> I'm sure you're not the only poorly-educated > well-out-of-college employee that uses jargon > he doesn't understand, that people with a > genuine education in CS have had to deal with.
Oh, be careful. There are a lot of schools that have IS or usage curricula which pretend they're CS curricula; many CS graduates don't have the faintest clue what to do with Knuth. I know a bunch of degree-holders that don't realize that CS is basically a math degree.
In short, you're assuming a particular type of ignorance. There are many forms of ignorance this cat could be displaying.
> Any language that I can't pick up in a day and > master in a week isn't worth my time...
Which is probably why you say
> At this point they all look the same (just a > little different syntax).
Perhaps you should consider trying a language that takes you more than a week. Or, maybe just putting more than a week into one. I've met a bunch of the Big Names, and none of them could master the hairiness of C++ in a week (this is going to draw a lot of flame; preemptive strike: you're not qualified to attack this until you've read at least the three big Meyers books (ec++, mec++, estl) ).
> Ever do OO in assembly.... it's a bitch....
Not really. Unless you're implementing polymorphism or something. Even so, though, it's not much more difficult than it would be in C; the reason it's a bitch is because you're making nontrivial mechanisms.
Also, if you're wasting your time recreating high level abstractions in low level languages and ignoring the years of optimization by hundreds of programmers in things like GCC, well,... you can probably guess what I think of you.
> If you know C++ well, and use the language > effectively, there is very, very little > housekeeping
Whereas I agree with you in sentiment, because I'm a picky bastard, I would like to point out a semantic difference - it's not that there's less housekeeping, but that housekeeping can be done in most situations by the code doing the work in the first place, and therefore can be automated away when the original code is being written.
Then again, the same can be said for most modern languages, and IMNSHO, for good reason.
> There seems to be a lot of life left in that > corpse,
I would take that to a doctor. It seems to be a terminal lack of humor.
Things like the auctioning of languages to free software societies and the reference to Eliza as authority are generally key indicators. Also, the use of "charnel" on/. is a pretty big hint.
Also, perl and python haven't become Lisp. Lisp programmers are annoying in the same way that Perl and Java programmers are - they see all other languages in terms of their home turf.
OTOH, I'm a C++ programmer, and everyone knows why we're annoying - most of everything being broken the way it is is our fault.
" Microsoft's 2005 version of its Windows operating system, apes features that have been in Apple's OS X operating system since 2001. "... which in turn have been in NeXTstep since the beginning of fucking time.
I think it'd be about the same as if you were to try to sell an audiobook you recorded of someone else's text. Free speech only covers things you created in the first place. It's not free usage of speech.
> Yes, you are correct, when you say he can > choose to ignore the feelings of those who feel > his ancestors robbed them. You suggest that no > one can make him care over this historical > injustice. And I would say this is, by > definition, insensitive.
That's disingenuous. If he were writing a book about what happened around that time in the valley, then and only then would he have the obligation to mention what may or may not have happened.
He is no more obligated to pay homage to the people who lived there previously in acquiescance and thanks to a family member than I would be to apologize to the person my mother killed in a car wreck when thanking her in the preface to one of my books.
It is not germane. It would have been kind, but it is certainly not germane.
> But it comes nowhere near Perl when you are > using mod_perl, I think our own/. is an attest > to that.
You think incorrectly, sahib. Even using mod_perl, PHP scales far better. Nice try, though.
The fact of the matter is that with enough attention, most languages scale acceptably. Slashdot runs on a pretty serious server set. I bet you could get a pretty good intercal server going if you tried hard enough.
> More expressive languages tend to have two > downsides that faviour less expressive > languages in many a case: 1) they tend to be > slower than the same system written in a more > traditional language, and no matter how fast > processors get, there will always be incentive > to squeeze the best performance out of them. 2) > they tend to be less general purpose which > limits their ability to become ubiquitous.
Er. There are some languages which move the speed problems of given tasks into the tasks rather than letting them plague the entire language. Consider C++; if you don't know what I'm on about, read the ARM.
