the last time I ran one one of the BSDs (netbsd 1.5) I nuked the TCP stack by doing massive amounts of NFS transfers over 10/100. That kind of stuff would never happen under Solaris.
BTW, that massive NFS transfer only took around 4 minutes before the stack died a terrible death, to the point where the netbsd box couldn't ping out due to some kind of buffer error. Oh well.
I can't tell if the references in this article are intentionally wrong so as to mock the ideas, or accidentally wrong. Either way, it's hilarious!
Francis Fukuyama is a historian, not an economist. His book, The End of History, sees "History" as the struggle between competing systems of economic organization. History is at an end (according to him) because after the fall of the Soviet Union, the alternate ideology of economic organization essentially debased itself.
Money was never based on physical assets, it was a proxy for stores of value that were difficult to carry around.
"Now it is just valued for its intrinsic value to people." This make no sense, nor does the argument in the paragraph. Wow!
Fiduciary doesn't mean what it's used as in the reply. Fiduciary: A person holding a position of confidence, such as a trustee, guardian or executor. Capital has become a person!
new economy: what new economy?
This reminded me of the old living color skit, where the guy would be spouting words that he didn't understand. All in all, a great post! The poster is a genius, subtly mocking not only the lack of understanding and reasoning ability of slashdot readers, but mocking in a way that appears to understand the underlying arguments of economic theory. Bravo!
General Magic and Active Paper shipped a handheld-based web browser that was wireless-capable in 1996. They also had POP3 and PPP access, if I remember right.
Before them, there was a web browser working on a HP-LX 100 or some weird handheld like that.
whoo! Go DirectTV! That's really classic!
Don't really understand why the DirectTV thing is so hard...can't they update the card on the fly, and use the last couple of bits from the last 5 seconds?
Those who have not worked or experienced the enterprise environment will not understand the question. In the enterprise environment, bugs are political issues, not technical issues. If you can say "Well, Microsoft said..." the CIO can go and hammer on Microsoft to get it fixed. This makes everyone happy: the tech doesn't have to deal with the problem, the CIO gets to abuse a big-ass vendor, and MS keeps the large revenue stream from the ELA. Ta-da!
The linux world believes it understands do-it-yourself support. However, when you're paying for support already, why the heck should your guys do the work? Politically, it's easier to blame a vendor than your own guys.
One other point, forgotten in the discussion, is linux provides no way to actually UPDATE 90,000 POS systems at once. Tivoli Enterprise runs on Linux, but I'm not sure what products are supported.
If you actually read the Florida Supreme Court decision, you'd retch in disgust. Their logic went something like this:
FL statue: the FL Secretary of State CAN certify results at her discretion, and there is a penalty associated with late filing of results that the FL Secretary of State accepts.
FL Supreme Court: because there is a penalty associated with filing late results, the FL secretary of state MUST accept late results; if late results were not acceptable, they would not be penalties associated with them, thus there is no discretion on the part of the FL Secretary of State when it comes to accepting vote counts.
US Supreme Court: huh?
What's lucky in this instance is the US Supremes were able to intervene because it's a Federal question. Most state cases that have logic like this probably aren't reviewed by anyone.
buy a new 10/100 switch, a wireless networking solution, or reinstall Windows 2000; any of them are a better use of your time than seeing this movie. This movie could have been squished into 10 minutes if the editor was more skilled and/or the director wasn't so full of himself. Heck, this guy makes Wim Wenders look like the poster boy for tight, cohesive movie-making.
I'd rather watch Until the End of the World 10 times than see 15 minutes of this film again...or get really, really high and watch Heat at half-speed, so as to really enjoy the lack of pacing.
Plus, the story is moronic. Maybe it's really intense for comic book fans who, say, stop at the sunday funnies, but for Sandman fans the plot is pretty lame. I mean geez...an antagonist who has to find/create his own protagonist. C'mon!
