No. Dot-com burst wasn't due to geeks, it was... guess who? The suits. They thought there were lots of golden "business opportunities" out there. This is obvious and I feel silly even pointing that out, but geeks would have been happy just putting up a simple web site and be done (or more likely, keep on tinkering). But suits not; they needed to get 100+ employee sweat shop creating web site that gives stuff out for free, and then selling these brilliant ideas to other suits and unfortunate investors.
And you say geeks are intrinsically bad at earning living, but suits not? And then point at dotcom bubble as an example. Oh boy.
Of course it's worth pointing out that it was epidemic of "green suits", not seasoned pointy-haired ones, that kept trainwreck going... at least initially; towards the end even normally level-headed people joined the party. And then it all imploded.
Just like sports salary's. You don't *have* to pay a given player $20 million, but if you don't there is another team that will. What's wrong with that?
Assuming skills of executives were tangible and measurable -- like they are in sports -- that'd be understandable. Wrong, unfair, but understandable.
However, whereas you can compare performances of athletes, especially since sports that can afford big money are mostly team based, performance of a CEO is impossible to reliable and objectively measure. If you could compare a dozen CEO under identical conditions, you would see the difference. Too bad it can not be done without access to parallel universes.
What this leads to, is the myth of magical super-powered leaders. Myth in that there's little connection to reality. One man does not usually make a huge (positive) difference in big corporations, not even the leaders. And one could
say, leaders especially do not make positive difference, at least not when they are compared to relative high compenstaion.
Still, you are right, in that the line of thinking you mention is still in fashion... amazingly, after dotcom burst and accounting scandals. And even though there's lots of criticism toward iconizing big high-salaried leaders... many people realize that really good leaders are often not in the limelight, don't have big mouths nor astronomical salaries. And interestingly enough, they are often committed to long-term prosperity, and do NOT do jobhopping although they would be in good position to do just that.
Yeah, just like Windows users were saying around '95, "who needs pre-emptive multitasking, I've never needed more than co-operative multitasking" (and Mac users until Mac Os X). Or linux users wondering why one needs a journaling file system, up until three of them get mature enough. And when the system does get what was formerly denounced as "unnecessary toy", it suddenly becomes indispensable commodity.
Now, SMP is not necessarily something people "start needing"; you don't usually realize one day that "gee, now I really need SMP". For many things it isn't strictly required, like you say -- after all, most systems only scale up decently to 4- or perhaps 8-way systems, and common rule of thum is that you get about sqrt(num_procs) output (assuming you get enough CPU load for all processors)... and so you can same amount of work done by getting twice as fast CPU (instead of 4-way SMP system).
However, having SMP capability as an option is a Good Thing. For servers it allows nice high-end scalability (esp. with Sun boxes, E10K and co. wouldn't rock if it wasn't their SMP-scalability coupled with kickass I/O.. but those beast scale well past 8 CPUs). For desktop systems it allows for better interactivity (especially on traditional unix[ish] system that have batch-job oriented
scheduling), smoother UI.
Yeah, I don't understand the popularity of screwdrivers... Why can't we all just get along and use hammers for getting screws installed, TVs fixed and kids disciplined! Why waste time and resources on developing and learning how to use multiple tools?
java: int, Integer -- what the **** is that supposed to be good for?
Performance. To get same level of performance from boxed primitives as 'native' primitives, compilers and VMs need to do more work. Nothing impossible, but originally Java was KISS - oriented language (auto-boxing would make it bit simpler to use, but more complicated internally).
As to multiple inheritance... I don't quite see how you think interface makes anything into a semantic problem. The thing interfaces are missing (from MI) is the default implementation, and that's a practical, not semantic problem?
