Subject: Re: CL & CORBA From: Erik Naggum <clerik@naggum.no> Date: 1998/09/11 Newsgroups: comp.lang.lisp Message-ID: <3114488102043328@naggum.no> * Bjoern-Falko Andreas <andreas@ki.informatik.uni-ulm.de> | There is still no standard of how to write maintainable code. why this clamoring for standards? | But the style of other people? You have to admit that understanding | their line of thought isn't always easy to understand. no, it isn't _always_ easy. but so? I said 10% were a horror, and 20% were uncomfortable, but the _rest_ is "easy". however, it sounds like you are of the "easy generation", for whom everything must be easy. | For instance, I had to track down a bug that showed up only 1 in a 1000 | times. I wasn't even able to locate where it originated from. After | spending 16+ hours of tracing in one session I found it. By sheer luck. | It's still preferrable not to have to do this. There is not enough | aspirin in the world to cure all the headaches browsing foreign code | produces. the cause of your headaches isn't "foreign code", it's your attitude to work in general. debugging is painful, it's a fucking waste. sometimes, it's better to rewrite the code from scratch. sometimes it's better to scrap the whole damn thing and consider it impossible. which is better of debugging, reimplementing, and abandoning takes some experience to know. if debugging is the best choice, at least you know the other two choices are more work and/or less rewarding. that's _something_. FYI, I spent nine months debugging and understandign 25,000 lines of horrible C code -- it was fun in the way that all meaningless challenges are. you'll have a very hard time convincing me you deserve sympathy for your 16+ hours. the most insidious bug in GNU Emacs I found was so hard to pin down I found it only after having worked on it for more than 300 hours over several months. you see, I _hate_ crashes, and this was a bug that _sometimes_ caused Emacs to attempt to redraw the screen before it had updated its size parameters after a resize, causing Emacs to write screen contents into all sorts of unrelated memory, some of which took until the next GC to crash Emacs, and some of which caused bizarre behavior elsewhere. this is not the proper forum for macho bug-killing stories, however. you just need to know that we've all been there and that we've all spent days, weeks, or months on trivial problems. it's what this trade requires of its practitioners, just like doctors have to live with patiens who die. it sucks, but if you can't hack it, get out. #:Erik -- http://www.naggum.no/spam.html is about my spam protection scheme and how to guarantee that you reach me. in brief: if you reply to a news article of mine, be sure to include an In-Reply-To or References header with the message-ID of that message in it. otherwise, you need to read that page.