This isn't quite "Capital." I'm not sure which volume you read, but what I remember is first a really interesting (and unfortunately totally pedantic - but what do you expect from a German philosopher) exploration of money and value systems, leading to an exposition on production, the means of production, and how that all relates to labor (especially in relation to money and value above) and, above all, rent.
What I vaguely remember (I wasn't reading for political content at the time) was that the "revolution" espoused by Marx was not a revolution as the term is described today; the revolution was the end result of capitalism, in the sense that "if capitalism goes down that path this is what will happen."
The basic question was "if scarcity doesn't exist then how does capitalism survive?" The answer was "it doesn't," at least not in its current form.
What I don't remember is how capitalism is defined in Marx's worldview, as at that time capitalism was not quite as it is today. While his ideas -were- enough to spark the 1848 revolutions, it's unclear to me what the deal was anymore; I suspect, but am not sure, that it was mainly the beginning of industrial dislocation brought on by the attack on the guild system and urbanization.
Let's face it - back then, large-scale capitalism wasn't the norm by far...and the bourgeoisie weren't the ones revolting, it was the poor. The poor can only revolt if there are enough of them around (in an agrarian society, the poor are too busy farming). But the poor, by definition, are inert until sparked. So who sparked them off? The people who could understand Marxism, of course! Those elite who either rejected their elite status or had some kind of grudge against the current order or wanted to grab the reins of power without going the Long Way.
Heck, in 1848 there weren't a lot of people who could spell, much less spell capitalism.
So where does this leave the above quote? In the land of "not quite accurate fantasyland." You can have your.10 back now.
Using GPS for ATC purposes is a spectacularly bad idea. While the public may think that GPS signals are a natural resource, like cosmic rays, the fact is that GPS is first and foremost a military application and is still under control of the us armed forces. The armed forces occasionally do interesting things to the GPS signals, for field exercises and various tests, and these things would cause Really Bad Things to happen to ATC systems.
For example, one test I heard about was a 5-mile displacement test, to see how well battlefield systems cope. The affect on ATC systems would be obvious.
hah! good qa/test is hard enough to get when you pay for it. for free? guess again. qa always sucks, especially in commercial firms; most qa people are total morons who couldn't get a real job.
the best qa people are sadistic technology freaks who JUST KNOW that the developer cheaped out on that corner case, AND who actually care that they ship a good product.
developers suck at QA. developers have other things to worry about besides corner cases, unless that developer had lots of free time or happens to be a qa zealot.
open source QA? Peer Review? More like mutual masturbation. yEr c0d3z great d00d!
the gist of the article is "we got screwed by our vendors promises' that didn't pan out." Hahahahaha! Welcome to corporate computing!
What it sounds like is sour grapes, and a power-user betrayed syndrome. MacOSX? Who cares? If you make $XXX/hr w/photoshop on a dual g4, no way you're gonna move to OSX unless there's a compelling reason. Duh. Likewise all those users 'left behind': who cares? Home users don't buy software, they steal it. And they don't do much anyway. A little AOL, a little napster, a little AIM, maybe a stolen copy of word/excel, maybe print shop deluxe. Only hardware and AOL makes money of that segment.
And it's not like a box is useless because it can't run the latest and greatest. It's ironic that the dude doesn't understand that; after all, he makes emulators for defunct OS's. Photoshop 2.0 can edit just as well as photoshop 5.0, unless you want the hairy stuff. Heck, I personally run Word for Windows 2.0, because Word 2000/1997/whatever suck.
Don't listen to the marketing hype! don't listen to the beautiful songs of the marketeers! Listen to what you need to do.
oh, checked again. Look for stuff by Arthur W. Combs et al. The book I mentioned was "Perceptual Psychology: a humanistic approach to the study of persons." Looks like Combs has been doing lots of stuff (that one was published in the 50's?). Just ordered a bunch of stuff myself, just to see how the field has been doing.
this is an old, old branch of psychology that has the following premise, if I remember correctly: all human behavior at a moment in time is determined by the person's perception of the situation at that given time. By understanding or manipulating that perception, you can predict and/or control the behavior.