Re:The day of a single very powerful CPU is over..
on
End In Sight For Alpha
·
· Score: 2
You may think so, but somehow how distributed computing has "been coming" for decades now. Just like neural networking, intelligent agents and quantum computing (ok, last one not as long as others). Every few years new modification of DC tags along; right now it's "grid computing" (throw in whatever CPUs you got an manage them transparently), tomorrow it'll be something else (at least name changes for marketing).
There are a few niches where distributed computing is really useful. So far best examples have been, let's see, cracking encryption and searching for ET.
There are many reasons why distributed computing is not "the" general solution for need-for-speed... here are some of the obvious ones:
In many cases, algorithms need enough messaging that overhead just kills any benefits from scaling.
It's much more difficult to parallelize algorithms to scale properly, outside the obviously scalable cases (like cracking encryption)
Maintenance for N individual systems is much more expensive than maintaining lesser number of more powerful systems.
And finally, like somebody else mentioned, perhaps you are mixing SMP and DC here a bit; SMP has fewer problems and is where I see the future, more than with distribution (actually, more than SMP, processor-level multi-threading).
Amen, Greg Bear is one of my all time favourites, along with Philip K Dick. I first read "Blood music" (novel version, later read the original short story)... nice thought-provoking story.
ps. Anyone remember who was the author of novel "Greenhouse" (or whatever it was in english... only read the translation)? I remember it was written decades ago, and still it's got to be one of best sci-fi novels written... the idea of humans brains actually being just a parasite is at the same time funny, scary, and deeply ironic. And the idea of global warming (which lead to humans brains leaving human skulls as it became too warm for them!) wasn't much of a theory back then.
One thing I've noticed, related to credibility of books (movies etc) is that story does not have to be realistic, but it has to be somewhat believable. For non-experts in the related field many more alternatives appear equally believable, and one can choose what fits best with the story.
This applies to all kinds of things, naturally; most people even consider movie gunshot sounds more authentic than real ones (ditto for most other sound fx).
As to original comments, I'm not sure realistic people are absolutely necessary. I'd vote for interesting, varied, multidimensional characters, but I guess authors who can create such characters are usually more than able to make them believable as an added bonus.:-)
In theory, H1B can NOT be used to get salaries down -- the very regulation for H1B mandates paying market rate salary for H1B employees. In practise, there is some room for 'optimizing the salary', since data median salary is based on is from couple of years ago... but in this economy that's not as helpful as it was a year or two ago.
Two (minor?) problems with H1B visas one would have to solve are the fact that H1B is aimed at technical employess (but there is a separate visa class for managers, so just replace H1 with whatever the code for managers is... or is there just one for execs?), and the limitation on terms of employment. Latter restricts both job title and job location for employee in question; in theory H1B employee can not be promoted or moved to another position without filing a new application.
Oh, by they way, today they also voted for a $4,000 per year salary increase for each member of Congress.
There may be some symbolic value about this raise, but fiscally it's pretty meaningless (about 2M$/year?). Contrast that to 800k people; each would get lofty 2.5 dollar benefits if it was given to them instead.
It's very popular whining subject to complain about direct compensation for politicians. I'd prefer salaries to be high enough to both attract at least mediocre talent, and ensure they don't need extracurricular activities to get them by.
I am much more concerned about indirect compensation (both donations and after-the-term board positions etc) politicians get than about money they more or less honestly earn by representing some part of the population (voters of the district they were voted from).
In fact, even better might be allocating money for election campaigning, based on existing number of seats party has. This is commonly done in european countries, as even though it sucks having to pay for campaigning with your tax dollars, at least it makes it possible (if not guarantee) that it's not just mr. Big Bucks that they listen, because mr. Bucks paid for the campaign.
Most of the laws that effect our lives should be at the state level, where people have more control (most states at least have props, people can easily move to another state). For example, if ill Californians want to smoke pot, they should be allowed to, but unfortunatly the will of folks in other states dictate otherwise.
Why restrict it to state level? Why not let each county decide all the laws locally? Hell, how about letting each citizen decide whatever (s)he wants to do! Why need laws at all!
