shouldn't be much more interesting than 0.5Ghz Dragon (if it wasn't for advertising). And if raw CPU power is meaningless, currently "secondary" things like price and power consumption could become real factors in designing what to use.
... yeah, it's like somebody took an OS (unix) and just cloned its features, just renamed the results (linux). Who'd of thank?
Interesting thing, however, is why they chose MIPS ISA over, say, Alpha? Both are reasonably clean RISC designs... perhaps it's just that MIPS is extensively documented in quite a few basic processor design books? Hell, even I had to design and implementation of the first MIPS processor in VHDL (for my proc. architectures course), and I'm not an EE major.
That's unfortunately completely false argument. Leverage does NOT mean simple "use", nor "indirect use". Simply put, if word "use" is good fit, "leverage" is not.
This because leverage means (in more general sense than in its strict definition of using a lever to move an object) using smaller force (or amount etc) to achieve similar effect as using bigger force (etc). This is what a lever does -- it "converts distance to force" (same amount of work is done by either applying smaller amount of force over longer distance as bigger force over shorter distance).
And that's why many people (myself included) do think that majority of usage cases for "leverage" are in fact buzzspeak. People who think leverage is "just fancier way of saying 'use'" clearly do not understand meaning of the word.
There are reasons why loose coupling (distributed systems, Beowulf etc) is sometimes better, and tight coupling (SMP, virtual domains in Solaris) is sometimes better.
Good things about "intra-box scalability" include:
Lower maintenance costs. Much of hardware can
be dynamically allocated and/or shared, either in real-time or at least without rebooting.
Much better I/O throughput between different computational tasks. Potentially back-plane / bus speeds as opposed to network speeds (order of magnitude faster inter-process communication).
Potentially lower hardware costs (related to first item). This depends a lot though... but in case of, say, Sun servers, it makes sense to buy couple of more CPUs instead of more systems.
Easier maintenance / administration (related to first item). Depends on tools, but it is generally easier to maintain single "big" box than multiple "small" ones.
Good point... and I think same applies to genres in computer games and literature (although to a lesser degree in literature I guess). I might argue that there's always room for _some_ improvement, but after a while it probably makes more sense for most people to either move on (change genre, or break its constraints), or to enjoy what has been achieved as is.
Everything's already been done, or so it seems, so really original and entertaining gameplay+graphics is a tough combination to create.
I disagree; I think the idea of "everything's already been done at least twice" is a common phallacy. Some people claim that all good music has already been done, or all good movies, or even all good paintings.
And yet amount of permutations for basic components of music (melodic, harmonic, rhythmic), literature (themes, characters, time, style) or, computer games (ones similar to literature), is pretty much infinite. There will _always_ be room for new things in any of above-mentioned forms of art.
I agree with the article. Lack of innovative games has more to do with business objectives of predictable revenue than with not having room to explore that limits original games. 20 years ago technology was limiting things much more; nowadays it's almost a moot point, at least from game idea point of view. Any interesting non-novelty gameplay idea can probably be implemented on standard gaming system of choice. But since coming up with a new idea IS more difficult than refining an existing idea (I'm not arguing otherwise), the risks associated just make it so much more compelling to "just write yet another sequel of a hit".
Funnily enough, this is just one of those problems with short-sighted businesses. Without new innovative hits, there won't be chance for new predictably profitable sequels. You can only do so many sequels from a certain theme, with lowering profitability... and then have to move on.
According to NPR if Sadaam burns his own oil fields analysts predict prices will rise to $80 a barrel! This would make gas $4.00 a gallon!
Well... this is off-topic, but... why on earth would Saddam burn "his own" oil fields?
What puzzles me is why people (even smart ones) make the logical leap from "he burnt Kuwaits oild fields" to "he will burn Iraq's oil fields", without thinking of obvious differences.
In case of Kuwait, SH was just playing sour loser and trying to prevent Kuwaitians (plus allies) from benefiting from their oil, when we knew he'd be asskicked out. In case of Iraq, it's "his" oil. Iraq will still have the oil, no matter what the outcome (unless you really believe in extreme conspiracy theories). There's hardly point in burning any of it. It won't slow US troops down; it's like panting in your pants to prevent cops from arresting you. Unlikely to work to say the least.
