Re:Daddy, why are we hiding from the police?
on
FreeBSD 4.2 Is Out
·
· Score: 1
Yes, I use that autoindent frequently... but it's not nearly as good as Emacs'. vi's autoindent just places point at the last line's indent position. Emacs will indent appropriately, line up braces, et cetera, as you type. Very helpful. For instace, if you type this
01: while (someCondition)
02: {
03: do_something ();
04: }
05:
06:/* lots more code */
and later, mindlessly/drunkenly/accidentally delete the right brace at line 4, Emacs will automatically indent everything afterwards appropriately, which should make it obvious that you have an open brace. This greatly reduces the number of "stupid little programmer errors" that are time-consuming to isolate at compile time.
Normally, this is what indent(1) is for. However, while GNU indent works with C, C++, and Java, the indent I have avilable at work (/usr/ucb/indent on AIX 4.3) only understands C. If I'm using Perl or some other non-C family language, Emacs is not only the best choice for automatic source formatting (IMHO), but it is the only choice in some cases.
While vi's autoindent still requires you to manually indent the first line in a block, and also requires you in un-indent closing braces, Emacs does everything for you, and will also tell you, as you type, if you have mismatched parenthesis, braces.
That's why I'm willing to give up my favorite editor now and then. I mean, Emacs always takes up at least 1MB of memory; it must do something useful, right?;-)
Wow, I don't think a vi user has ever been as kind to Emacs as I just have.
All generalizations are false.
Re:Daddy, why are we hiding from the police?
on
FreeBSD 4.2 Is Out
·
· Score: 1
Yes, I have that shirt.:-) But the funny thing was, I ordered it, and they sent me the Emacs version! I was so pissed! Luckily, I was able to exchange it quickly. I think the first day I wore it to work was the day my boss lost all hope for my sanity.
For the record, I very much prefer vi, but often use GNU Emacs when editing large programs because I like the auto-indent. I also don't mind the syntax highlighting available when running from X. Yes, I've used Vim. Its syntax highlighting leaves much to be desired. The C, C++, Java, and Perl modes are pretty good, but I've had consistent problems with Vim's interpretation of quoted strings in SQL mode. And in any case, it doesn't to the fab auto-indent. (Which is a finger-saver for those who use spaces (as opposed to hard tabs) for indenting!!) It's a pity, because otherwise think Vim is an excellent program, and it doesn't conflict too much with the Berkeley and SVR4 vi's I'm used to.
I've tried Emacs' various vi-alike modes, but they just don't cut it. See, "viper" and the like are just vi-like enough so that I get comfortable, and then I discover that it doesn't emulate all of vi's keymaps, and it fucks me up.
Anyway, even syntax highlighting and auto-indent aren't to lure me away from vi. God bless BJ.
...and I just upgraded my boxes to 4.1 a few weeks ago! Oh well. 4.1 was the first release to show BSDi's influence, and I'm curious to see if this will continue to become more apparent as time goes on.
I don't have a PS2 yet, so I keep a Win98 box around with a DVD-ROM and a RealMagic PCI card to watch movies.
The RealMagic sucks. Even on Win98, the only platform where the damned thing realiably works, the quality is pretty bad. Hopefully the PS2 will be an improvement.
Anyway, back on topic... the RealMagic software won't let you take screenshots on copyrighted DVDs, and there's not a way around it in my experience. (The video card never really "sees" the decoded DVD data, and any attempts to take a screenshot by other means will leave you with an image of a big black box. After playing with it for few hours once, I was able to find a workaround, but it didn't work on American DVDs. Odd.) That was my original reason for downloading DeCSS, to take screenshots.
I dig DVDs, but the whole "we must completely control your viewing experience" thing is starting to piss me off. Maybe even enough to go back to... VHS? Nahh...
Well, a phone can only be jammed up an ass so many times before it has to be thrown away.
After the tenth Anal Exploration, the little speaker holes tend to get clogged up with shit. Even with a paper clip and some Kleenex, there's only so much you can do. The amount of shit that can be removed from the phone is calculated with this equation:
lc / n = g
where n is number of ass-stuffings, l is the length of the phone cord, and c is the average consistency of the Stuffer's shit. Common values for c are 5 (for average, well packed poo), 10, (for rock-hard, constipation-compressed feces), or 1 (for syruppy diarrhea with the occasional chunky, greasy fart). The quotient g is the number of grams of fecal matter that can be expected to be salvaged from the phone's inner workings without damaging the microphone or speaker.