Actually I don't think that getting law making process more local would be a good thing at all. As much as compromises suck -- at high level you always need compromises because there are so many people involved -- the alternative, local populistic or mob rule, would suck even more. That's also why having just 2 (significant) parties is bad -- it's only one less than in totalitarian countries (plus, with just 2 parties, leverage power of fringes of both parties, currently especially GOP, is way too big). If there were more parties -- most democratic countries have 3 or 4 major parties, plus often couple of minor additional ones.
It's not a coincidence that right wing extermist groups (anti-abortion, gun liberals, bible thumpers) want more local law making power. It'd be much easier to leverage voice of minority that way. What is more interesting is why leftist groups are not doing the same. They are proposing as many "extreme" local laws as the other side (in areas that have strong leftist tradition), but they are not fighting to move more power there.
Yah, imagine that. You work for company and they try to prevent you from switcing to the competition directly after seeing the prized assets. Imagine that. You mean you cant work for company A for 30 days, see all their trade secrets, and then go the company B 10 days later? REally? 'elp, 'elp, I'm being oppres'd!
That's silly. Moving to a competitor does not mean handing out trade secrets. Not that most peons would really know anything secret in the first place. But even giving "just confidential" information would be a decent grounds for legal action, plus your new employer would make 150% sure you do NOT make THEM vulnerable to legal action, and thus explicitly request you to make sure you do NOT reveal such information.
Now, I've been employed by 2 US software companies; one had clauses preventing joining direct competitor's in 3 specified states (which pretty much indicated nr. 1 company that's forbidden); the other one had no restrictions whatsoever. Both had fairly draconian IP restrictions, which are enough to protect against supposed "trade secret" violations.
And FWIW, my current employer has made "proper business conduct" classes obligatory for all engineers, and they specifically dealt with subject like this, emphasising dangers for THEM for employees willingly giving out classified info from previous jobs.
Or you could do as Canada does and simply integrate them into your society as citizens,
rather than immigrants, teach them, and make sure they do as good a job as any other
citizen, for the same level of pay.
Believe it or not, but that's pretty similar to what USA does, although methods are slightly different. USA attracts highly-educated professional (through H1 program), allows them to apply for a green card and potentially become citizens. The biggest difference is just that this is more "laissez faire" system; there's not much in a way of support (teaching language, culture, trying to speed up integration etc). Immigrants are free to integrate, in their own time and using their own money. And the 'carrot' is just that if you do not assimilate, your social status is bit limited ("those weird foreigners"), despite the fact you are a citizen (or permanent resident). The last part is somewhat race-dependant; even though USA in general is not very racist country, skin colour still counts (that is, for white europeans things are still easier than for, say, people coming from Asian countries).
Most/many people seem to think H1(b)s are hastily educated second-rate cheap-o workers that crank out code (or whatever), and just leech in the states. However, when applying for H1B, employees have to specify market rate salary (only caveat being statistics used to be couple of years behind, meaning it was market rate of yesteryear); have to proof no qualified citizen was available (and can NOT discriminate based on salary when trying to hire 'natives').
Plus, in 3-5 years, most H1Bs have either gotten their green card, are in process of getting it soon, or have left (having been laid-off and not found a new H1B eligible job).
Yes, it is hard to believe a government would actually stand up to its ideals; ideals of considering death penalty being barbaric act no civilized society should impose on anyone.
It sure seems much more likely french gov't just took the word of this very likely guilty person.</sarcasm>
Believe me, french people despise murderers just as much as anybody else, but:
They (like many europeans) have doubts about fairness of US legal system (whether that's fair or not is another issue altogether)
They (majority of people, not nearly all) think death penalty is about as honorable act as slavery, torture or cannibalism.
