No its not

Post #25,139 by tuberculosis 1/22/02 2:28:57 PM 8/21/07 5:54:34 AM Reply	No its not I am running into a class of younger programmers that only know Java. Pointers frighten them and are disdained. I doubt most of them could write a decent string copy in C. Javazation of the workforce is producing a class of dumbed down programmer than only knows what Sun tell him. Since we can't get enough good programmers, we work to enable the shitty ones. Its a bit sad, really. The average hunter gatherer works 20 hours a week. The average farmer works 40 hours a week. The average programmer works 60 hours a week. What the hell are we thinking? Edited by tuberculosis Aug. 21, 2007, 05:54:34 AM EDT No its not I am running into a class of younger programmers that only know Java. Pointers frighten them and are disdained. I doubt most of them could write a decent string copy in C. Javazation of the workforce is producing a class of dumbed down programmer than only knows what Sun tell him. Since we can't get enough good programmers, we work to enable the shitty ones. Its a bit sad, really. The average hunter gatherer works 20 hours a week. The average farmer works 40 hours a week. The average programmer works 60 hours a week. What the hell are we thinking? Collapse All History
Post #25,144 by pwhysall 1/22/02 2:39:04 PM Reply	OT: Your Pic You might want to crop it square and squish it down to about 60x60. Peter Shill For Hire [link\|http://www.kuro5hin.org\|There is no K5 Cabal]
Post #25,272 by tuberculosis 1/23/02 12:57:58 PM 8/21/07 5:57:32 AM Reply	Re: OT: Your Pic Whats a good free tool on Lose200 for that? The average hunter gatherer works 20 hours a week. The average farmer works 40 hours a week. The average programmer works 60 hours a week. What the hell are we thinking? Edited by tuberculosis Aug. 21, 2007, 05:57:32 AM EDT Expand All History
Post #25,328 by pwhysall 1/23/02 4:41:32 PM Reply	Re: OT: Your Pic www.gimp.org Peter Shill For Hire [link\|http://www.kuro5hin.org\|There is no K5 Cabal]
Post #25,179 by marlowe 1/22/02 5:30:08 PM Reply	Pointers scare me a bit, too. And it's not because I don't understand them. It's because I do. My feelings about Java are very mixed. But the freedom from pointers is an item in the plus column. [link\|http://www.angelfire.com/ca3/marlowe/index.html\|http://www.angelfir...e/index.html] Sometimes "tolerance" is just a word for not dealing with things.
Post #25,182 by ChrisR 1/22/02 5:46:55 PM Reply	Depends on the language To do anything and everything in C, you had to do it with pointers. OTOH, in a language like Pascal pointers were quite useful and not nearly as awkward.
Post #25,239 by CRConrad 1/23/02 7:37:17 AM Reply	See that, Barry R? See? Muahahahaha... :-) (Thanks, Critter! :-)
Post #25,190 by wharris2 1/22/02 6:26:00 PM Reply	Scare me, too Mostly because people's careless use of pointers. Or deliberately "sophisticated" use of pointers. One application I had to maintain for a time was built on structures (yes, multiple) of pointers to pointers to pointers. Naturally if you get one of those suckers corrupted, you're up the creek unless you're really good at debugging, or very lucky. For a number of years now, I've taken up the "free(pointer); pointer = NULL;" metaphor; at least it makes core dumps easier to debug. "Beware of bugs in the above code; I have only proved it correct, not tried it." -- Donald Knuth
Post #25,205 by nking 1/22/02 9:27:06 PM Reply	pointers don't scare me but the misuse of pointers and sloppy code does. Esp when someone gives me the project to work on that someone else messed up. "Will code Visual BASIC for cash."