For instance, if my phone cord is 4 feet long, I've stuffed the phone up my ass 6 times, and my average shit consistency is 3 (shapely and well-formed, but rather mushy turds), I would use the equation
g = 4(3)/6
to determine that 2 grams of shit can be safely removed from the phone. For values of n above 12, the difficulty of shit removal overshadows the cost of a new phone, and therefore I would usually head out to Best Buy and purchase a new one after the tenth or eleventh ass-stuffing.
Unfortnately, I doubt that ass-related damage was covered by Ma Bell's warranty, so the benefit of a subscription phone plan is negligible to the ass-stuffing phone owner. Perhaps if this phone system were still around today, Ma Bell would offer an ass-stuffer's insurance which would be more financially convenient to the ass-stuffing consumer than purchasing a new phone from Best Buy (long known to be the ass-stuffing phone enthusiast's superstore for ass-stuffing goods and services).
Go try to buy a copy of Oracle and you'll have an option to buy a perpetual license, a 2-year license, or one of a few other options.
Yeah, well Oracle's a little weird. Do they still make you compute those Magical Oracle Power Units before purchasing? I'm trying to remember the forumla... something like (number of computers) * (number of processors) * (speed of processors) * (architecture rating)... whatever.
Right now, for instance, buying direct from MS is usually the most expensive option, since they sell almost everything at full MSRP.
With the types of sever-side technologies available today (i.e. Java, or in Microsoft's case, C#), I think we'll eventually see ASPs relying on web browser-based applications. Even today, a subscription word processor could easily be implemented as an applet commuicating with a servlet, CGI application, or ActiveX application over an SSL connection.
Why do I bring this up? Well, I do the majority of my non-trivial text processing in GNU Emacs, but there are situations when a copy of Microsoft Word would come in handy. When revising my résumé, for instance, although usually I write it in XHTML, many potential employers expect you to send a Word document. If I could access Microsoft Word through an applet-capable UNIX web browser, sure, I'd be willing to pay $5 for an hour's access. It's a lot cheaper than buying the $300 (or however much it costs!) full version, and even better, I don't have to worry about giving up Solaris or GNU/Linux to use it.
I believe that eventually the WWW will be the only application platform that matters for most business users. It's Sun's "WORA" philosophy taken to a whole new level.
I think that progress in this direction is held back by two problems. The first is bandwidth. Even if the WWW-based work processor described above were extremely modular, there's no way you could have 5000 employees using the application over a few T3's. Clearly the "buy the media and licenses" model will be around for a while, but if the bandwidth trickle-down continues at its current pace, I think this not an unreasonable goal for the next decade.
The second problem is many UNIX programmers' reliance on old-school server-side programming methodologies. Like it or not, Java and its lightweight object-oriented brethren are here to stay. We're in the middle of a period of evolution -- much like the jump from assembler to C -- when program abstraction is moving to a new level. (I can no longer say things like "business logic" and laugh.) Unfortunately, I see a bigotry in many UNIX users towards anything but the "C/Perl CGI" model of server-side programming. Even C++ is sometimes looked down upon. These attitudes will have to change; there will always be a place for those technologies, but as web programming more fully embraces OOP (1) (and database-driven OOP apps), I see more and more server-side programming being done in Java (generally), C# (for microsoft-freaks), and C++ (when neither of the first two is fast, powerful, or flexible enough). My experiences in the last year have convinced me that this approach is not only does increase programmer productivity (less time designing, AND less time coding), but allows us to create systems of a complexity that would be unthinkable in ANSI C or Perl 5.(2)
Perl has the foundations to make a place in this new world, but it needs better support for high-level programming, and better database support. I've spend a good portion of the past two months converting my employers old Perl scripts to Java servlets, simply because Java has better database support (gotta love JDBC connection pooling) and servlets scale much better on high-load sites than Perl CGI(3), at least with my employer's setup.
I have faith that Larry Wall and crew can pull this off. Listen to the webcast of Larry's "Camel Lot #6" speech (4)from mid-October. C++ and Java programmers who've dabbled with Perl should enjoy the latter half of the speech.
