well, excuse *me* for agreeing with "yet another dumb game that requires a 3d card". I'm not whining, in fact I think it's cool that they released this for Linux, and I hope they'll keep doing it with the rest of their games. But *personally*, I'm not in the very least interested. count me out of the "sophisticated game crowd", I just don't care about 1st person shooters nor about games having to be visual show-offs. I'm not retro, I'll take a nice looking thing over a crappy looking one if it's at least as good and interesting, but hell, tetris is more interesting than all of these bloated 3d games together.
Kaa, you're completely missing the point. It's not that some women are upset because they can't be "as beautiful" as top models, the problem is that the whole model industry, and more generally the media-based fads, are pushing the common man's (and woman's) definition of "beautiful", in a way that is actually *harmful* for many of those who are influenced by it and will try to push their own bodies towards it. Like someone else said, it's hard not to find these top models nice to *look at*, but they're usually so skinny and unhealthy looking that they're not anywhere close to being the most attractive, at least to me.
you got it exactly. it was an action thriller, possibly even a good one (i can't tell, action/kungfu thrillers are generally crap to me), NOT a worthwhile sf movie, nor even an attempt at one.
erf! this was one of these films where I kept watching the time "is this ever gonna end or what?". waht a total waste of 2+ hours... I can't find a single redeeming value in it. perty graphics + action does NOT a good film make, not even a good sf film. the whole thing made absolutely no sense, and didn't even try to be original in any way. hell, I usually *like* sf, and i've read tons of it, including quite a few good takes on "reality is actually virtual" (latest and best: Guy Thuillier's _Le Dixieme Cercle_, which also was mixed with tons of religious imagery, but actually interesting that time). and now they're making sequels of it? *shrug* wahtever floats their boat, but to me this is the windows95 of sf movies.
OTOH, my experience with Linux servers is that they stay up without issue, even if they're mostly just left alone. no magical admin incantations needed to keep them up. the admins get to configure the servers, etc, but a slight mistake won't make the box die with a blue screen every few days.
speaking of... does anyoen know of the legal situation for crypto in other EU countries? we only ever hear about France, UK and now Germany.... any clues about Spain? Italy? Belgium? the Scandinavian countries?
Re:No, France's laws are still as bad as the US's
on
Germany Frees Crypto
·
· Score: 1
no, France's policy is still *worse* than the US's! you need to get official forms and declarations in order to be able to sell most kinds of strong crypto-enabled products in France.
duh. if you're going to compare ISAPI to something, be fair and compare it to Apache's module API. CGI is dead old and slow, if you have CGI scripts, at least convert them to FastCGI or something like that. From the tests I've seen, Apache and IIS are not far away at all in terms of speed. Apache is faster at some things, IIS at others, but the difference isn't great anywhere. IIS doesn't outperform Apache "in every imaginable way", and for fsck's sake, even Steve Ballmer admits that in the article that started this thread.
my take is that raw speed is a bit of a red herring. 5% more or less in speed won't let you use the same box for much longer, if your site grows; in other words, you don't want to be within 5% or 10% of your machine's max throughput. so if it's too much for a single server, yes, install several ones, whether it's Apache or IIS, Linux or FreeBSD or Solaris or Tru64 or NT. Just give them each one IP, and several A records for your www.whatever.com hostname, and make them all access the same DB server if there's a DB. btw, web servers typically don't need SMP, they do need lots of RAM (and I do mean lots) and a fast processor, but SMP isn't usually the best way to throw money at a webserver. by the time a PII-450 with 256Mb of RAM is getting hammered, I'd say first optimize your stuff a bit (like, serve your images from a separate Apache process with minimal options and no mod_{perl,php3}, then start thinking about getting more servers. putting the images on a separate box, for example ; that doens't even need any load balancing.
also: stability is much more important than speed, if only because it's much harder to see problems coming. if you absolutely want the very best speed, then there are faster alternatives than Apache; start by looking at Roxen. but I'd keep Apache anyway.
nothing wrong with dynamic pages -- but you have to put a bit of a limit to them. I'd say a good rule of thumb is, make them dynamic when they truly need to change for each user, and pre-compute everyhthing that doesn't. I think it's quite stupid to do a (data, template) merge at every hit if there's just one or two possible templates for that page, and you could have done it instead at the time the data was entered or the page was changed. then again, a lot of web programmers seem to disagree.