Expressiveness isn't anathema to machine speed, it's just that many languages prefer ease of use and maintainability to software efficiency and machine nearness. Languages like C++, Forth, Ocaml, and so on stand as counter-examples.
> PHP 4 and previous versions taught most of us > that if you run it even a 13-year-old script- > kiddy can 0wn your site.
Unless of course you have another thirteen year old doing security at your site. PHP is arguably just as hard to get into as Tomcat, ASP, or most CGI applications (read: not very difficult at all.)
This has more to do with the software written in the language than the language itself. Yes, I know there have been many PHP security holes; there have been security holes for *everyone*. Progressive squashing is the name of the game, and if you're not QMail, you're part of the problem.
Don't blame industry problems on one language, please. It's obnoxious.
Leave it to a Java guy to make asinine comments about practitioners of another langauge. Leave it to a C++ programmer (me) to compound the error.
> Of course, Perl has had all this for some time.
Yah, so have lots of languages. That makes it entering PHP no less exciting.
> Either way, PHP makes for a good interface > language for web apps - I guess. You can throw > it on top of an application layer to do the > real work.
My, how casually dismissive. PHP scales far better than Perl does. Go talk to the yahoos at Yahoo!. (There are good arguments raging about PHP vs. Java in scalability, and so I won't start that here, as I don't want a bad argument to be my fault. Go read if you want to, but don't just assume that because you can't write scalable PHP it can't be done.)
> Now, I'm not sure if that's a faultof PHP > itself, or that some of the "easier" features > of PHP tend to attract less competent > programmers and/or misconfigurations?
Arguably, it's most likely because PHP is built into Apache by default, so lots of people running a server out of their dorm room use it. PHP is the most likely scripting environment on small machines because, let's be frank, Tomcat is a huge pain in the ass to set up.
That said, like C++, PHP is a language which is very easy to put hidden ineffiencies into. So that may compliment the problem.
> because PHP wouldn't scale to a large project > (taking 4-5 seconds to load and generate a page > on a hefty server, the codebase was only about > a meg and a half of PHP),
Sure it will. You just need to use the language well. PHP's got just as many secret places for inefficiency as C++ does.
That said, since I know you're gonna say "nuh-uh", just look at that Yahoo! just switched over to PHP. It scales. Your code just doesn't. Not the language's fault.
> As well as being higher resolution, it performs > sub-pixel anti-aliasing, where the Gameboy does > not.
Ahem. DESI's Asteroids for AGB uses subpixel AA. http://www.gaming-age.com/cgi-bin/specials/sp ecial.pl?spec=charla&pagenum=1
It would help to understand what sub-pixel AA is, so that when I point out to you that there are even amateur AGB games which perform subpixel AA, it won't soften the blow that it's NOT A HARDWARE ISSUE. (Well, at least given that you have an LCD screen.)
Subpixel AA is shading the color channels in a fashion that allows color-channel (rather than pixel) level curvilinear shading. It's largely popularly known through M$' screen resolution enhancement packages, but I'm aware that it goes back at least as far as the Apple ][ (go Woz!)
There's something else that sometimes gets called subpixel AA, that being AA which is rendered finer than a pixel. That's a misnomer, but let's be clear on something: neither the camera nor the AGB do that. That's something for 3D rendering hardware, neither of which the camera or the game system have.
So, look. There are a bunch of schemes for it for Windows. This alone should point out how not a hardware issue it is, but let's give some links instead. And then, when you're nice and full, realize that subpixel AA is a software-only thing, that a camera pretty much can't do subpixel AA for images (it's a technique that's useful only in situations where the target device has more color channels than the source image), and that mostly you're just reporting what you've been told.
(I was going to insert a snide "check google first" comment, but surprisingly Google doesn't find anything in english which isn't 404ed.)