However you feel about Microsoft, it's interesting how long ago they saw this coming, and how well they've prepared for it. Windows 2000 will continue to kick our ass in this arena for a few years, because the operating system has capabilities that we UNIX goons need application servers (iPlanet Application server, BEA WebLogic, et cetera) for.
Amusing... in ten years, it might not be uncommon for every worker to have a diskless X terminal feeding of a central (UNIX, GNU/Linux, NT) server, running "productivity" applications (read: Office) over the Internet with Internet Explorer 27 or Netscape 6 (grin;-). It's amusing because that's the same model that we were supposed to have escaped when PC's got cheap and every employee had his own. (A new era of workplace surveillance? Likely. A new era of employment for UNIX sysadmins. Definitely.)
Anyway, that's enough ranting from me.
Footnotes.
1. Perl isn't a good OOP language. The basics are there, but it doesn't even support encapuslation properly. This will change, however.
2. It also allows us to create systems that can bring today's mighty servers to their knees. If I were a conspiracy theorist, and I am, I would postulate that one of Sun's motivations for spreading Java's use is that it creates a need for their monster hardware setups again.;-) "Haha! Just try and run four JVMs on that 600MHz Athlon, kiddies!"
3. The normal response to this is, "What about Slashdot? They use Perl and they're a high-load site!" The only reasons that Slash is able to run a site like this are mod_perl and the Arrowpoints. Mod_perl is awesome, but my employer runs iPlanet on RS/6000s, so it's not an option. True, Java is fucked above 100k HTTP connections, but that's what clustering is for, right?;-)
4. http://www.technet cas t.com/tnc_play_stream.html?stream_id=375
Now the new wave and electronic styles of the 80's are re-emerging.
You mean to say that maybe in a year, I'll be able to play my Kraftwerk CDs loud without shame?;-)
Now, re: your.sig... could you possibly point me to the source of that quote? (I code Java, but I have a sense of humour. I think it's almost necessary, if you catch my drift.:-)
Seriously. True "death metal" has never been part of American mainstream music. The closest thing to "death metal" that was ever considered popular was pre-Youthanasia Megadeth (Mustaine isn't anything special without his cocaine;-) and much of Slayer's material, but even that is pretty tame compared to your average European metal band.
The original post was flamebait. Personally, I agree with the anti-Backstreet Boys sentiment: I can't respect artists who don't even write their own music. They're amateurs, and couldn't survive as real musicians without support of a major record label and MTV. The fact that the poster referred to Manson as "death metal" just clarifies the fact that he doesn't know what he's talking about or is simply trolling.
Washingtonpost.com is a "free service", like Slashdot. They sell advertising, like Slashdot. They produce original material, as well as crib content from other sources, like Slashdot. The people who run the site do so for a living, like those who run Slashdot. But unlike Slashdot, they run the site in a profressional manner.
The moment that this site became profitable, Rob and the rest of those goons lost their excuses for such fuck-ups. Our pageviews are what sells advertising; indirectly, we pay for Rob's weed, so we have a right to complain when the level of "service" drops. The management has the right not to listen (which they do very well). The duplicate story problem is well known and recurs often. The management doesn't care. If it really bothers you, stop reading. The management won't care if you stop reading, but they will care when they start losing advertising revenue. If you say nothing, you are accepting that the service rendered is adequate. So go ahead and complain, fellow Slashdotters. Or even better, stop visiting the site.
I used to say, "oh, leave Rob and the rest alone; it's all just a fun geek site, right?", but then I thought, what else does Malda do all day? He stops playing Diablo 2 a few times a day to post these lame stories, and he can't even check for duplicates? That pissed me off. Why should we support Rob's right to be a lazy sod and get rich off of a second-rate site?
Slashdot is big business now. Corporate sponsors, national media exposure, the works. It's time for the management to realize this, and stop doing such a half-assed job. If you want to continue deluding yourself into thinking Slashdot was the same place it was two years ago, or even a year ago, or even six months ago, go ahead. But it's not the same. Dell Computers got their start in a college dorm room, and now they're a major corporation. Consequently, they're expected to act more professionally then when they were run from a dorm room. We should expect no less of Slashdot. If Rob's such a Perl wizard, why can't modify Slashcode so that every accepted story submission is put into a DB table, and compared against previous submissions for similar keywords and links? The editor can then maually review possible duplicate submissions. What else is so pressing that you can't find the time to actually maintain the site that made you a millionare, Rob?