It's been 20 years since Symbolics time (ie. this thing has been going on for a while now), so doomsday prophecies seem bit silly. But more importantly, there are lots of examples of exactly the opposite -- IBM PC clones were what made them the standard, not IBM that created them. Photoshop clones, in many ways, just make the 'original' product more dominant aping its UI and feature set.
Plus, creating a better mousetrap has often been a good recipe for success: the common conception is that that's what made Japan a hi-tech super power (and later on, Taiwan).
In fact, I would claim this is a very good and sustainable situation. Original innovators have to keep on innovating, to stay ahead of copycats, instead of milking the monopoly fees (there exists one well-known exception to the rule... for now). And customers benefit; through improvements in products, by lower costs (when earlier feature sets become a commodity), by more de-facto standardization.
There is (and will be) lots of incentive to be the first one, to create the pioneering product. Arrival of cloners is usually a good indication that you did something right. And that happens on every other area too, not just in software development; it's just that software development is rather light on capital requirements. It's one of few things you do not need heavy machiner or financial backing to start.
Well, fair enough. I completely agree on importance of interoperability. However, I think reliance on a proprietary file formats that were never designed to be interoperable -- such as MS Office format -- is a mistake. Even though it is difficult (some would claim impossible or impractical) to move to something truly standard (especially no clean alternative exists...), that's how it should be done. And that's why I don't expect complete facsimile transfer from/to Word; it was never intended to happen, and is not something Microsoft would even like to see (if they did, they would drive adoption of Word document format as a standard).
As to cloning, I'm afraid one really has to clone all functionality, completely, to get close to 100% compatibility. All oddities, workarounds and bugs big baggage-ridden apps (like Word) have, have to be duplicated, since file format is very tightly integrated with the feature set.
People can complain as much as they like about HTML, but it is much more ubiquitous and better standard than, say, Word. It wasn't designed to map feature set of a single app (well, occasionally it was, first by Netscape, then by MS... and that causes biggest flaws in HTML), and thus allows reasonable interoperability. Same thing should be done with word processors -- decide on common file format and compete on implementations. And hopefully also learning from HTML mistakes.
So true. When we tested Word import capability in -99 or so, we tried 3 products; different version of Word (ie. later version reading output from older one) on Windows, Applixware Office on Linux, and WP on Linux.
Winner was Applixware, Word came in second, and WP was the last. All of them produced pretty close matches, though.
If you want Honda Civic, buy Honda Civic. If you want MS Word, buy MS Word license. I don't see the point of replacing Word with an exact clone, nor that this is a pre-requisite for success of an office suite.
If OpenOffice gains market share from MS Word, it will be because of its licensing, availablity on multiple platforms, good enough compatibility with other office suites, and its user-friendliness. It's not because it can just clone exact feature set of Word, including 99.99% reliable conversions (100% is impossible for any apps, including different versions of MS Word itself).
For me, "reasonably good job" is perfectly acceptable, and I'd rather they not waste too much time on reverse-engineering their competitor's file format, but concentrate on improving their own product.
Are you saying that the problem is that margins can not be reliably specified, or that there are 'significant' (for anal-retentive lawyers at least) rounding errors even when things are fully specified? I guess I need to ask my printing guru friend for details (he actually currently writes PCL drivers for a commercial desktop publishing product, albeit prefers postscript)...
...
But OpenOffice didn't include the
same fonts as my documents, forcing the program to select alternate fonts that messed
up the spacing between words.
When I fixed the spacing and re-opened the documents in
Word and PowerPoint, the spacing was now messed up by the return to the original font.
No offense but this is an area where not much more can be done. You are taking in "alien" content, modifying it natively, then once again converting to alien format. These kinds of conversions are lossy by nature, esp. since MS Office formats are proprietary ancient messy "standard" (proprietary although not secret any more). This is especially true with layout related information, as file format doesn't really define how to use information; and also because fonts themselves usually can not be freely shipped. To get truly ubiquitous precise layout, Postscript/PDF should be used. In future hopefully a real office document interchange format emerges; OpenOffice has been active in this area (although it's not likely their XML-based format will become standard, it hopefully leads the way, showing how standard could be defined).