Here's Jeff Froweihn's excellent beginner's explanation of Subpixel AA and how to do it on the AGB, from April of 2001 (two years ago): http://grc.com/ctwhat.htm
Blowhole features a subpixel rendered font; subpixel AA can be in constant data! http://www.somethingscrewy.com/
Clearize is a subpixel reducer on a gameboy development site that'll allow you to try ti for yourself. Only works on an LCD monitor. http://www.pineight.com/pc/
Here's a subpixel scroller for the GameBoy *Color*, not even the advance. (Again, this'll only work on real hardware, or an emulator which supports per-channel screen zoom.) http://www.emucamp.com/demoroms/list.cgi?s ystem=gb (it's near the bottom, just search for "pixel")
> > Same thing with TiVo. What's the purpose for > > excluding it from their advertisements? > > Probably because the target audience for TiVos > are humans. Seriously, Dude, we're "Geeks," and > just _some_ of us care about a device's OS.
Nonono. He meant, "why won't sharp say that they have sync stuff for linux?" And it's perfectly reasonable, because they mention the other platforms they can sync to. Why not just add it at the end of the list?
For a PDA which has Linux as one of its *selling* points, I think you underestimate the interested portion of their ownership populace.
> Games have gone mainstream so just like you get > easy funding for middle of the road MPAA crap, > EA seems to have a good living putting out > endless sequels to FIFA, formulaic platformers > etc.
I often get the impression that people think that because there are hundreds of video games every year, that the sell-outs have it easy.
It's almost impossible even to sell out. Do you have any idea how many programmers want to be game developers? Half the people I knew in college went into CS because they heard the money was good; now that there's the glut, they're looking for any easy way out they can find. Video games are a common target; the people who wanted in for the love of writing games are simply being drowned out by numbers.
Maybe try entering the market before talking about how easy it is to do so. I suspect that you're in for a dry shock.
> I've been to the last couple of GDCs and seen > independent gaming's "best of the best". I've > also downloaded hundreds of demos from > independent developers. They're not very good.
True as all of this may be, it's also a shallow view of the problem. I've been writing demo carts for years; I've only been able to afford to go to E3 once, and the people who set my interviews up for me fucked up something awful (only one of my eleven interviews was with a publisher who even touched my intended target platform.) I'm not complaining - they also got me in and did a lot of very important explaining-of-things to me - but still, the bulk of the reason that GDC games suck is that the GDC does an abominable job of getting the talent to itself.
They expect people who are spending every red cent they have on developing a game while holding a minimum-wage job to also transport themselves to a con, put themselves up in a hotel, and provide promotional material.
The GDC's mistake is that it should be making money charging publishers to enter the building, and then turning half of that money into locating and sponsoring the tiny developers which would in turn make the GDC a worthwhile place for these publishers to go. I bet (and I could be very wrong about this) that a publisher would pay $10,000 to enter a con where they could fight to take up the next three iDs.
It's too hard to find the wheat amongst the chaff, because the wheat isn't getting the resources to grow, and there's so much f*cking chaff.
> No one with a clue is suggesting that indies > try to get on store shelves. You can't get into > Walmart, don't even try.
Actually, that's exactly where KO Interactive, now GamesParlor, got in. And it's the only successful thing they've ever done. But they've sunk seven figures into multiple failed business investments simply because they had one successful title. (A $10 CD of shareware games they bought, and sold at walmart.)
Frankly, I've often considered writing just such a thing to get my start. I know if I had anywhere near the funding they did, I could make it. (I watched someone dump $50K into an untested advertising program which fell flat there. Then, they couldn't pay my paycheck. Most galling thing I've ever seen.)
Slashdot Mirror
> I'm sure you're not the only poorly-educated
> well-out-of-college employee that uses jargon
> he doesn't understand, that people with a
> genuine education in CS have had to deal with.
Oh, be careful. There are a lot of schools that have IS or usage curricula which pretend they're CS curricula; many CS graduates don't have the faintest clue what to do with Knuth. I know a bunch of degree-holders that don't realize that CS is basically a math degree.
In short, you're assuming a particular type of ignorance. There are many forms of ignorance this cat could be displaying.
> Any language that I can't pick up in a day and
... you can probably guess what I think of you.
> master in a week isn't worth my time...
Which is probably why you say
> At this point they all look the same (just a
> little different syntax).