Basically, the idea is that we carry models of our world around with us, and we act in accordance of that given model at all times. We also try and perserve those models, which leads to various things like defensive behavior, acting out, etc.
I'm not sure if this branch of psychology is active, but a search on amazon for 'perceptual psychology' turned up 100 books.
This kind of stuff is amazingly useful to interface designers, because it gives a good framework for understanding the user(s).
unix UIs will never be ported to native OSX, because OSX will have that crappy X interface ported to it. Yuck.
Most X (or web) interfaces suck, because for a really good UI, you need to track massive amounts of state information -and- be able to present information in realtime, something that the web and X are pretty bad at doing (X is better at that than the web, but - realtime UI on X = bandwidth).
Good UI needs thought, and not just implementation thought. It needs the kind of thought that says "who is my user, what are they trying to do, and how do I make it easier for them to do that." Copying a UI isn't good enough; you should be able to explain why something is the way it is, and how it helps the user.
Plus, if you copy a UI blindly, you don't learn why/why not. A UI decision today may be the wrong one next year, but if you don't remember (or never knew) why one thing or another happened, you're screwed.
Take frameworks, for example. The 1.0 AWT was terrible, just terrible. For those of us used to using really good stuff (powerplant, macapp, etc) the AWT was a Great Leap Backward. The API was awful, there was too much subclassing, and you had to understand the internals of the AWT to actually do anything. It's obvious that the AWT folks at that time didn't have any idea who their market was, and what the state-of-the-art was.
Well, I guess sun got some graphical API users, because the AWT 1.1 was a lot, lot better: it actually made sense! (Actually, they got the IFC guys, I think, which is still better than the AWT/swing IMO).
What this shows is that just throwing an interface out there isn't enough...look around, and steal what's good, and think about your users.
Because it's hard to come up with examples that make sense and exercise the language. The strange languages also run up against conceptual problems, since most programmers don't understand the abstractions really well. Note that most of the 'successful' languages are literal, and most programmers are clueless when it comes to OO. A language abstracted even farther from the machine would be, well, too complicated.
For example, why would you use an OODL? I'll bet less than 20 people on/. would be able to explain the benefits of dynamic language features.
The question, though, is moot. PHP is a functional language, and is used extensively, as is FileMaker scripting, SQL, etc. It's a complexity thing.
john katz tries and make us believe that some yoyo demolishing a McDonalds in France is standing up for not only his liberty, but yours. Oh, please.
In JK's mind, acts of violence like the pathetic attack on someone's franchise are symbols of french anti-corporatism and a blow for indvidual liberty against the capitalist juggernaut. Indeed, Jose may soon be leading frustrated frenchmen on a campaign to dismantle their ultra-statist economy, freeing the little guy from the chains of corporate slavery.
Yeah, right.
What Jose has captivated is the imagination of millions of frustrated liberals who are looking for a new ideology to replace the bankrupted leftist canon of communism and socialism. The left, in effect, is looking for a fix for its withdrawl from righteousness anger at the system, and searching desperately for an ideal in which to deify, so as to make their miserable lives meaningful in some way. They get a buzz from the righteousness and militancy of protests and arrests, because they are being victimized and branded by the state who's recognition they so desperately need to self-justify their existence.
Their causes are not causes, they are events in which they are able to finally express their supposed superiority to the masses, a pathetic exclamation of "I AM" to the empty void of the viewing public.
The US revolution was not a media event. It was not a revolution that tore down a new structure in order to put up a flawed ideology. Except for Samuel Adams and the rest of the Mass. contingent, it was a revolution by sober men who were quite aware that they were most likely going to lose their substantial wealth, standing, and lives. And, of course, they built to last, unlike those French losers who tended to go off and kill everyone in the flush of victory (thanks anyway, LaFayette! You still owe us one!)
Unfortunately for katz and his krowd, the corporate/capailist state is much more amenable to the vast majority of the population of the Western World. Except for disaffected youth (and aren't they always disaffected these days?), existence is relatively pastoral. And most of us have no problem at all with that, none at all. Values are expressed at the personal level, not the political level. And that is just fine.