Post #25,209 by wharris2 1/22/02 10:07:31 PM Reply	One problem How many C programmers are careful and use pointers properly? Look at the Microsoft buffer overflow security problems. I am not privy to the code, but I would guess 95% of them are pointer problems. "Beware of bugs in the above code; I have only proved it correct, not tried it." -- Donald Knuth
Post #25,251 by marlowe 1/23/02 9:42:18 AM Reply	Pointers don't cause buffer overrun holes. People cause buffer overrun holes. A lot of them. And they use pointers to do it. Maybe we should take their pointers away from them. It's a lot more humane than killing all sloppy programmers. I could be convinced to tolerate pointers and pointer arithmetic, in the interests of efficient code, if only there were some safeguards. But in C and C++, there are no safgeguards. None. They're completely abusable. [link\|http://www.angelfire.com/ca3/marlowe/index.html\|http://www.angelfir...e/index.html] Sometimes "tolerance" is just a word for not dealing with things.
Post #25,275 by tuberculosis 1/23/02 1:07:06 PM 8/21/07 5:57:39 AM Reply	You're arguing for Java "If programmer's can't use pointer's correctly, lets take them away." So they did. Welcome to Java. OTOH, the String class in Java is so amazingly lame and underpowered that I find myself missing basic pointer arithmetic everytime I have to do string ops. The buffer overflow problems that are so famous have nothing to do with pointers and everything to do with the introduction and continued availability of patently unsafe api in the C standard library. strcat, strcpy, the scanf's are all insanely defined. Just because terrorists use explosives doesn't mean that the miners shouldn't have them. The average hunter gatherer works 20 hours a week. The average farmer works 40 hours a week. The average programmer works 60 hours a week. What the hell are we thinking? Edited by tuberculosis Aug. 21, 2007, 05:57:39 AM EDT Expand All History
Post #25,288 by ben_tilly 1/23/02 1:28:38 PM Reply	No As Larry Wall said: I do quarrel with logic that says, "Stupid people are associated with X, therefore X is stupid." Stupid people are associated with everything. The same applies to languages as well. Just because an idea is associated with Java doesn't mean that Java's mistakes are a condemnation of the idea. The idea of not having pointers because they aren't safe, and instead using only safe constructs is not original to Java. A few random languages that do the same thing include Lisp, Smalltalk, Perl, Python, Haskell, and so on. While has strengths and weaknesses, I strongly prefer the string handling in every one of these languages to what is commonly used in either C or Java. Cheers, Ben PS While problems in the C standard library lead to many buffer overflows, I have seen C programmers produce enough without that assistance for me to think that pointers are a problem in and of themselves.
Post #25,335 by tuberculosis 1/23/02 5:00:32 PM 8/21/07 6:00:41 AM Reply	You've completely confused me. Yes, pointers in the wrong hands are a problem. Yes, the average Java programmer is the wrong hands. Sadly, programmers capable of writing good C are getting more and more scarce. Yes, most biz programming doesn't need the level of efficiency of C. To get slightly more on topic, I'm pretty sick of biz app programming. Its all getting to be like shoveling the stables. Nothing interesting in it at all anymore. I think I need a real programming job. System software or something... I'm also finding the web to be more and more boring everyday. The average hunter gatherer works 20 hours a week. The average farmer works 40 hours a week. The average programmer works 60 hours a week. What the hell are we thinking? Edited by tuberculosis Aug. 21, 2007, 06:00:41 AM EDT Expand All History
Post #25,369 by admin 1/23/02 8:28:33 PM Reply	Interesting C biz programming... We've got dozens of TCP connections feeding us bond information at rates such that our database can't keep up with the traffic on a 12-way Sparc box. We're moving that bit into C code... :-) Very interesting stuff. Regards, -scott anderson "Welcome to Rivendell, Mr. Anderson..."
