Well then what's really needed is a way for you to easily tell the system, "I have XFree86 version x.y". Or better yet, make install should do the right thing and add the installed package to the database.
Yeah, the fact that TeX (and therfore LaTeX) is Turing complete makes it a PITA for many document description purposes. PostScript has the same problem too.
Of course, HTML has this problem too, thanks to embedded scripting. How do you write a program which analyses a web page and determines whether it has any pop-ups? You can't, because it's equivalent to solving the halting problem. (note that having the brwser simply ignore requests to open new windows is not what I'm talking about. I'm talking about analysing a page without executing the content. After all, the content might not ever terminate). A restricted subset of HTML that lacks <SCRIPT> and the ONWHATEVER attributes can be analysed for the most part though.
HTML has another advantage over LaTeX: it's a hell of a lot easier to parse. It's even easier to make an XHTML parser.
The point of what I wrote was that him seeing five or six people's "dark secrets" just felt somehow gratuitous.
So if the first person he'd bumped into was the garbage man, wouldn't you be complaining about how unrealistic it was that he "got lucky" the first time he bumped into someone?
The impression I got was that when he first started bumping into people, he was sort of disoriented. It wasn't obvious what he could do, and he was probably somewhat disturbed by his ability. When he finally bumped the garbage man, what he saw was so disturbing, that it jolted him into action. At least, that's the impression I got.
he always see's things after-the-fact when it is too late too stop them.
He only sees what they've done, but it isn't too late to do something. He knew about the guy with the gun ayt the stadium, and he did save the kids from the garbage man. So yes, he can't predict the future, but he can often make a reasonable extrapolation based on the past events that he does see. I mean really, how many comic book heros stop criminals before they commit any crimes?
Package dependencies exist for far more than just libarry requirements. RPM actually does have support for saying that a package requires a certain file to exist. Most RPMs use package dependencies rather than file dependencies though, because they're more reliable and can, in general, express more than file dependencies
For example, suppose a program relies on a specific version of diff. What file should it look for?
So what's your suggestion then? How do you propose that a package installer determine what was "actually installed", without resorting to the use of magic?
I don't thing the Hashtable vs HashMap thing is so bad. Having both synchronized and unsynchronized hash table is classes is useful. Of course, the names could be better (replacing "Map" with "table" doesn't really indicate much of anything)
The API does have numerous warts. The lack of anything like select() (as you mentioned), the completely awful design of the File class, numerous quirks in earlier versions of AWT, the rather poorly defined Image* classes, things in the wrong packages, packages with bad names (ie: java.sql probably should've been called javax.sql or javax.jdbc), the incredibly stupidly named Calendar class, and the list goes on.
Things are getting better though, and the nasty stuff is slowly being deprecated away. Pre-ANSI C was pretty icky too. It took roughly 15 years for ANSI C to be developed after C was created. Java's a fairly young language, so it isn't too surprising that it'll have some growing pains. It's interesting to note that most of these problems are in the API too, not in the language itself.
Actually HashMap can be more efficient that Hashtable, because it doesn't synchronize everything. Of course, that also means you have to manage the synchronization aspects yourself, if multiple threads will be manipulating it.
I think you two are just playing semantics now, because I agree with everything both of you said. You just seem to define "performance problems" differently. Compare SimonK's statement:
I don't consider overuse of linear search, refetching of already fetched data, and so on, to be performance problems. They're sheer carelessness, and they mess up the structure and expressiveness of the code. They reflect a lack of thought in the design.
to what you just said:
You're right that it doesn't make sense to optimize every last bit of the code if it is rarely executed; but if you don't think about the performance impact of architectural decisions up front you're going to have real problems later when you start to worry about performance.
I think what it boils down to is this: getting it working first is more important that optimizing every single inner loop and applying every possible optimization. At the same time though, the high level design needs to be done with a consideration of what can be done efficiently.
or the most part the voice actors hired on many projects suck... just watch most of the old "Streamline titles" or "warriors of the wind" (warrning: if you've seen the original Naussica do NOT watch warroiors of the wind)
The problem with Warriors of the Wind wasn't the voice acting. The problem was that they decided to rewrite the entire story, and cut out huge chunks of it. From what I remember, the voice acting wasn't that bad. Of course, the last time I saw Warriors of the Wind, Knight Rider was still showing on TV (and I don't mean reruns), so my memory could be a bit hazy...