Yeah, and don't get me started on SIGTERM and SIGKILL! now *that* is really amazing! whoops I wasn't supposed to tell... damn, I think the cops are outside
the main reason the HP48 was so slow is that most of the high-level internal code was written in system-RPN, which means many levels of indirection before it would reach assembly. they claim to have rewritten a lot of it in assembly for the hp49, which I think is a bit of a bad move... I'd much rather they'd optimized their system-RPN! there was one real easy way to optimize it from what the hp48 did: store the length of all composite objects (such as programs and sub-programs) next to their prolog, instead of calculating it on the fly by recursive descent.
I must have crashed my old hp48 thousands of times. then again, I used to program it in assembler and system RPL, and these things weren't particularily well documented at the time...
huh? rebol threatens some status quo? it looks like just another weird-ish smart language that has some good ideas but is too different for difference's sake to actually get mindshare. I could see myself learning it for fun on a lost afteroon, but I really don't think I'll ever need it for anything.
nah, it's not linux 2.3.1 that will seal bsd's grave.. the BSD people are competent enough to pick up the enhancements too. I can see maybe one of the BSDs going away from lack of interest in a few years, because having 3 of htem is a bit too much for the number of people interested.. but otherwise my guess is that BSD development will continue indefinitely, just like that of Linux.
hmm? what standards worth being complied with does vim fail to follow? if you don't have a.vimrc it sets a lot of options to their vi-like settings. if you're missing something specific, I suggest you mail it to the vim developpers.
well, I woudl gladly accept award money from the KKK (not that I'd do anythign to deserve an award from them, but anyway, we're just speculating). and then I'd make a point of using it in a way completly contrary to what they stand for, or more likely donating it to some institution that specializes in doing that.
no kidding it's very biased! it's full of plain wrong claims all long. whoever wrote this seems to believe that if a GPL program incorporates code from MIT-X-licenced one, then the original suddently disappears from the servers, so only the GPL'd copy remains! and it seems he's never heard of the BSDL's obnoxious advertising clause.
I wish people would take licensing issues a bit more coldly. the MIT-X and the GPL are just not made for teh same purposes; one is a "do what you want", the other is a "copyleft". there's more than enough room for both, and they're not best for the same things. it's really simple: you write an app that you don't want proprietary forked versions of, you use the GPL. if you don't mind, you use the MIT-X-l. one case where I'd definitely use the MIT-X license would be for the implementation of some protocol that I wanted to push, and that I wanted many programs to be interoprable with. in other cases, like programs made just to "make a point", public domain can be the most appropriate.
Solaris doesn't have a framebuffer console??? $ ls -l/dev/fb lrwxrwxrwx 1 root root 58 Mar 16 00:34/dev/fb ->/devices/iommu@f,e0000000/sbus@f,e0001000/cgsix@2, 0:cgsix0
Slashdot Mirror
well, excuse *me* for agreeing with "yet another dumb game that requires a 3d card". I'm not whining, in fact I think it's cool that they released this for Linux, and I hope they'll keep doing it with the rest of their games. But *personally*, I'm not in the very least interested. count me out of the "sophisticated game crowd", I just don't care about 1st person shooters nor about games having to be visual show-offs. I'm not retro, I'll take a nice looking thing over a crappy looking one if it's at least as good and interesting, but hell, tetris is more interesting than all of these bloated 3d games together.
Kaa, you're completely missing the point. It's not that some women are upset because they can't be "as beautiful" as top models, the problem is that the whole model industry, and more generally the media-based fads, are pushing the common man's (and woman's) definition of "beautiful", in a way that is actually *harmful* for many of those who are influenced by it and will try to push their own bodies towards it. Like someone else said, it's hard not to find these top models nice to *look at*, but they're usually so skinny and unhealthy looking that they're not anywhere close to being the most attractive, at least to me.
you got it exactly. it was an action thriller, possibly even a good one (i can't tell, action/kungfu thrillers are generally crap to me), NOT a worthwhile sf movie, nor even an attempt at one.
erf! this was one of these films where I kept watching the time "is this ever gonna end or what?". waht a total waste of 2+ hours... I can't find a single redeeming value in it. perty graphics + action does NOT a good film make, not even a good sf film. the whole thing made absolutely no sense, and didn't even try to be original in any way. hell, I usually *like* sf, and i've read tons of it, including quite a few good takes on "reality is actually virtual" (latest and best: Guy Thuillier's _Le Dixieme Cercle_, which also was mixed with tons of religious imagery, but actually interesting that time). and now they're making sequels of it? *shrug* wahtever floats their boat, but to me this is the windows95 of sf movies.