Unfortunately, for people who have to work with Office docs end-to-end, things will never be very easy unless they stick to using MS Office. But that's only a small part of functionality OpenOffice (and StarOffice providew), and judging the suite solely based on this feature is rather unfair for it.
True. And this is exactly the reason why consumer need to have option of getting Linux pre-installed with, say, Compaq or Gateway... and why Microsoft is fighting that with all of their might, although behind the scenes.
Installing Windows on crappy h/w is a bit as well, what with trying to hunt down working drivers and all. It's just that with name brands this is already being taken care of.
think patent system is Bad For You (tm) and doesn't want to patent the damn thing... but wants to prevent others from patenting it
lacks resources to patent the thing (no money, bad at patent legalese)
has no plans to use it, thus patent fees being total wate
might want to avoid hassles with employer; with patent (s)he might well lose patent to employer (even if inventing it outside scope of work employer might just try its luck at court). Disclosing the thing can not be undone, and is less likely to get the knee-jerk reaction (depends totally on employer of course)... there's nothing left to take over (idea is still there of course, but not patentability)
Additionally, material support could be interpreted to include publicity and propaganda.
No, "material" was probably used exactly as to prevent it from applying to non-material things like moral support, publicity, propaganda and such.
And as to proxy servers... it's only "hosting stuff" from anal-retentively physical point-of-view. Semantically it's just doing local buffering. Not that some smart lawyers couldn't twist it to appear to be something much more sinister.:-/
...GIMP would seriously benefit from having being a single app/MDI type of thing
This is matter of taste I know, but personally I just can't stand MDI. I absolutely hate that, and I know I'm not alone here. I do know many like it, as well. It's not widely used outside Windows world is it? (Java supports MDI windows in Swing, but natively?). Then again, I'm in minority in that I never full-size windows to overtake the whole screen either.:-)
Still, UI certainly could/should be improved, and MDI is not the only way to improve handling of multiple windows (ie. menu/toolbars on windows instead of global tool etc. window). If possible perhaps having option between MDI and multiple 'real' windows (without container like in MDI)... I remember seeing such apps, but can't remember names.
Slashdot Mirror
And you say geeks are intrinsically bad at earning living, but suits not? And then point at dotcom bubble as an example. Oh boy.
Of course it's worth pointing out that it was epidemic of "green suits", not seasoned pointy-haired ones, that kept trainwreck going... at least initially; towards the end even normally level-headed people joined the party. And then it all imploded.
Assuming skills of executives were tangible and measurable -- like they are in sports -- that'd be understandable. Wrong, unfair, but understandable. However, whereas you can compare performances of athletes, especially since sports that can afford big money are mostly team based, performance of a CEO is impossible to reliable and objectively measure. If you could compare a dozen CEO under identical conditions, you would see the difference. Too bad it can not be done without access to parallel universes.
What this leads to, is the myth of magical super-powered leaders. Myth in that there's little connection to reality. One man does not usually make a huge (positive) difference in big corporations, not even the leaders. And one could say, leaders especially do not make positive difference, at least not when they are compared to relative high compenstaion.
Still, you are right, in that the line of thinking you mention is still in fashion... amazingly, after dotcom burst and accounting scandals. And even though there's lots of criticism toward iconizing big high-salaried leaders... many people realize that really good leaders are often not in the limelight, don't have big mouths nor astronomical salaries. And interestingly enough, they are often committed to long-term prosperity, and do NOT do jobhopping although they would be in good position to do just that.
He must have been referring to former east Germany, which was referred to as DDR ("Deutsche Democratic Republic" or something?)