And for those who are unable to define themselves except in respect to others, well, why not watch Fight Club and emulate the protaganist(s)? That guy was going to get REAL results.
Because mac users actually expect quality software that just works, with no configuration hassles. That aesthetic informs all [good] MacOS developers.
Open-source, when you get away from the hype, is basically a cabal of *nix developers, who tend to be willing to jump through all kinds of hoops to get things done. MacOS users, though, earn $$ from their Macs, and lose $$ when they have to mess with config crap. When you're billing $400/hr, the last thing you want to do is tweak that annoying config file with the option that you read about 2 years ago and isn't in the man page.
Sofware that's too hard to use = lost time, and time in the Mac world is more valuable than it is in other worlds. If your product doesn't save users time, they won't use it.
"Welcome to the Real World."
And, just to whip it out, I've configured virtual domains in apache, used perl and shell on 4 *nices enough to hate it, and wondered why the dumb gnu configure script doesn't stash its info in a repository so it doens't have to look for it over and over and over every time you compile another tarball of stuff. I've forgotten more *nix system administration that I ever wanted to know, and I believe that compared to a real *nix like AIX or Solaris, Linux is 2-4 years off-the-ball. And yes, I do use smit(ty).
actually, AIX is probably the worst thing for telecoms. Weird things happen to AIX when you have large numbers of processes (though that may not be a requirement for telco stuff). An embedded would be much, much better.
There are fewer mac viruses than PC viruses, and the PC has a single-file philosophy. Oy!
Indeed, this sounds like a religious argument. Make it easy...for me! the developer!
However, deployment considerations should be thought about. While you, joe user, may in the comfort of your own home be willing to push random files in random directories around, there are people who do actually deploy software on an enterprise scale.
Indeed, software versioning is the Final Frontier in *nix system management. RPM, while nice try, is a code abortion compared to AIX packages.
Your metadirectory idea, btw, is exactly how packages are implemented in MacOSX. This, however, requires attribute data and some funky hand-holding to preserve the semantics of the current *nix OSs. In the NeXT world packages existed only in the OpenStep desktop; go to a terminal, and you could see the raw structure.
It may be that gnome/enlightenment/eazel implements this kind of standard view in the future.
well actually, it is confusing, but only if you are looking at both levels at once as the author does.
There is the Carbon layer, and the BSD layer. The Carbon and BSD layer semantics are different, but the semantics are internally consistent within each layer. If you write Carbon, you think in Carbon. If you write BSD, you think in BSD. It's nowhere near as bad as the long pointer/short pointer stuff in Win16. It's not as bad as MFC vs Win32API.
sigh. Why must I reply to these slashdot puppies? Guess I have nothing better to do than whomp on the ignorant.
"poor" depends on your evaluation criteria.
Unix is also poor because the underlying security philosophies are multi-faceted and self-conflicting. Applications take advantage of this, and can breach the apparent security.
The MacOS is poor because the philosophy is "we're going to decide not to implement security, because we'd rather be doing UI."
What is the priority differential here?
Every OS is a combination of tradeoffs given the legacy of the initial design goals, targets, and hardware. *nix is case-sensitive because it's cheaper; the design decision was to sacrifice user semantics for performance. The MacOS is not robust because it was (is) a single-user system designed for a box with 128k of RAM and a bitmapped, wysiwyg display, so shortcuts were taken to reduce the OS footprint and complexity that, with hindsight, were skanky hacks. *nix commands are terse because the designers didn't like typing, and the pathname semantics may very well be due to the fact that the '/' key is easy to find on the keyboard.
So enlighten yourselves! Most, if not all computing was designed, and there is usually a reason behind everything you see. That reason has strange effects down the timeline, and the reason may not be what you expect.
Slashdot Mirror
the last time I ran one one of the BSDs (netbsd 1.5) I nuked the TCP stack by doing massive amounts of NFS transfers over 10/100. That kind of stuff would never happen under Solaris.