Of course, their are many other examples of anime with incredibly poor dubbing. The worst I'd ever seen/heard was a dub of Macross Plus into Cantonese with English subtitles. Let's just say that forcing someone to listen to this tape is probably against the Geneva Convention. (And the worst part is, Macross Plus is known for it's great soundtrack)
There's a difference between getting offended when someone says something, stopping someone from saying it.
Of course, it isn't the word itself one should be offended by, but the intended meaning. If you use a word like "nigger" or "kike" actually intending to insult the people those derogatory terms refer to, then that would be something I'd get offended about. On the other hand if you used those terms for another reason (like quoting Mark Twain, or this discussion), then I wouldn't be offended at all. Either way, I wouldn't attempt to take away your right to say it.
On CityTV they bleep out "fuck" and "motherfucker" during the day, but after a certain time (I think around 8:30PM or so) they only bleep out the "mother" part of "motherfucker". I'd always thought it strange that the innocent word "mother" somehow made a swear word even worse somehow.
If that's a troll, it's an awfully well-crafted one.
Pinging 127.0.0.1 will only tell you if the lo interface is up, and that isn't particularly useful in most situations.
I find it's best to ping the address of my DNS servers. The addresses are right there in/etc/resolv.conf (assuming you're not usinga a local caching DNS, in which case you want to go for the servers it refers to). But I usually only do that if pinging yahoo.com fails.
Guess what?...the CD I made sounds better - richer sound, deeper bass. My advice would be, if you own the vinyl and it's in good condition then get a friend with good equipment to make a CD for you.
It's no surprise that the CD you made sounds different. The fact that it sounds better to you is completely subjective. The fact that you're used to the one you made probably has something to do with it. Someone who was used to the commercial CD would probably think your CD would sound weird and muffled.
Does any of the money generated by the browser get back to Mozilla? I kinda doubt it.
Most of the development on Mozilla was done by full-time employees of Netscape, working on Netscape's time. So yes, a lot of the money generated by the browser goes back to Mozilla.
There is no unique software.... One does not completely pay 'd' to rewrite Apache because it doesn't have a feature, he pays the much smaller amount demanded by a programmer to make the change.
Yeah, and we all know Apache is really just cat with a few tweaks here and there.
H1B's can now leave their jobs for better paying ones, or jobs with a better work environment.
This will force H1B loving companies into competing fairly for their employees (including H1B's).
While I think the new legislation is great (guess who just switched jobs), it isn't quite enough to make things fair for all H1B holders. Most (or at least a very large fraction of) H1B holders also apply for a greencard. If you switch jobs during the 4-5 years of the greencard process, your entire application is nullified.
That doesn't affect me, personally. I'm fairly certain I'll be staying with my new employer well beyond getting my greencard. There are many other who are essentially stuck with their current employer for at least a few years because of this. That can result in employers giving far less compensation to, or requiring far more work from, employees who are in the greencard application process. This harms not only foreign workers, but also American workers. It harms American workers because it gives some employers an incentive to choose foreign workers rather than Americans.
Not all of us run Windows, so we can't beta-test IE 6 unless MS suddenly decides to start supporting platforms outside of Windows and Macintosh.
Actually, there is a version of IE "for UNIX". It's actually for Solaris (and maybe HPUX too?), but the only page that says this is the download page. Everywhere else they say "for UNIX", to make it sound like they support just about every platform in common usage...
I tried out IE for Solaris. It's missing a lot of the features of the Win32 version. If you install it, you'll also see that Microsoft doesn't quite understand this "UNIX" thing... half of the install goes in your home directory (rather than somewhere in/usr), but with the permissions set so that only root can execute the browser.
Slashdot Mirror
What do you need Perl for? Just use wget and md5sum.
/dev/null -O - http://www.microsft.com/ | md5sum -
[~]$ wget -o
721d953c6e14caeaadcc9e14d8d64691 -
Well then what's really needed is a way for you to easily tell the system, "I have XFree86 version x.y". Or better yet, make install should do the right thing and add the installed package to the database.