The bottom line is that Slashdot is too successful for him to not care.
Taco hasn't seen anything that cool in months? Yeah, two months sounds about right.:-) Between his spelling and his memory, I bet he qualifies for crippy plates by now.
I'd be much more wary of an OS developed by a software company whose CEO is a BASIC programmer. Complaining that the defrag tool was written by a Clam is like ordering a bowl of soup, being served a bowl of cold piss, and then complaining that the piss has a fly in it. What's the difference between Microsoft and CoS? One is a cult of mindless drones led by a maniacal dictator who will stoop to even the lowest, most vile means to coerce their enemies into submission, and the other is the CoS.
</mandatory MS bashing>
(Before you start flaming me, let me say that I do use NT (4 and 5) in addition to UNIX systems. It's not all that bad. But when you consider the almost infinite resources that Microsoft has available, don't you expect something more for $300?)
---------///---------- All generalizations are false.
Slashdot Mirror
All generalizations are false.
Normally, this is what indent(1) is for. However, while GNU indent works with C, C++, and Java, the indent I have avilable at work (/usr/ucb/indent on AIX 4.3) only understands C. If I'm using Perl or some other non-C family language, Emacs is not only the best choice for automatic source formatting (IMHO), but it is the only choice in some cases.
While vi's autoindent still requires you to manually indent the first line in a block, and also requires you in un-indent closing braces, Emacs does everything for you, and will also tell you, as you type, if you have mismatched parenthesis, braces.
That's why I'm willing to give up my favorite editor now and then. I mean, Emacs always takes up at least 1MB of memory; it must do something useful, right? ;-)
Wow, I don't think a vi user has ever been as kind to Emacs as I just have.
All generalizations are false.
For the record, I very much prefer vi, but often use GNU Emacs when editing large programs because I like the auto-indent. I also don't mind the syntax highlighting available when running from X. Yes, I've used Vim. Its syntax highlighting leaves much to be desired. The C, C++, Java, and Perl modes are pretty good, but I've had consistent problems with Vim's interpretation of quoted strings in SQL mode. And in any case, it doesn't to the fab auto-indent. (Which is a finger-saver for those who use spaces (as opposed to hard tabs) for indenting!!) It's a pity, because otherwise think Vim is an excellent program, and it doesn't conflict too much with the Berkeley and SVR4 vi's I'm used to.
I've tried Emacs' various vi-alike modes, but they just don't cut it. See, "viper" and the like are just vi-like enough so that I get comfortable, and then I discover that it doesn't emulate all of vi's keymaps, and it fucks me up.
Anyway, even syntax highlighting and auto-indent aren't to lure me away from vi. God bless BJ.
Okay, that was offtopic. :-)
All generalizations are false.
All generalizations are false.
Hey, where's that Java 2 I was promised?
Anyway, this release is predisposed to good karma because of 4.2BSD.
All generalizations are false.
All generalizations are false.
The RealMagic sucks. Even on Win98, the only platform where the damned thing realiably works, the quality is pretty bad. Hopefully the PS2 will be an improvement.
Anyway, back on topic... the RealMagic software won't let you take screenshots on copyrighted DVDs, and there's not a way around it in my experience. (The video card never really "sees" the decoded DVD data, and any attempts to take a screenshot by other means will leave you with an image of a big black box. After playing with it for few hours once, I was able to find a workaround, but it didn't work on American DVDs. Odd.) That was my original reason for downloading DeCSS, to take screenshots.
I dig DVDs, but the whole "we must completely control your viewing experience" thing is starting to piss me off. Maybe even enough to go back to... VHS? Nahh...
All generalizations are false.
All generalizations are false.