BTW, that massive NFS transfer only took around 4 minutes before the stack died a terrible death, to the point where the netbsd box couldn't ping out due to some kind of buffer error. Oh well.
what is EEC PC133? A search on outpost gives 2 kinds of EEC ram:
PC133 EEC REG CL3
PC133 EEC CL3
So what's the difference between a registered DIMM and an unregistered DIMM?
I can't tell if the references in this article are intentionally wrong so as to mock the ideas, or accidentally wrong. Either way, it's hilarious!
Francis Fukuyama is a historian, not an economist. His book, The End of History, sees "History" as the struggle between competing systems of economic organization. History is at an end (according to him) because after the fall of the Soviet Union, the alternate ideology of economic organization essentially debased itself.
Money was never based on physical assets, it was a proxy for stores of value that were difficult to carry around.
"Now it is just valued for its intrinsic value to people." This make no sense, nor does the argument in the paragraph. Wow!
Fiduciary doesn't mean what it's used as in the reply. Fiduciary: A person holding a position of confidence, such as a trustee, guardian or executor. Capital has become a person!
new economy: what new economy?
This reminded me of the old living color skit, where the guy would be spouting words that he didn't understand. All in all, a great post! The poster is a genius, subtly mocking not only the lack of understanding and reasoning ability of slashdot readers, but mocking in a way that appears to understand the underlying arguments of economic theory. Bravo!
General Magic and Active Paper shipped a handheld-based web browser that was wireless-capable in 1996. They also had POP3 and PPP access, if I remember right. Before them, there was a web browser working on a HP-LX 100 or some weird handheld like that.
whoo! Go DirectTV! That's really classic! Don't really understand why the DirectTV thing is so hard...can't they update the card on the fly, and use the last couple of bits from the last 5 seconds?
Those who have not worked or experienced the enterprise environment will not understand the question. In the enterprise environment, bugs are political issues, not technical issues. If you can say "Well, Microsoft said..." the CIO can go and hammer on Microsoft to get it fixed. This makes everyone happy: the tech doesn't have to deal with the problem, the CIO gets to abuse a big-ass vendor, and MS keeps the large revenue stream from the ELA. Ta-da!
The linux world believes it understands do-it-yourself support. However, when you're paying for support already, why the heck should your guys do the work? Politically, it's easier to blame a vendor than your own guys.
One other point, forgotten in the discussion, is linux provides no way to actually UPDATE 90,000 POS systems at once. Tivoli Enterprise runs on Linux, but I'm not sure what products are supported.
If you actually read the Florida Supreme Court decision, you'd retch in disgust. Their logic went something like this:
FL statue: the FL Secretary of State CAN certify results at her discretion, and there is a penalty associated with late filing of results that the FL Secretary of State accepts.
FL Supreme Court: because there is a penalty associated with filing late results, the FL secretary of state MUST accept late results; if late results were not acceptable, they would not be penalties associated with them, thus there is no discretion on the part of the FL Secretary of State when it comes to accepting vote counts.
US Supreme Court: huh?
What's lucky in this instance is the US Supremes were able to intervene because it's a Federal question. Most state cases that have logic like this probably aren't reviewed by anyone.
buy a new 10/100 switch, a wireless networking solution, or reinstall Windows 2000; any of them are a better use of your time than seeing this movie. This movie could have been squished into 10 minutes if the editor was more skilled and/or the director wasn't so full of himself. Heck, this guy makes Wim Wenders look like the poster boy for tight, cohesive movie-making.
I'd rather watch Until the End of the World 10 times than see 15 minutes of this film again...or get really, really high and watch Heat at half-speed, so as to really enjoy the lack of pacing.
Plus, the story is moronic. Maybe it's really intense for comic book fans who, say, stop at the sunday funnies, but for Sandman fans the plot is pretty lame. I mean geez...an antagonist who has to find/create his own protagonist. C'mon!
This isn't quite "Capital." I'm not sure which volume you read, but what I remember is first a really interesting (and unfortunately totally pedantic - but what do you expect from a German philosopher) exploration of money and value systems, leading to an exposition on production, the means of production, and how that all relates to labor (especially in relation to money and value above) and, above all, rent.