Perhaps you should consider trying a language that takes you more than a week. Or, maybe just putting more than a week into one. I've met a bunch of the Big Names, and none of them could master the hairiness of C++ in a week (this is going to draw a lot of flame; preemptive strike: you're not qualified to attack this until you've read at least the three big Meyers books (ec++, mec++, estl) ).
> Ever do OO in assembly.... it's a bitch....
Not really. Unless you're implementing polymorphism or something. Even so, though, it's not much more difficult than it would be in C; the reason it's a bitch is because you're making nontrivial mechanisms.
Also, if you're wasting your time recreating high level abstractions in low level languages and ignoring the years of optimization by hundreds of programmers in things like GCC, well,
> but for the math gifted, most of them will code
> in Occaml
That's mighty presumptuous, especially considering the lambda calculus, OCaml's not so great history with parallelization, and Eiffel.
Besides, they're all gonna program in C++. Why? Because nobody has a choice, that's why. Otherwise, we'd all be up to our necks in Ruby and Lua.
> If you know C++ well, and use the language
> effectively, there is very, very little
> housekeeping
Whereas I agree with you in sentiment, because I'm a picky bastard, I would like to point out a semantic difference - it's not that there's less housekeeping, but that housekeeping can be done in most situations by the code doing the work in the first place, and therefore can be automated away when the original code is being written.
Then again, the same can be said for most modern languages, and IMNSHO, for good reason.
> There seems to be a lot of life left in that
/. is a pretty big hint.
> corpse,
I would take that to a doctor. It seems to be a terminal lack of humor.
Things like the auctioning of languages to free software societies and the reference to Eliza as authority are generally key indicators. Also, the use of "charnel" on
Also, perl and python haven't become Lisp. Lisp programmers are annoying in the same way that Perl and Java programmers are - they see all other languages in terms of their home turf.
OTOH, I'm a C++ programmer, and everyone knows why we're annoying - most of everything being broken the way it is is our fault.
"Ha, ha." - Nelson
" ... which in turn have been in NeXTstep since the beginning of fucking time.
Microsoft's 2005 version of its Windows operating system, apes features that have been in Apple's OS X operating system since 2001.
"
I think it'd be about the same as if you were to try to sell an audiobook you recorded of someone else's text. Free speech only covers things you created in the first place. It's not free usage of speech.
> Yes, you are correct, when you say he can
> choose to ignore the feelings of those who feel
> his ancestors robbed them. You suggest that no
> one can make him care over this historical
> injustice. And I would say this is, by
> definition, insensitive.
That's disingenuous. If he were writing a book about what happened around that time in the valley, then and only then would he have the obligation to mention what may or may not have happened.
He is no more obligated to pay homage to the people who lived there previously in acquiescance and thanks to a family member than I would be to apologize to the person my mother killed in a car wreck when thanking her in the preface to one of my books.
It is not germane. It would have been kind, but it is certainly not germane.
> What do you mean Dragon Ball Z didn't
/.ers have more taste. ;)
> record?!?!
That's the point at which they become white-hat, and at which you should be thanking them.
But I assume you meant that as sarcasm in your example.
> No app just mysteriously segfaults
This is only true because Win2k is technically an operating system, not an application.
(ducks!)
Yeah, you were making a valid point. But mine is funnier.
> But it comes nowhere near Perl when you are /. is an attest
> using mod_perl, I think our own
> to that.
You think incorrectly, sahib. Even using mod_perl, PHP scales far better. Nice try, though.
The fact of the matter is that with enough attention, most languages scale acceptably. Slashdot runs on a pretty serious server set. I bet you could get a pretty good intercal server going if you tried hard enough.
> I'll begin doing some php only when microsoft
> will have released Visual MSPHP. Thats it.
I can't decide if that's tongue in cheek or if you're really that much of a troll.
> PHP is not a part of Apache by default.
Sorry, what I should have said was that there was a distro of Apache with PHP built in that's pre-built and ready for most platforms.
And honestly, in my opinion, thta's just as good as. But you're right to raise the error flag.