True. It's just like people who try to make relative sizes/volumes of things more understandable by using real life objects. Things like: "think of Sun as an orange... now Earth is a pea, and it's so far from the orange that you can't see it 'cause there's that building over there that blocks line of sight; but if you could see it it'd be somewhere beyond north-Dakota". Or, "if all Oreos sold in one day were stacked in one place, they'd be as big as USS Enterprise and one half of a destroyer, plus 5 lifeboats". Oh, I see, now it all suddenly makes perfect sense; I really understand exact relationship between huge things.
I mean; whenever the order of magnitude difference between things exceeds a threshold (whatever it may be for the person in question; anything from 2 to 6), there is just no way to really put things in perspective, by using silly analogies. And LOC seems like just another such analogy (perhaps even worse than the older "page of text" used instead of kilobyte... assuming 8-bit chars).
The interesting part behind XML is, you don't have to invent your own syntax and implement your own parser. As a result other applications can quite easily access your data, too.
Well. XML is human-readable, but not automatically human understandable. My pet peeve regarding most voal XML hype is the claim "XML is self-documenting" (most often used with SOAP but sometimes generalized to all XML data). Something that just isn't true. It's just like claiming C programs are self-documenting; after all, we do have common syntax with C as well as with XML.
XML can be understood by anyone only at syntactic (structural) level; you can see how elements are nested. What is not automatic is semantic meanings. Without knowing which XML application (as in XML terminology; not piece of code, but a language like XHTML or RDF) one is using (and knowing its semantics), it's not much use to know structure. It's like understanding syntactic structure of english language without knowing any of the words used. Knowing that a sentence has structure "noun predicate objective"
isn't enough to understand what is actually being said. You need to know the words (vocabulary) and their meaning.
So... while using XML does simplify task of interchangeability a bit (on top of Ascii/Unicode, that solves another lower level problem, on top of de factor 'standard' of using 8-bit byte instead, of, say, older 9-bit words, and... ok, you get the idea), it does only solve first 10% or so. Agreeing on common XML application solves perhaps next 50% or so. And then actually specifying remaining semantics (that can not be expressed by DTD or even XML Schema) is the final part.
Slashdot Mirror
This may be true for simplest of assembly line jobs, although even there I'd doubt that. But for any developer job (or similar, jobs that need skilled professionals) there's huge differences in productivity.
My estimation is that difference between minimally adequate (below average but still nominally skilled and/or motivated, that is, able to work to some degree) and average ("normal" skills and motivation) worker is in the order of 3 - 5 times, and between average and excellent further 2 - 3 times. So, roughly speaking, best of employees get as much done as 10 or more "sucky" employees. And that's ignoring the fundamental limitations of "bunch of cheap incompetent idiots" approach, which is that there are some demanding tasks where you just can NOT get things done without skilled, talented, motivated individuals (not all tasks, but some key tasks).
So, if an employer follows tactics you outlined, here's the likely scenario:
Another big mistake is misunderstanding the role of (monetary) compensation to motivation. Rewards are good, almost independent of size of reward (except of insulting small ones, like those Larry Ellison dolls given as annual bonus). But above and beyond that, money is not much of an incentive to otherwise content people. And for uncontent people, well, it's only temporary relief that soon is forgotten.
However, the opposite (ie. short-changing employees) does affect morale drastically and quickly. Whereas giving a raise helps a bit in short run, and stabilizes things in long run, salary reduction (or "too low" starting salary) is a quick and effective tool of demoralization. No matter what the situation, that's equivalent to middle finger salute.
No, not ONLY in real life. If you have ever played MUDs, the exotic combination of non-real world and 'real' characters (as in played by real people more or less acting as if they were not) is the killer substance that gets you hooked.
Also consider "but you can interact socially in real life" aspect a bit; wouldn't it be interesting if, unlike in Real World, you could actually tell your boss exactly what you think about him/her, act in totally different manner you otherwise do (without burning bridges in your real life)... Thus, "doing something you just can't do otherwise" is definitely not unique aspect of combat games or simulators. Rather, it (often referred to as role-playing) is one of main attractions of (massively) multi-player games.