I think the fact that existing phones simply wouldn't be able to dial these numbers is a significantly larger problem.
Yeah, the fact that TeX (and therfore LaTeX) is Turing complete makes it a PITA for many document description purposes. PostScript has the same problem too.
Of course, HTML has this problem too, thanks to embedded scripting. How do you write a program which analyses a web page and determines whether it has any pop-ups? You can't, because it's equivalent to solving the halting problem. (note that having the brwser simply ignore requests to open new windows is not what I'm talking about. I'm talking about analysing a page without executing the content. After all, the content might not ever terminate). A restricted subset of HTML that lacks <SCRIPT> and the ONWHATEVER attributes can be analysed for the most part though.
HTML has another advantage over LaTeX: it's a hell of a lot easier to parse. It's even easier to make an XHTML parser.
The point of what I wrote was that him seeing five or six people's "dark secrets" just felt somehow gratuitous.
So if the first person he'd bumped into was the garbage man, wouldn't you be complaining about how unrealistic it was that he "got lucky" the first time he bumped into someone?
The impression I got was that when he first started bumping into people, he was sort of disoriented. It wasn't obvious what he could do, and he was probably somewhat disturbed by his ability. When he finally bumped the garbage man, what he saw was so disturbing, that it jolted him into action. At least, that's the impression I got.
he always see's things after-the-fact when it is too late too stop them.
He only sees what they've done, but it isn't too late to do something. He knew about the guy with the gun ayt the stadium, and he did save the kids from the garbage man. So yes, he can't predict the future, but he can often make a reasonable extrapolation based on the past events that he does see. I mean really, how many comic book heros stop criminals before they commit any crimes?
Package dependencies exist for far more than just libarry requirements. RPM actually does have support for saying that a package requires a certain file to exist. Most RPMs use package dependencies rather than file dependencies though, because they're more reliable and can, in general, express more than file dependencies
For example, suppose a program relies on a specific version of diff. What file should it look for?
So what's your suggestion then? How do you propose that a package installer determine what was "actually installed", without resorting to the use of magic?
I don't thing the Hashtable vs HashMap thing is so bad. Having both synchronized and unsynchronized hash table is classes is useful. Of course, the names could be better (replacing "Map" with "table" doesn't really indicate much of anything)
The API does have numerous warts. The lack of anything like select() (as you mentioned), the completely awful design of the File class, numerous quirks in earlier versions of AWT, the rather poorly defined Image* classes, things in the wrong packages, packages with bad names (ie: java.sql probably should've been called javax.sql or javax.jdbc), the incredibly stupidly named Calendar class, and the list goes on.
Things are getting better though, and the nasty stuff is slowly being deprecated away. Pre-ANSI C was pretty icky too. It took roughly 15 years for ANSI C to be developed after C was created. Java's a fairly young language, so it isn't too surprising that it'll have some growing pains. It's interesting to note that most of these problems are in the API too, not in the language itself.
Actually HashMap can be more efficient that Hashtable, because it doesn't synchronize everything. Of course, that also means you have to manage the synchronization aspects yourself, if multiple threads will be manipulating it.
I think you two are just playing semantics now, because I agree with everything both of you said. You just seem to define "performance problems" differently. Compare SimonK's statement:
I don't consider overuse of linear search, refetching of already fetched data, and so on, to be performance problems. They're sheer carelessness, and they mess up the structure and expressiveness of the code. They reflect a lack of thought in the design.
to what you just said:
You're right that it doesn't make sense to optimize every last bit of the code if it is rarely executed; but if you don't think about the performance impact of architectural decisions up front you're going to have real problems later when you start to worry about performance.
I think what it boils down to is this: getting it working first is more important that optimizing every single inner loop and applying every possible optimization. At the same time though, the high level design needs to be done with a consideration of what can be done efficiently.
This is our good friends the RIAA in action.
No. The RIAA has nothing to do with it. It's the MPAA you're thinking of. The RIAA is audio, MPAA video. It really isn't that hard to remember.
Of course, their are many other examples of anime with incredibly poor dubbing. The worst I'd ever seen/heard was a dub of Macross Plus into Cantonese with English subtitles. Let's just say that forcing someone to listen to this tape is probably against the Geneva Convention. (And the worst part is, Macross Plus is known for it's great soundtrack)
Yeah, then you get up and go to the next room for a sec, and suddenly realize you can't understand what they're saying unless you look at the screen!