Now, SMP is not necessarily something people "start needing"; you don't usually realize one day that "gee, now I really need SMP". For many things it isn't strictly required, like you say -- after all, most systems only scale up decently to 4- or perhaps 8-way systems, and common rule of thum is that you get about sqrt(num_procs) output (assuming you get enough CPU load for all processors)... and so you can same amount of work done by getting twice as fast CPU (instead of 4-way SMP system).
However, having SMP capability as an option is a Good Thing. For servers it allows nice high-end scalability (esp. with Sun boxes, E10K and co. wouldn't rock if it wasn't their SMP-scalability coupled with kickass I/O.. but those beast scale well past 8 CPUs). For desktop systems it allows for better interactivity (especially on traditional unix[ish] system that have batch-job oriented scheduling), smoother UI.
Yeah, I don't understand the popularity of screwdrivers... Why can't we all just get along and use hammers for getting screws installed, TVs fixed and kids disciplined! Why waste time and resources on developing and learning how to use multiple tools?
Performance. To get same level of performance from boxed primitives as 'native' primitives, compilers and VMs need to do more work. Nothing impossible, but originally Java was KISS - oriented language (auto-boxing would make it bit simpler to use, but more complicated internally).
As to multiple inheritance... I don't quite see how you think interface makes anything into a semantic problem. The thing interfaces are missing (from MI) is the default implementation, and that's a practical, not semantic problem?
There are a few niches where distributed computing is really useful. So far best examples have been, let's see, cracking encryption and searching for ET.
There are many reasons why distributed computing is not "the" general solution for need-for-speed... here are some of the obvious ones:
And finally, like somebody else mentioned, perhaps you are mixing SMP and DC here a bit; SMP has fewer problems and is where I see the future, more than with distribution (actually, more than SMP, processor-level multi-threading).
ps. Anyone remember who was the author of novel "Greenhouse" (or whatever it was in english... only read the translation)? I remember it was written decades ago, and still it's got to be one of best sci-fi novels written... the idea of humans brains actually being just a parasite is at the same time funny, scary, and deeply ironic. And the idea of global warming (which lead to humans brains leaving human skulls as it became too warm for them!) wasn't much of a theory back then.
This applies to all kinds of things, naturally; most people even consider movie gunshot sounds more authentic than real ones (ditto for most other sound fx).
As to original comments, I'm not sure realistic people are absolutely necessary. I'd vote for interesting, varied, multidimensional characters, but I guess authors who can create such characters are usually more than able to make them believable as an added bonus. :-)
In theory, H1B can NOT be used to get salaries down -- the very regulation for H1B mandates paying market rate salary for H1B employees. In practise, there is some room for 'optimizing the salary', since data median salary is based on is from couple of years ago... but in this economy that's not as helpful as it was a year or two ago.
Two (minor?) problems with H1B visas one would have to solve are the fact that H1B is aimed at technical employess (but there is a separate visa class for managers, so just replace H1 with whatever the code for managers is... or is there just one for execs?), and the limitation on terms of employment. Latter restricts both job title and job location for employee in question; in theory H1B employee can not be promoted or moved to another position without filing a new application.
There may be some symbolic value about this raise, but fiscally it's pretty meaningless (about 2M$/year?). Contrast that to 800k people; each would get lofty 2.5 dollar benefits if it was given to them instead.
It's very popular whining subject to complain about direct compensation for politicians. I'd prefer salaries to be high enough to both attract at least mediocre talent, and ensure they don't need extracurricular activities to get them by. I am much more concerned about indirect compensation (both donations and after-the-term board positions etc) politicians get than about money they more or less honestly earn by representing some part of the population (voters of the district they were voted from).
In fact, even better might be allocating money for election campaigning, based on existing number of seats party has. This is commonly done in european countries, as even though it sucks having to pay for campaigning with your tax dollars, at least it makes it possible (if not guarantee) that it's not just mr. Big Bucks that they listen, because mr. Bucks paid for the campaign.