And like many people have pointed out, social interaction need not just be talking and such; tactics and communication in otherwise boring 1st person shoot-em-ups also counts as form of social interaction.
That is pretty bad and shows how far computer games were behind technology wise.
Well, games on _x86 PC_ platform were behind. All the other existing so-called "home computers" were able to do side-scrollers from day one (C64, Atari boxes, Apple II, Amiga, Atari ST). Of course, like you said, tables were turned with 1st person shoot-em-ups; there raw CPU power was needed, not (fairly simple) hardware graphics processor features for allowing pixel offsetting.
As to gaming world being dead, no, not really. Neither one person nor one new genre (that Doom, or rather Castle Wolfenstein started) really prevented any major catastrophe IMO. There were so many different genres that I find it interesting that for a while FP shooters were so much dominating game scene. Why were gamers like lemmings?
What would have happened is that someone else would have sooner or later (sooner I'd guess) came up with same idea and decent enough implementation. Really, the idea of using 1st person perspective and realistic-enough looking 3D display was nothing new. Implementation was innovative, and gave just enough advantage for ID to be the first one to get it done well enough; otherwise it would have taken bit longer for CPU speed to catch up.
But more to the point... you can't really even publish your copyrighted work, unless you are rich, or have produced something others (publishing company) thinks can make them rich (companies that are very likely to be "rich" compared to most individuals). Meaning that most publishable works can certainly earn enough money to allow them to be copyrighted for as long as they remain productive. So, even a poor (at least originally poor) author/composer can get copyright extended, as long as work in question is considered good enough (as measured by people buying copies of the work).
Plus, don't confuse copyright with authorship. Even if one doesn't hold a copyright on, say, a book, doesn't mean (s)he has lost authorship. That person just can not prevent others from reproducing (aka copying) the work in question.
Even more importantly, when interpreting constitution, judges do try to consider not just literal meaning, but also intention. Clearly intention wasn't to allow anything but whatever was considered reasonable limited period of time.
Well, that's only every 20 years or so, so as far as conspiracy theories go this isn't all that good. Or perhaps it was meant to be a joke?
Sure it is his choice, if he chooses to make that. That doesn't guarantee it happens, but it's his choice. But where did he claim it's his right, or that he's entitled to it? He was barely saying he intends not to try to change the way things are, not that it's a god-given birth right, nor that it's the way things even should be.
I'm not saying you need to agree that things are just fine, but don't fight the strawmen, and read more into parent post than is there.
Note that while in general I mostly agree with what you were saying before this statement, I'm reasonably sure this statement is an urban legend. World's food output is not dramatically higher than need (currently), by any standards. It's barely keeping up with demand. Potential for food production is bigger, but that would require lots of changes, plus absolutely requires use of 'industrial' approach to food production (unlike many "green food production" proponents believe... industrial fertilizers at least are pretty much a must, just need to make sure usage is optimized so as not to cause problems with nutrient washout etc).
One factor behind myth of "having way too much food yet still hunger" is that in world market there is always "surplus" food on sale, and they are oftentimes hard to sell, too. Amounts, however, are reasonably small (compared to total consumption), and problem is more about poorest, most needy, countries not having means to purchase the goods than about actually having real excess food production capacity.
Plus, at least here one can still go flipping burgers, and earn more than those 5 billion others (I might claim it's less than 5, perhaps 4, what with upcoming middle class of China and India, but what the hell).
And even if you were _really_ poor, for many foreigers stereotypic image of, say, american, would still be rich white anglosaxon middle-aged man. So his statement would be true. :-)
Thing I was wondering, however, was that salary rates for US workers seemed awfully low:
"fifth of those in the United States, with programmers earning $250 to $700 a month, compared with $1,600 to $3,600"
I didn't think I was hugely overpaid, but I earn twice as much as "high end" programmers are supposed to earn.
Perhaps that was just a typo or something, since they do mention hourly wages of 45$ later on (that'd translate to ~7200 instead of 3600).