I am ashamed to admit how many times this has happened to me... (nevertheless, I still prefer subbed over dubbed)
1) it's kind of difficult to write
PDF also make extensive use of LZW which we all know is a Bad Thing thanks to Unisys.
There's a difference between getting offended when someone says something, stopping someone from saying it.
Of course, it isn't the word itself one should be offended by, but the intended meaning. If you use a word like "nigger" or "kike" actually intending to insult the people those derogatory terms refer to, then that would be something I'd get offended about. On the other hand if you used those terms for another reason (like quoting Mark Twain, or this discussion), then I wouldn't be offended at all. Either way, I wouldn't attempt to take away your right to say it.
Take "garglesnarf" for example.
Watch your mouth!
On CityTV they bleep out "fuck" and "motherfucker" during the day, but after a certain time (I think around 8:30PM or so) they only bleep out the "mother" part of "motherfucker". I'd always thought it strange that the innocent word "mother" somehow made a swear word even worse somehow.
ObFuckDomain: fuckedcompany.com
the U.S. is the nation that has become the most powerful nation with the highest standard of living (for its size)
When you say "for its size", I assume you mean population-wise. If that's the case, then the same could be said about China.
If you're actually referring to land-area, Canada has a higher standard of living that the US.
If that's a troll, it's an awfully well-crafted one.
/etc/resolv.conf (assuming you're not usinga a local caching DNS, in which case you want to go for the servers it refers to). But I usually only do that if pinging yahoo.com fails.
Pinging 127.0.0.1 will only tell you if the lo interface is up, and that isn't particularly useful in most situations.
I find it's best to ping the address of my DNS servers. The addresses are right there in
Guess what? ...the CD I made sounds better - richer sound, deeper bass. My advice would be, if you own the vinyl and it's in good condition then get a friend with good equipment to make a CD for you.
It's no surprise that the CD you made sounds different. The fact that it sounds better to you is completely subjective. The fact that you're used to the one you made probably has something to do with it. Someone who was used to the commercial CD would probably think your CD would sound weird and muffled.
Does any of the money generated by the browser get back to Mozilla? I kinda doubt it.
Most of the development on Mozilla was done by full-time employees of Netscape, working on Netscape's time. So yes, a lot of the money generated by the browser goes back to Mozilla.
There is no unique software. ... One does not completely pay 'd' to rewrite Apache because it doesn't have a feature, he pays the much smaller amount demanded by a programmer to make the change.
Yeah, and we all know Apache is really just cat with a few tweaks here and there.
H1B's can now leave their jobs for better paying ones, or jobs with a better work environment.
This will force H1B loving companies into competing fairly for their employees (including H1B's).
While I think the new legislation is great (guess who just switched jobs), it isn't quite enough to make things fair for all H1B holders. Most (or at least a very large fraction of) H1B holders also apply for a greencard. If you switch jobs during the 4-5 years of the greencard process, your entire application is nullified.
That doesn't affect me, personally. I'm fairly certain I'll be staying with my new employer well beyond getting my greencard. There are many other who are essentially stuck with their current employer for at least a few years because of this. That can result in employers giving far less compensation to, or requiring far more work from, employees who are in the greencard application process. This harms not only foreign workers, but also American workers. It harms American workers because it gives some employers an incentive to choose foreign workers rather than Americans.
Not all of us run Windows, so we can't beta-test IE 6 unless MS suddenly decides to start supporting platforms outside of Windows and Macintosh.
/usr), but with the permissions set so that only root can execute the browser.
Actually, there is a version of IE "for UNIX". It's actually for Solaris (and maybe HPUX too?), but the only page that says this is the download page. Everywhere else they say "for UNIX", to make it sound like they support just about every platform in common usage...
I tried out IE for Solaris. It's missing a lot of the features of the Win32 version. If you install it, you'll also see that Microsoft doesn't quite understand this "UNIX" thing... half of the install goes in your home directory (rather than somewhere in
That explains client server. I guess a 3-tier architecture would be a... never mind.