After the tenth Anal Exploration, the little speaker holes tend to get clogged up with shit. Even with a paper clip and some Kleenex, there's only so much you can do. The amount of shit that can be removed from the phone is calculated with this equation:
where n is number of ass-stuffings, l is the length of the phone cord, and c is the average consistency of the Stuffer's shit. Common values for c are 5 (for average, well packed poo), 10, (for rock-hard, constipation-compressed feces), or 1 (for syruppy diarrhea with the occasional chunky, greasy fart). The quotient g is the number of grams of fecal matter that can be expected to be salvaged from the phone's inner workings without damaging the microphone or speaker. For instance, if my phone cord is 4 feet long, I've stuffed the phone up my ass 6 times, and my average shit consistency is 3 (shapely and well-formed, but rather mushy turds), I would use the equation to determine that 2 grams of shit can be safely removed from the phone. For values of n above 12, the difficulty of shit removal overshadows the cost of a new phone, and therefore I would usually head out to Best Buy and purchase a new one after the tenth or eleventh ass-stuffing.Unfortnately, I doubt that ass-related damage was covered by Ma Bell's warranty, so the benefit of a subscription phone plan is negligible to the ass-stuffing phone owner. Perhaps if this phone system were still around today, Ma Bell would offer an ass-stuffer's insurance which would be more financially convenient to the ass-stuffing consumer than purchasing a new phone from Best Buy (long known to be the ass-stuffing phone enthusiast's superstore for ass-stuffing goods and services).
Thank you. I'll have another G&T, please.
All generalizations are false.
All generalizations are false.
All generalizations are false.
Why do I bring this up? Well, I do the majority of my non-trivial text processing in GNU Emacs, but there are situations when a copy of Microsoft Word would come in handy. When revising my résumé, for instance, although usually I write it in XHTML, many potential employers expect you to send a Word document. If I could access Microsoft Word through an applet-capable UNIX web browser, sure, I'd be willing to pay $5 for an hour's access. It's a lot cheaper than buying the $300 (or however much it costs!) full version, and even better, I don't have to worry about giving up Solaris or GNU/Linux to use it.
I believe that eventually the WWW will be the only application platform that matters for most business users. It's Sun's "WORA" philosophy taken to a whole new level.
I think that progress in this direction is held back by two problems. The first is bandwidth. Even if the WWW-based work processor described above were extremely modular, there's no way you could have 5000 employees using the application over a few T3's. Clearly the "buy the media and licenses" model will be around for a while, but if the bandwidth trickle-down continues at its current pace, I think this not an unreasonable goal for the next decade.
The second problem is many UNIX programmers' reliance on old-school server-side programming methodologies. Like it or not, Java and its lightweight object-oriented brethren are here to stay. We're in the middle of a period of evolution -- much like the jump from assembler to C -- when program abstraction is moving to a new level. (I can no longer say things like "business logic" and laugh.) Unfortunately, I see a bigotry in many UNIX users towards anything but the "C/Perl CGI" model of server-side programming. Even C++ is sometimes looked down upon. These attitudes will have to change; there will always be a place for those technologies, but as web programming more fully embraces OOP (1) (and database-driven OOP apps), I see more and more server-side programming being done in Java (generally), C# (for microsoft-freaks), and C++ (when neither of the first two is fast, powerful, or flexible enough). My experiences in the last year have convinced me that this approach is not only does increase programmer productivity (less time designing, AND less time coding), but allows us to create systems of a complexity that would be unthinkable in ANSI C or Perl 5.(2)
Perl has the foundations to make a place in this new world, but it needs better support for high-level programming, and better database support. I've spend a good portion of the past two months converting my employers old Perl scripts to Java servlets, simply because Java has better database support (gotta love JDBC connection pooling) and servlets scale much better on high-load sites than Perl CGI(3), at least with my employer's setup.
I have faith that Larry Wall and crew can pull this off. Listen to the webcast of Larry's "Camel Lot #6" speech (4)from mid-October. C++ and Java programmers who've dabbled with Perl should enjoy the latter half of the speech.
However you feel about Microsoft, it's interesting how long ago they saw this coming, and how well they've prepared for it. Windows 2000 will continue to kick our ass in this arena for a few years, because the operating system has capabilities that we UNIX goons need application servers (iPlanet Application server, BEA WebLogic, et cetera) for.
Amusing... in ten years, it might not be uncommon for every worker to have a diskless X terminal feeding of a central (UNIX, GNU/Linux, NT) server, running "productivity" applications (read: Office) over the Internet with Internet Explorer 27 or Netscape 6 (grin ;-). It's amusing because that's the same model that we were supposed to have escaped when PC's got cheap and every employee had his own. (A new era of workplace surveillance? Likely. A new era of employment for UNIX sysadmins. Definitely.)