Anyway; if you do read the page, try also talking to someone on H1B (and/or co-worker or manager of some such person) and see if reality matches the sensationalist claims of h1b."info", to get balanced view of the whole mess.
For what it's worth my understanding is that it's not worth envying H1B peon's position, and claiming they are here just leeching here stealing jobs. Job security with H1B is really really bad; visas really are temporary, and if you do lose your job, finding a new one is next to impossible (currently, didn't use to be, but that was same for 'natives')
Plus, right now managing to get a NEW H1B visa is next to impossible for most programmers. Extending existing ones (or transferring) is ok, but proving there is shortage of qualified american programmers is tricky. Like it should be, I might add. All this ranting about "increasing H1B visa quota" is so clueless... they didn't even have to use all 200K they had for last year, and even getting past 65K might be tough call (that quota is only used against new H1B visas, not extensions).
Finally, there's one more thing many people even here are unaware of. Since it's possible to apply for a green card after getting a H1B visa, majority of all current or previous H1B holders are or are becoming permanent residents (and eventually citizens if they choose to). So, "closing the gates" wouldn't make that much of a difference either way. Influx of foreign work force has already happened in high tech. Many of those indian DBAs or chinese programmers (etc. etc.) are actually NOT on H1B any more.
But that's not what he was saying. Try to avoid knee-jerk reactions. What he (and most everyone else) is saying is that presentation information should be available somehow. No one's asking it to necessarily be embedded in content, just be available, either on different file, or be defined as static specification (like HTML default rendering suggestions or de facto 'standards').
Without presentation information there just is no way to know how to adequately render content for human usage. And MS certainly stores and uses such information, but apparently they just don't want to share that... since that would make the somewhat open format much much more useful for competitors.
MS should just look at how OpenOffice folks did their format. It's not perfect (far from it), but it adequately both stores presentation information and separates actual structured content.
Perhaps because they could cut costs and earn more money? AMD chips have traditionally been less expensive (comparing similarly performing CPUs). Companies are known to be interested in getting bigger profit margins you know, and getting price of one of costlier components should help.
Although I do agree with your points in general, it's worth nothing that for "non power users" pretty much any OS does ok nowadays (any as in a Unix/Linux, MacOs, Windows). All have decent browsers, enough plugins for multimedia content, basic word proc. and spreadsheet apps, email clients etc. etc. So, general attitude from "mainstream" people really should just be "whatever".
So, what you have here is essentially a reply to strawman argument.
Well I don't know if I'd call a watered down 500mhz MIPS based chip "high powered". Maybe once you lash 128 of them together you'll have a decently powered box, but individually, it's way less than yesterdays tech.
You are ignoring his point, though; export restrictions aren't aimed at preventing "power users" in China from getting fastest running single-processor system, but from building "super computers", that are without an exception (massively) multi-processor systems.
And for such systems, distinction between 3 GHz vs. 500Mhz (assuming CPU power was directly proportional to clock rate) is not all that critical. Especially when you combine SMP with distributed systems.
Then again, export controls don't really protect against this even without this new CPU; it'd be possible to use freely traded Intel/AMD chips as well. However, it may be easier to design the big system using these CPUs if/when they are properly designed for SMP systems (since x86 procs in general are not designed for bigger than, say, 4-way SMP systems).
It is entirely possible to do totally different design trade-offs if raw computing power is not THE goal. So, although this is just speculation, here are some potential improvements it could offer:
Basically it all comes down to this: for most things, any modern CPU (even one that runs at 500mhz) is enough. Thus, 3Ghz Octium shouldn't be much more interesting than 0.5Ghz Dragon (if it wasn't for advertising). And if raw CPU power is meaningless, currently "secondary" things like price and power consumption could become real factors in designing what to use.
Interesting thing, however, is why they chose MIPS ISA over, say, Alpha? Both are reasonably clean RISC designs... perhaps it's just that MIPS is extensively documented in quite a few basic processor design books? Hell, even I had to design and implementation of the first MIPS processor in VHDL (for my proc. architectures course), and I'm not an EE major.