> More expressive languages tend to have two
> downsides that faviour less expressive
> languages in many a case: 1) they tend to be
> slower than the same system written in a more
> traditional language, and no matter how fast
> processors get, there will always be incentive
> to squeeze the best performance out of them. 2)
> they tend to be less general purpose which
> limits their ability to become ubiquitous.
Er. There are some languages which move the speed problems of given tasks into the tasks rather than letting them plague the entire language. Consider C++; if you don't know what I'm on about, read the ARM.
Expressiveness isn't anathema to machine speed, it's just that many languages prefer ease of use and maintainability to software efficiency and machine nearness. Languages like C++, Forth, Ocaml, and so on stand as counter-examples.
> PHP 4 and previous versions taught most of us
> that if you run it even a 13-year-old script-
> kiddy can 0wn your site.
Unless of course you have another thirteen year old doing security at your site. PHP is arguably just as hard to get into as Tomcat, ASP, or most CGI applications (read: not very difficult at all.)
This has more to do with the software written in the language than the language itself. Yes, I know there have been many PHP security holes; there have been security holes for *everyone*. Progressive squashing is the name of the game, and if you're not QMail, you're part of the problem.
Don't blame industry problems on one language, please. It's obnoxious.
> Leave it to a Perl guy to compare PHP to Java.
Leave it to a Java guy to make asinine comments about practitioners of another langauge. Leave it to a C++ programmer (me) to compound the error.
> Of course, Perl has had all this for some time.
Yah, so have lots of languages. That makes it entering PHP no less exciting.
> Either way, PHP makes for a good interface
> language for web apps - I guess. You can throw
> it on top of an application layer to do the
> real work.
My, how casually dismissive. PHP scales far better than Perl does. Go talk to the yahoos at Yahoo!. (There are good arguments raging about PHP vs. Java in scalability, and so I won't start that here, as I don't want a bad argument to be my fault. Go read if you want to, but don't just assume that because you can't write scalable PHP it can't be done.)
> Now, I'm not sure if that's a faultof PHP
> itself, or that some of the "easier" features
> of PHP tend to attract less competent
> programmers and/or misconfigurations?
Arguably, it's most likely because PHP is built into Apache by default, so lots of people running a server out of their dorm room use it. PHP is the most likely scripting environment on small machines because, let's be frank, Tomcat is a huge pain in the ass to set up.
That said, like C++, PHP is a language which is very easy to put hidden ineffiencies into. So that may compliment the problem.
> because PHP wouldn't scale to a large project
> (taking 4-5 seconds to load and generate a page
> on a hefty server, the codebase was only about
> a meg and a half of PHP),
Sure it will. You just need to use the language well. PHP's got just as many secret places for inefficiency as C++ does.
That said, since I know you're gonna say "nuh-uh", just look at that Yahoo! just switched over to PHP. It scales. Your code just doesn't. Not the language's fault.
> As well as being higher resolution, it performs
p ecial .pl?spec=charla&pagenum=1
s ystem=gb
> sub-pixel anti-aliasing, where the Gameboy does
> not.
Ahem. DESI's Asteroids for AGB uses subpixel AA.
http://www.gaming-age.com/cgi-bin/specials/s
It would help to understand what sub-pixel AA is, so that when I point out to you that there are even amateur AGB games which perform subpixel AA, it won't soften the blow that it's NOT A HARDWARE ISSUE. (Well, at least given that you have an LCD screen.)
Subpixel AA is shading the color channels in a fashion that allows color-channel (rather than pixel) level curvilinear shading. It's largely popularly known through M$' screen resolution enhancement packages, but I'm aware that it goes back at least as far as the Apple ][ (go Woz!)
There's something else that sometimes gets called subpixel AA, that being AA which is rendered finer than a pixel. That's a misnomer, but let's be clear on something: neither the camera nor the AGB do that. That's something for 3D rendering hardware, neither of which the camera or the game system have.
So, look. There are a bunch of schemes for it for Windows. This alone should point out how not a hardware issue it is, but let's give some links instead. And then, when you're nice and full, realize that subpixel AA is a software-only thing, that a camera pretty much can't do subpixel AA for images (it's a technique that's useful only in situations where the target device has more color channels than the source image), and that mostly you're just reporting what you've been told.