Anyway, that's enough ranting from me.
Footnotes. ;-) "Haha! Just try and run four JVMs on that 600MHz Athlon, kiddies!"
;-)
1. Perl isn't a good OOP language. The basics are there, but it doesn't even support encapuslation properly. This will change, however.
2. It also allows us to create systems that can bring today's mighty servers to their knees. If I were a conspiracy theorist, and I am, I would postulate that one of Sun's motivations for spreading Java's use is that it creates a need for their monster hardware setups again.
3. The normal response to this is, "What about Slashdot? They use Perl and they're a high-load site!" The only reasons that Slash is able to run a site like this are mod_perl and the Arrowpoints. Mod_perl is awesome, but my employer runs iPlanet on RS/6000s, so it's not an option. True, Java is fucked above 100k HTTP connections, but that's what clustering is for, right?
4. http://www.technet cas t.com/tnc_play_stream.html?stream_id=375
All generalizations are false.
All generalizations are false.
All generalizations are false.
All generalizations are false.
Now, re: your .sig... could you possibly point me to the source of that quote? (I code Java, but I have a sense of humour. I think it's almost necessary, if you catch my drift. :-)
All generalizations are false.
The original post was flamebait. Personally, I agree with the anti-Backstreet Boys sentiment: I can't respect artists who don't even write their own music. They're amateurs, and couldn't survive as real musicians without support of a major record label and MTV. The fact that the poster referred to Manson as "death metal" just clarifies the fact that he doesn't know what he's talking about or is simply trolling.
All generalizations are false.
All generalizations are false.
The moment that this site became profitable, Rob and the rest of those goons lost their excuses for such fuck-ups. Our pageviews are what sells advertising; indirectly, we pay for Rob's weed, so we have a right to complain when the level of "service" drops. The management has the right not to listen (which they do very well). The duplicate story problem is well known and recurs often. The management doesn't care. If it really bothers you, stop reading. The management won't care if you stop reading, but they will care when they start losing advertising revenue. If you say nothing, you are accepting that the service rendered is adequate. So go ahead and complain, fellow Slashdotters. Or even better, stop visiting the site.
I used to say, "oh, leave Rob and the rest alone; it's all just a fun geek site, right?", but then I thought, what else does Malda do all day? He stops playing Diablo 2 a few times a day to post these lame stories, and he can't even check for duplicates? That pissed me off. Why should we support Rob's right to be a lazy sod and get rich off of a second-rate site?
Slashdot is big business now. Corporate sponsors, national media exposure, the works. It's time for the management to realize this, and stop doing such a half-assed job. If you want to continue deluding yourself into thinking Slashdot was the same place it was two years ago, or even a year ago, or even six months ago, go ahead. But it's not the same. Dell Computers got their start in a college dorm room, and now they're a major corporation. Consequently, they're expected to act more professionally then when they were run from a dorm room. We should expect no less of Slashdot. If Rob's such a Perl wizard, why can't modify Slashcode so that every accepted story submission is put into a DB table, and compared against previous submissions for similar keywords and links? The editor can then maually review possible duplicate submissions. What else is so pressing that you can't find the time to actually maintain the site that made you a millionare, Rob?
The bottom line is that Slashdot is too successful for him to not care.
All generalizations are false.
All generalizations are false.
All generalizations are false.
All generalizations are false.
Actually, most distros started shipping Vim 5.7 quite a while ago. But I'm an nvi man myself.
Oh, wait...
All generalizations are false.
---------///----------
All generalizations are false.
I'd be much more wary of an OS developed by a software company whose CEO is a BASIC programmer. Complaining that the defrag tool was written by a Clam is like ordering a bowl of soup, being served a bowl of cold piss, and then complaining that the piss has a fly in it. What's the difference between Microsoft and CoS? One is a cult of mindless drones led by a maniacal dictator who will stoop to even the lowest, most vile means to coerce their enemies into submission, and the other is the CoS.
</mandatory MS bashing>
(Before you start flaming me, let me say that I do use NT (4 and 5) in addition to UNIX systems. It's not all that bad. But when you consider the almost infinite resources that Microsoft has available, don't you expect something more for $300?)
---------///----------
All generalizations are false.