Why restrict it to state level? Why not let each county decide all the laws locally? Hell, how about letting each citizen decide whatever (s)he wants to do! Why need laws at all!
Actually I don't think that getting law making process more local would be a good thing at all. As much as compromises suck -- at high level you always need compromises because there are so many people involved -- the alternative, local populistic or mob rule, would suck even more. That's also why having just 2 (significant) parties is bad -- it's only one less than in totalitarian countries (plus, with just 2 parties, leverage power of fringes of both parties, currently especially GOP, is way too big). If there were more parties -- most democratic countries have 3 or 4 major parties, plus often couple of minor additional ones.
It's not a coincidence that right wing extermist groups (anti-abortion, gun liberals, bible thumpers) want more local law making power. It'd be much easier to leverage voice of minority that way. What is more interesting is why leftist groups are not doing the same. They are proposing as many "extreme" local laws as the other side (in areas that have strong leftist tradition), but they are not fighting to move more power there.
Yah, imagine that. You work for company and they try to prevent you from switcing to the competition directly after seeing the prized assets. Imagine that. You mean you cant work for company A for 30 days, see all their trade secrets, and then go the company B 10 days later? REally? 'elp, 'elp, I'm being oppres'd!
That's silly. Moving to a competitor does not mean handing out trade secrets. Not that most peons would really know anything secret in the first place. But even giving "just confidential" information would be a decent grounds for legal action, plus your new employer would make 150% sure you do NOT make THEM vulnerable to legal action, and thus explicitly request you to make sure you do NOT reveal such information.
Now, I've been employed by 2 US software companies; one had clauses preventing joining direct competitor's in 3 specified states (which pretty much indicated nr. 1 company that's forbidden); the other one had no restrictions whatsoever. Both had fairly draconian IP restrictions, which are enough to protect against supposed "trade secret" violations.
And FWIW, my current employer has made "proper business conduct" classes obligatory for all engineers, and they specifically dealt with subject like this, emphasising dangers for THEM for employees willingly giving out classified info from previous jobs.
Believe it or not, but that's pretty similar to what USA does, although methods are slightly different. USA attracts highly-educated professional (through H1 program), allows them to apply for a green card and potentially become citizens. The biggest difference is just that this is more "laissez faire" system; there's not much in a way of support (teaching language, culture, trying to speed up integration etc). Immigrants are free to integrate, in their own time and using their own money. And the 'carrot' is just that if you do not assimilate, your social status is bit limited ("those weird foreigners"), despite the fact you are a citizen (or permanent resident). The last part is somewhat race-dependant; even though USA in general is not very racist country, skin colour still counts (that is, for white europeans things are still easier than for, say, people coming from Asian countries).
Most/many people seem to think H1(b)s are hastily educated second-rate cheap-o workers that crank out code (or whatever), and just leech in the states. However, when applying for H1B, employees have to specify market rate salary (only caveat being statistics used to be couple of years behind, meaning it was market rate of yesteryear); have to proof no qualified citizen was available (and can NOT discriminate based on salary when trying to hire 'natives').
Plus, in 3-5 years, most H1Bs have either gotten their green card, are in process of getting it soon, or have left (having been laid-off and not found a new H1B eligible job).
It sure seems much more likely french gov't just took the word of this very likely guilty person.</sarcasm>
Believe me, french people despise murderers just as much as anybody else, but:
It's been 20 years since Symbolics time (ie. this thing has been going on for a while now), so doomsday prophecies seem bit silly. But more importantly, there are lots of examples of exactly the opposite -- IBM PC clones were what made them the standard, not IBM that created them. Photoshop clones, in many ways, just make the 'original' product more dominant aping its UI and feature set.
Plus, creating a better mousetrap has often been a good recipe for success: the common conception is that that's what made Japan a hi-tech super power (and later on, Taiwan).