.10 back now.
What I vaguely remember (I wasn't reading for political content at the time) was that the "revolution" espoused by Marx was not a revolution as the term is described today; the revolution was the end result of capitalism, in the sense that "if capitalism goes down that path this is what will happen."
The basic question was "if scarcity doesn't exist then how does capitalism survive?" The answer was "it doesn't," at least not in its current form.
What I don't remember is how capitalism is defined in Marx's worldview, as at that time capitalism was not quite as it is today. While his ideas -were- enough to spark the 1848 revolutions, it's unclear to me what the deal was anymore; I suspect, but am not sure, that it was mainly the beginning of industrial dislocation brought on by the attack on the guild system and urbanization.
Let's face it - back then, large-scale capitalism wasn't the norm by far...and the bourgeoisie weren't the ones revolting, it was the poor. The poor can only revolt if there are enough of them around (in an agrarian society, the poor are too busy farming). But the poor, by definition, are inert until sparked. So who sparked them off? The people who could understand Marxism, of course! Those elite who either rejected their elite status or had some kind of grudge against the current order or wanted to grab the reins of power without going the Long Way.
Heck, in 1848 there weren't a lot of people who could spell, much less spell capitalism.
So where does this leave the above quote? In the land of "not quite accurate fantasyland." You can have your
what's the big deal? You can get this stuff for free here:
http://mrcla.com/XonX/
Duh. Why pay when you can get the crappy x-window interface for free?
Using GPS for ATC purposes is a spectacularly bad idea. While the public may think that GPS signals are a natural resource, like cosmic rays, the fact is that GPS is first and foremost a military application and is still under control of the us armed forces. The armed forces occasionally do interesting things to the GPS signals, for field exercises and various tests, and these things would cause Really Bad Things to happen to ATC systems.
For example, one test I heard about was a 5-mile displacement test, to see how well battlefield systems cope. The affect on ATC systems would be obvious.
hah! good qa/test is hard enough to get when you pay for it. for free? guess again. qa always sucks, especially in commercial firms; most qa people are total morons who couldn't get a real job.
the best qa people are sadistic technology freaks who JUST KNOW that the developer cheaped out on that corner case, AND who actually care that they ship a good product.
developers suck at QA. developers have other things to worry about besides corner cases, unless that developer had lots of free time or happens to be a qa zealot.
open source QA? Peer Review? More like mutual masturbation. yEr c0d3z great d00d!
the gist of the article is "we got screwed by our vendors promises' that didn't pan out." Hahahahaha! Welcome to corporate computing!
What it sounds like is sour grapes, and a power-user betrayed syndrome. MacOSX? Who cares? If you make $XXX/hr w/photoshop on a dual g4, no way you're gonna move to OSX unless there's a compelling reason. Duh. Likewise all those users 'left behind': who cares? Home users don't buy software, they steal it. And they don't do much anyway. A little AOL, a little napster, a little AIM, maybe a stolen copy of word/excel, maybe print shop deluxe. Only hardware and AOL makes money of that segment.
And it's not like a box is useless because it can't run the latest and greatest. It's ironic that the dude doesn't understand that; after all, he makes emulators for defunct OS's. Photoshop 2.0 can edit just as well as photoshop 5.0, unless you want the hairy stuff. Heck, I personally run Word for Windows 2.0, because Word 2000/1997/whatever suck.
Don't listen to the marketing hype! don't listen to the beautiful songs of the marketeers! Listen to what you need to do.
that domain-specific expertise isn't necessarily fungible to other domains.
oh, checked again. Look for stuff by Arthur W. Combs et al. The book I mentioned was "Perceptual Psychology: a humanistic approach to the study of persons." Looks like Combs has been doing lots of stuff (that one was published in the 50's?). Just ordered a bunch of stuff myself, just to see how the field has been doing.
this is an old, old branch of psychology that has the following premise, if I remember correctly: all human behavior at a moment in time is determined by the person's perception of the situation at that given time. By understanding or manipulating that perception, you can predict and/or control the behavior.