This because leverage means (in more general sense than in its strict definition of using a lever to move an object) using smaller force (or amount etc) to achieve similar effect as using bigger force (etc). This is what a lever does -- it "converts distance to force" (same amount of work is done by either applying smaller amount of force over longer distance as bigger force over shorter distance).
And that's why many people (myself included) do think that majority of usage cases for "leverage" are in fact buzzspeak. People who think leverage is "just fancier way of saying 'use'" clearly do not understand meaning of the word.
Good things about "intra-box scalability" include:
Good point... and I think same applies to genres in computer games and literature (although to a lesser degree in literature I guess). I might argue that there's always room for _some_ improvement, but after a while it probably makes more sense for most people to either move on (change genre, or break its constraints), or to enjoy what has been achieved as is.
I disagree; I think the idea of "everything's already been done at least twice" is a common phallacy. Some people claim that all good music has already been done, or all good movies, or even all good paintings.
And yet amount of permutations for basic components of music (melodic, harmonic, rhythmic), literature (themes, characters, time, style) or, computer games (ones similar to literature), is pretty much infinite. There will _always_ be room for new things in any of above-mentioned forms of art.
I agree with the article. Lack of innovative games has more to do with business objectives of predictable revenue than with not having room to explore that limits original games. 20 years ago technology was limiting things much more; nowadays it's almost a moot point, at least from game idea point of view. Any interesting non-novelty gameplay idea can probably be implemented on standard gaming system of choice. But since coming up with a new idea IS more difficult than refining an existing idea (I'm not arguing otherwise), the risks associated just make it so much more compelling to "just write yet another sequel of a hit".
Funnily enough, this is just one of those problems with short-sighted businesses. Without new innovative hits, there won't be chance for new predictably profitable sequels. You can only do so many sequels from a certain theme, with lowering profitability... and then have to move on.
Well... this is off-topic, but... why on earth would Saddam burn "his own" oil fields?
What puzzles me is why people (even smart ones) make the logical leap from "he burnt Kuwaits oild fields" to "he will burn Iraq's oil fields", without thinking of obvious differences.
In case of Kuwait, SH was just playing sour loser and trying to prevent Kuwaitians (plus allies) from benefiting from their oil, when we knew he'd be asskicked out. In case of Iraq, it's "his" oil. Iraq will still have the oil, no matter what the outcome (unless you really believe in extreme conspiracy theories). There's hardly point in burning any of it. It won't slow US troops down; it's like panting in your pants to prevent cops from arresting you. Unlikely to work to say the least.
I mean; whenever the order of magnitude difference between things exceeds a threshold (whatever it may be for the person in question; anything from 2 to 6), there is just no way to really put things in perspective, by using silly analogies. And LOC seems like just another such analogy (perhaps even worse than the older "page of text" used instead of kilobyte... assuming 8-bit chars).
The interesting part behind XML is, you don't have to invent your own syntax and implement your own parser. As a result other applications can quite easily access your data, too.
Well. XML is human-readable, but not automatically human understandable. My pet peeve regarding most voal XML hype is the claim "XML is self-documenting" (most often used with SOAP but sometimes generalized to all XML data). Something that just isn't true. It's just like claiming C programs are self-documenting; after all, we do have common syntax with C as well as with XML.
XML can be understood by anyone only at syntactic (structural) level; you can see how elements are nested. What is not automatic is semantic meanings. Without knowing which XML application (as in XML terminology; not piece of code, but a language like XHTML or RDF) one is using (and knowing its semantics), it's not much use to know structure. It's like understanding syntactic structure of english language without knowing any of the words used. Knowing that a sentence has structure "noun predicate objective" isn't enough to understand what is actually being said. You need to know the words (vocabulary) and their meaning.
So... while using XML does simplify task of interchangeability a bit (on top of Ascii/Unicode, that solves another lower level problem, on top of de factor 'standard' of using 8-bit byte instead, of, say, older 9-bit words, and... ok, you get the idea), it does only solve first 10% or so. Agreeing on common XML application solves perhaps next 50% or so. And then actually specifying remaining semantics (that can not be expressed by DTD or even XML Schema) is the final part.