Post #25,470 by tuberculosis 1/24/02 12:16:11 PM 8/21/07 6:02:54 AM Reply	OT Question This is an oracle db? Did you know Oracle keeps saying they are dropping support for C libs and only want to support JDBC client libs going forward? Would this screw you up? The average hunter gatherer works 20 hours a week. The average farmer works 40 hours a week. The average programmer works 60 hours a week. What the hell are we thinking? Edited by tuberculosis Aug. 21, 2007, 06:02:54 AM EDT Expand All History
Post #25,510 by admin 1/24/02 1:50:30 PM Reply	Didn't know that. Yes, it would completely fuck us up. Of course, we're still on 8.1.7, and they have the support in 9.X so far, so we're good for at least another year. :-) I find it hard to believe they'd drop that, though. WAY too much ProC stuff out there. Regards, -scott anderson "Welcome to Rivendell, Mr. Anderson..."*
Post #25,531 by wharris2 1/24/02 2:29:08 PM Reply	Au contraire! I find it hard to believe they'd drop that, though. WAY too much ProC stuff out there. I can give one example that is somewhat similar. Informix dropped support for their C-language API libraries in the early or mid 90's. (They had support for embedded SQL, but the people using their application language library were out of luck - except to scramble to convert everything to embedded SQL, another language, or another database.) Granted, by that time Informix (once a leader in databases) was falling increasingly further behind Oracle, so their decision probably effected far fewer people, and I suppose most of their customers were probably using a 4GL rather than C, but at least one fairly major database player has* done something like that. Come to think of it, it is a bit odd that they got rid of an API interface entirely; I know DB2 has one and obviously Oracle still has one - at least for a while. Doing things like that may have been one reason they completely fell apart in the late 90's. "Beware of bugs in the above code; I have only proved it correct, not tried it." -- Donald Knuth
Post #25,554 by admin 1/24/02 3:29:57 PM Reply	Oracle I think is a different situation Informix was much smaller, and as you say, they didn't survive. I'd like to see a link (Todd) that references them giving up the C interface. Regards, -scott anderson "Welcome to Rivendell, Mr. Anderson..."
Post #25,579 by wharris2 1/24/02 4:56:12 PM Reply	Taking it to databases "Beware of bugs in the above code; I have only proved it correct, not tried it." -- Donald Knuth
Post #25,393 by ben_tilly 1/23/02 10:33:43 PM Reply	There shouldn't be any confusion I am saying that saying that arguing for a particular design decision that Java happens to have accepted is not an argument for Java any more than it is for any other language that makes the same decision. Furthermore the fact that Java did a bad job with it doesn't mean that it can't be done well. Finally I would argue that not only doesn't the average bizz application not need the performance of C, but it should not be written in C because most programmers can't be trusted with pointers. (The ones who can be trusted with them tend to be people like Dan Bernstein, whose first order of business is to write a library that protects him, and then writes to that library so he has as few opportunities as possible to f*ck up.) As for whether someone should have access to that power, well of course. You can't bootstrap yourself up the toolchain without someone, somewhere, creating the safety gear that everyone else uses. And for speed boosts, it should be possible to speed up the 5% of the code which is your bottleneck. But most programmers do not need and should not have access to that level. About your on topic comments, well of course. Most business programming is not done because it is interesting. It is done because someone needs it done and is willing to pay to see it so. Same goes for janitors, secretaries, and people working cash registers. And the web didn't become important because it is cool, interesting, and fun. It won because it was lowest denominator to the point where non-technical people could figure it out. As a development environment it sucks. And people who want to put stuff on it, well either they don't want to pay (in which case it is probably stuff that only they care about) or else they want to pay (in which case you still don't care about it). Either way you are unlikely to care much about it. Such is life. Let me give a different example of the same principle. I love books. I always have, probably always will. But when I walk into a bookstore I know what sections to look in. 90% of the shelves for me are an obstacle course between me and the sections I like. Nobody is asking me to help make the books I don't care about be available, so I don't often have cause to lament the situation. But if they did, I would. Cheers, Ben
Post #25,294 by marlowe 1/23/02 1:55:54 PM Reply	What Ben said. I'm not so much arguing for Java as arguing against C. I've got some more concrete ideas of what I'm actually for, but we're already way off topic here. Maybe in another forum... [link\|http://www.angelfire.com/ca3/marlowe/index.html\|http://www.angelfir...e/index.html] Sometimes "tolerance" is just a word for not dealing with things.

Welcome to IWETHEY!