Basically, the idea is that we carry models of our world around with us, and we act in accordance of that given model at all times. We also try and perserve those models, which leads to various things like defensive behavior, acting out, etc.
I'm not sure if this branch of psychology is active, but a search on amazon for 'perceptual psychology' turned up 100 books.
This kind of stuff is amazingly useful to interface designers, because it gives a good framework for understanding the user(s).
unix UIs will never be ported to native OSX, because OSX will have that crappy X interface ported to it. Yuck.
Most X (or web) interfaces suck, because for a really good UI, you need to track massive amounts of state information -and- be able to present information in realtime, something that the web and X are pretty bad at doing (X is better at that than the web, but - realtime UI on X = bandwidth).
Good UI needs thought, and not just implementation thought. It needs the kind of thought that says "who is my user, what are they trying to do, and how do I make it easier for them to do that." Copying a UI isn't good enough; you should be able to explain why something is the way it is, and how it helps the user.
Plus, if you copy a UI blindly, you don't learn why/why not. A UI decision today may be the wrong one next year, but if you don't remember (or never knew) why one thing or another happened, you're screwed.
Take frameworks, for example. The 1.0 AWT was terrible, just terrible. For those of us used to using really good stuff (powerplant, macapp, etc) the AWT was a Great Leap Backward. The API was awful, there was too much subclassing, and you had to understand the internals of the AWT to actually do anything. It's obvious that the AWT folks at that time didn't have any idea who their market was, and what the state-of-the-art was.
Well, I guess sun got some graphical API users, because the AWT 1.1 was a lot, lot better: it actually made sense! (Actually, they got the IFC guys, I think, which is still better than the AWT/swing IMO).
What this shows is that just throwing an interface out there isn't enough...look around, and steal what's good, and think about your users.
Because it's hard to come up with examples that make sense and exercise the language. The strange languages also run up against conceptual problems, since most programmers don't understand the abstractions really well. Note that most of the 'successful' languages are literal, and most programmers are clueless when it comes to OO. A language abstracted even farther from the machine would be, well, too complicated.
/. would be able to explain the benefits of dynamic language features.
For example, why would you use an OODL? I'll bet less than 20 people on
The question, though, is moot. PHP is a functional language, and is used extensively, as is FileMaker scripting, SQL, etc. It's a complexity thing.
john katz tries and make us believe that some yoyo demolishing a McDonalds in France is standing up for not only his liberty, but yours. Oh, please.
In JK's mind, acts of violence like the pathetic attack on someone's franchise are symbols of french anti-corporatism and a blow for indvidual liberty against the capitalist juggernaut. Indeed, Jose may soon be leading frustrated frenchmen on a campaign to dismantle their ultra-statist economy, freeing the little guy from the chains of corporate slavery.
Yeah, right.
What Jose has captivated is the imagination of millions of frustrated liberals who are looking for a new ideology to replace the bankrupted leftist canon of communism and socialism. The left, in effect, is looking for a fix for its withdrawl from righteousness anger at the system, and searching desperately for an ideal in which to deify, so as to make their miserable lives meaningful in some way. They get a buzz from the righteousness and militancy of protests and arrests, because they are being victimized and branded by the state who's recognition they so desperately need to self-justify their existence.
Their causes are not causes, they are events in which they are able to finally express their supposed superiority to the masses, a pathetic exclamation of "I AM" to the empty void of the viewing public.
The US revolution was not a media event. It was not a revolution that tore down a new structure in order to put up a flawed ideology. Except for Samuel Adams and the rest of the Mass. contingent, it was a revolution by sober men who were quite aware that they were most likely going to lose their substantial wealth, standing, and lives. And, of course, they built to last, unlike those French losers who tended to go off and kill everyone in the flush of victory (thanks anyway, LaFayette! You still owe us one!)