(I was going to insert a snide "check google first" comment, but surprisingly Google doesn't find anything in english which isn't 404ed.)
Here's Jeff Froweihn's excellent beginner's explanation of Subpixel AA and how to do it on the AGB, from April of 2001 (two years ago):
http://grc.com/ctwhat.htm
Blowhole features a subpixel rendered font; subpixel AA can be in constant data!
http://www.somethingscrewy.com/
Clearize is a subpixel reducer on a gameboy development site that'll allow you to try ti for yourself. Only works on an LCD monitor.
http://www.pineight.com/pc/
Here's a subpixel scroller for the GameBoy *Color*, not even the advance. (Again, this'll only work on real hardware, or an emulator which supports per-channel screen zoom.)
http://www.emucamp.com/demoroms/list.cgi?
(it's near the bottom, just search for "pixel")
I mean, really, did I read that right? You charge your watch and it's good to take incoming calls for a little over three hours?
> > Same thing with TiVo. What's the purpose for
> > excluding it from their advertisements?
>
> Probably because the target audience for TiVos
> are humans. Seriously, Dude, we're "Geeks," and
> just _some_ of us care about a device's OS.
Nonono. He meant, "why won't sharp say that they have sync stuff for linux?" And it's perfectly reasonable, because they mention the other platforms they can sync to. Why not just add it at the end of the list?
For a PDA which has Linux as one of its *selling* points, I think you underestimate the interested portion of their ownership populace.
> in the US? I remeber seeing a couple of old
> people riding motorized bikes but ive only seen
> them once.
I believe they're mostly made by these guys.
> Games have gone mainstream so just like you get
> easy funding for middle of the road MPAA crap,
> EA seems to have a good living putting out
> endless sequels to FIFA, formulaic platformers
> etc.
I often get the impression that people think that because there are hundreds of video games every year, that the sell-outs have it easy.
It's almost impossible even to sell out. Do you have any idea how many programmers want to be game developers? Half the people I knew in college went into CS because they heard the money was good; now that there's the glut, they're looking for any easy way out they can find. Video games are a common target; the people who wanted in for the love of writing games are simply being drowned out by numbers.
Maybe try entering the market before talking about how easy it is to do so. I suspect that you're in for a dry shock.
> I've been to the last couple of GDCs and seen
> independent gaming's "best of the best". I've
> also downloaded hundreds of demos from
> independent developers. They're not very good.
True as all of this may be, it's also a shallow view of the problem. I've been writing demo carts for years; I've only been able to afford to go to E3 once, and the people who set my interviews up for me fucked up something awful (only one of my eleven interviews was with a publisher who even touched my intended target platform.) I'm not complaining - they also got me in and did a lot of very important explaining-of-things to me - but still, the bulk of the reason that GDC games suck is that the GDC does an abominable job of getting the talent to itself.
They expect people who are spending every red cent they have on developing a game while holding a minimum-wage job to also transport themselves to a con, put themselves up in a hotel, and provide promotional material.
The GDC's mistake is that it should be making money charging publishers to enter the building, and then turning half of that money into locating and sponsoring the tiny developers which would in turn make the GDC a worthwhile place for these publishers to go. I bet (and I could be very wrong about this) that a publisher would pay $10,000 to enter a con where they could fight to take up the next three iDs.
It's too hard to find the wheat amongst the chaff, because the wheat isn't getting the resources to grow, and there's so much f*cking chaff.
> No one with a clue is suggesting that indies
> try to get on store shelves. You can't get into
> Walmart, don't even try.
Actually, that's exactly where KO Interactive, now GamesParlor, got in. And it's the only successful thing they've ever done. But they've sunk seven figures into multiple failed business investments simply because they had one successful title. (A $10 CD of shareware games they bought, and sold at walmart.)
Frankly, I've often considered writing just such a thing to get my start. I know if I had anywhere near the funding they did, I could make it. (I watched someone dump $50K into an untested advertising program which fell flat there. Then, they couldn't pay my paycheck. Most galling thing I've ever seen.)