OTOH, my experience with Linux servers is that they stay up without issue, even if they're mostly just left alone. no magical admin incantations needed to keep them up. the admins get to configure the servers, etc, but a slight mistake won't make the box die with a blue screen every few days.
speaking of... does anyoen know of the legal situation for crypto in other EU countries? we only ever hear about France, UK and now Germany.... any clues about Spain? Italy? Belgium? the Scandinavian countries?
no, France's policy is still *worse* than the US's! you need to get official forms and declarations in order to be able to sell most kinds of strong crypto-enabled products in France.
my take is that raw speed is a bit of a red herring. 5% more or less in speed won't let you use the same box for much longer, if your site grows; in other words, you don't want to be within 5% or 10% of your machine's max throughput. so if it's too much for a single server, yes, install several ones, whether it's Apache or IIS, Linux or FreeBSD or Solaris or Tru64 or NT. Just give them each one IP, and several A records for your www.whatever.com hostname, and make them all access the same DB server if there's a DB. btw, web servers typically don't need SMP, they do need lots of RAM (and I do mean lots) and a fast processor, but SMP isn't usually the best way to throw money at a webserver. by the time a PII-450 with 256Mb of RAM is getting hammered, I'd say first optimize your stuff a bit (like, serve your images from a separate Apache process with minimal options and no mod_{perl,php3}, then start thinking about getting more servers. putting the images on a separate box, for example ; that doens't even need any load balancing.
also: stability is much more important than speed, if only because it's much harder to see problems coming. if you absolutely want the very best speed, then there are faster alternatives than Apache; start by looking at Roxen. but I'd keep Apache anyway.
nothing wrong with dynamic pages -- but you have to put a bit of a limit to them. I'd say a good rule of thumb is, make them dynamic when they truly need to change for each user, and pre-compute everyhthing that doesn't. I think it's quite stupid to do a (data, template) merge at every hit if there's just one or two possible templates for that page, and you could have done it instead at the time the data was entered or the page was changed. then again, a lot of web programmers seem to disagree.
wtf is a SUV?
whoops I wasn't supposed to tell... damn, I think the cops are outside
the main reason the HP48 was so slow is that most of the high-level internal code was written in system-RPN, which means many levels of indirection before it would reach assembly. they claim to have rewritten a lot of it in assembly for the hp49, which I think is a bit of a bad move... I'd much rather they'd optimized their system-RPN! there was one real easy way to optimize it from what the hp48 did: store the length of all composite objects (such as programs and sub-programs) next to their prolog, instead of calculating it on the fly by recursive descent.
RPM is *way* better than Forth. RPM is strongly typed, dynamic, and object-based (but not object oriented, thank Zoinx). can't beat that!
I must have crashed my old hp48 thousands of times. then again, I used to program it in assembler and system RPL, and these things weren't particularily well documented at the time...
huh? rebol threatens some status quo? it looks like just another weird-ish smart language that has some good ideas but is too different for difference's sake to actually get mindshare. I could see myself learning it for fun on a lost afteroon, but I really don't think I'll ever need it for anything.
nah, it's not linux 2.3.1 that will seal bsd's grave.. the BSD people are competent enough to pick up the enhancements too. I can see maybe one of the BSDs going away from lack of interest in a few years, because having 3 of htem is a bit too much for the number of people interested.. but otherwise my guess is that BSD development will continue indefinitely, just like that of Linux.
hmm? what standards worth being complied with does vim fail to follow? if you don't have a .vimrc it sets a lot of options to their vi-like settings. if you're missing something specific, I suggest you mail it to the vim developpers.
well, I woudl gladly accept award money from the KKK (not that I'd do anythign to deserve an award from them, but anyway, we're just speculating). and then I'd make a point of using it in a way completly contrary to what they stand for, or more likely donating it to some institution that specializes in doing that.
I guess I get to be one of the few "Luddites" who once averaged 12 hours of logged-in-ness a day :)
you got it, it was an obvious troll. thanks for responding anyway :)
update your browser... Lynx can read it perfectly.
I wish people would take licensing issues a bit more coldly. the MIT-X and the GPL are just not made for teh same purposes; one is a "do what you want", the other is a "copyleft". there's more than enough room for both, and they're not best for the same things. it's really simple: you write an app that you don't want proprietary forked versions of, you use the GPL. if you don't mind, you use the MIT-X-l. one case where I'd definitely use the MIT-X license would be for the implementation of some protocol that I wanted to push, and that I wanted many programs to be interoprable with. in other cases, like programs made just to "make a point", public domain can be the most appropriate.
Solaris doesn't have a framebuffer console??? $ ls -l /dev/fb lrwxrwxrwx 1 root root 58 Mar 16 00:34 /dev/fb -> /devices/iommu@f,e0000000/sbus@f,e0001000/cgsix@2, 0:cgsix0
yeah but Ireland is full of *catholics*. even more so than Spain, and that's a lot.