Unfortunately for katz and his krowd, the corporate/capailist state is much more amenable to the vast majority of the population of the Western World. Except for disaffected youth (and aren't they always disaffected these days?), existence is relatively pastoral. And most of us have no problem at all with that, none at all. Values are expressed at the personal level, not the political level. And that is just fine.
And for those who are unable to define themselves except in respect to others, well, why not watch Fight Club and emulate the protaganist(s)? That guy was going to get REAL results.
Because mac users actually expect quality software that just works, with no configuration hassles. That aesthetic informs all [good] MacOS developers.
Open-source, when you get away from the hype, is basically a cabal of *nix developers, who tend to be willing to jump through all kinds of hoops to get things done. MacOS users, though, earn $$ from their Macs, and lose $$ when they have to mess with config crap. When you're billing $400/hr, the last thing you want to do is tweak that annoying config file with the option that you read about 2 years ago and isn't in the man page.
Sofware that's too hard to use = lost time, and time in the Mac world is more valuable than it is in other worlds. If your product doesn't save users time, they won't use it.
"Welcome to the Real World."
And, just to whip it out, I've configured virtual domains in apache, used perl and shell on 4 *nices enough to hate it, and wondered why the dumb gnu configure script doesn't stash its info in a repository so it doens't have to look for it over and over and over every time you compile another tarball of stuff. I've forgotten more *nix system administration that I ever wanted to know, and I believe that compared to a real *nix like AIX or Solaris, Linux is 2-4 years off-the-ball. And yes, I do use smit(ty).
actually, AIX is probably the worst thing for telecoms. Weird things happen to AIX when you have large numbers of processes (though that may not be a requirement for telco stuff). An embedded would be much, much better.
d00d! u r c00l!
Well, well, well.
There are fewer mac viruses than PC viruses, and the PC has a single-file philosophy. Oy!
Indeed, this sounds like a religious argument. Make it easy...for me! the developer!
However, deployment considerations should be thought about. While you, joe user, may in the comfort of your own home be willing to push random files in random directories around, there are people who do actually deploy software on an enterprise scale.
Indeed, software versioning is the Final Frontier in *nix system management. RPM, while nice try, is a code abortion compared to AIX packages.
Your metadirectory idea, btw, is exactly how packages are implemented in MacOSX. This, however, requires attribute data and some funky hand-holding to preserve the semantics of the current *nix OSs. In the NeXT world packages existed only in the OpenStep desktop; go to a terminal, and you could see the raw structure.
It may be that gnome/enlightenment/eazel implements this kind of standard view in the future.
well actually, it is confusing, but only if you are looking at both levels at once as the author does.
There is the Carbon layer, and the BSD layer. The Carbon and BSD layer semantics are different, but the semantics are internally consistent within each layer. If you write Carbon, you think in Carbon. If you write BSD, you think in BSD. It's nowhere near as bad as the long pointer/short pointer stuff in Win16. It's not as bad as MFC vs Win32API.
It is a bit different, though.
sigh. Why must I reply to these slashdot puppies? Guess I have nothing better to do than whomp on the ignorant.
"poor" depends on your evaluation criteria.
Unix is also poor because the underlying security philosophies are multi-faceted and self-conflicting. Applications take advantage of this, and can breach the apparent security.
The MacOS is poor because the philosophy is "we're going to decide not to implement security, because we'd rather be doing UI."
What is the priority differential here?
Every OS is a combination of tradeoffs given the legacy of the initial design goals, targets, and hardware. *nix is case-sensitive because it's cheaper; the design decision was to sacrifice user semantics for performance. The MacOS is not robust because it was (is) a single-user system designed for a box with 128k of RAM and a bitmapped, wysiwyg display, so shortcuts were taken to reduce the OS footprint and complexity that, with hindsight, were skanky hacks. *nix commands are terse because the designers didn't like typing, and the pathname semantics may very well be due to the fact that the '/' key is easy to find on the keyboard.
So enlighten yourselves! Most, if not all computing was designed, and there is usually a reason behind everything you see. That reason has strange effects down the timeline, and the reason may not be what you expect.