In fact, I would claim this is a very good and sustainable situation. Original innovators have to keep on innovating, to stay ahead of copycats, instead of milking the monopoly fees (there exists one well-known exception to the rule... for now). And customers benefit; through improvements in products, by lower costs (when earlier feature sets become a commodity), by more de-facto standardization.
There is (and will be) lots of incentive to be the first one, to create the pioneering product. Arrival of cloners is usually a good indication that you did something right. And that happens on every other area too, not just in software development; it's just that software development is rather light on capital requirements. It's one of few things you do not need heavy machiner or financial backing to start.
As to cloning, I'm afraid one really has to clone all functionality, completely, to get close to 100% compatibility. All oddities, workarounds and bugs big baggage-ridden apps (like Word) have, have to be duplicated, since file format is very tightly integrated with the feature set.
People can complain as much as they like about HTML, but it is much more ubiquitous and better standard than, say, Word. It wasn't designed to map feature set of a single app (well, occasionally it was, first by Netscape, then by MS... and that causes biggest flaws in HTML), and thus allows reasonable interoperability. Same thing should be done with word processors -- decide on common file format and compete on implementations. And hopefully also learning from HTML mistakes.
Winner was Applixware, Word came in second, and WP was the last. All of them produced pretty close matches, though.
If OpenOffice gains market share from MS Word, it will be because of its licensing, availablity on multiple platforms, good enough compatibility with other office suites, and its user-friendliness. It's not because it can just clone exact feature set of Word, including 99.99% reliable conversions (100% is impossible for any apps, including different versions of MS Word itself).
For me, "reasonably good job" is perfectly acceptable, and I'd rather they not waste too much time on reverse-engineering their competitor's file format, but concentrate on improving their own product.
Are you saying that the problem is that margins can not be reliably specified, or that there are 'significant' (for anal-retentive lawyers at least) rounding errors even when things are fully specified? I guess I need to ask my printing guru friend for details (he actually currently writes PCL drivers for a commercial desktop publishing product, albeit prefers postscript)...
When I fixed the spacing and re-opened the documents in Word and PowerPoint, the spacing was now messed up by the return to the original font.
No offense but this is an area where not much more can be done. You are taking in "alien" content, modifying it natively, then once again converting to alien format. These kinds of conversions are lossy by nature, esp. since MS Office formats are proprietary ancient messy "standard" (proprietary although not secret any more). This is especially true with layout related information, as file format doesn't really define how to use information; and also because fonts themselves usually can not be freely shipped. To get truly ubiquitous precise layout, Postscript/PDF should be used. In future hopefully a real office document interchange format emerges; OpenOffice has been active in this area (although it's not likely their XML-based format will become standard, it hopefully leads the way, showing how standard could be defined).
Unfortunately, for people who have to work with Office docs end-to-end, things will never be very easy unless they stick to using MS Office. But that's only a small part of functionality OpenOffice (and StarOffice providew), and judging the suite solely based on this feature is rather unfair for it.
Installing Windows on crappy h/w is a bit as well, what with trying to hunt down working drivers and all. It's just that with name brands this is already being taken care of.
No, "material" was probably used exactly as to prevent it from applying to non-material things like moral support, publicity, propaganda and such.
And as to proxy servers... it's only "hosting stuff" from anal-retentively physical point-of-view. Semantically it's just doing local buffering. Not that some smart lawyers couldn't twist it to appear to be something much more sinister. :-/
This is matter of taste I know, but personally I just can't stand MDI. I absolutely hate that, and I know I'm not alone here. I do know many like it, as well. It's not widely used outside Windows world is it? (Java supports MDI windows in Swing, but natively?). Then again, I'm in minority in that I never full-size windows to overtake the whole screen either. :-)
Still, UI certainly could/should be improved, and MDI is not the only way to improve handling of multiple windows (ie. menu/toolbars on windows instead of global tool etc. window). If possible perhaps having option between MDI and multiple 'real' windows (without container like in MDI)... I remember seeing such apps, but can't remember names.