Slashdot Mirror


User: ebh

ebh's activity in the archive.

Stories
0
Comments
803
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 803

  1. Re:Harlequin is in a funny state. on Salon on JWZ/Emacs/Mozilla/AOL and Nightclubs · · Score: 1
    Well, I guess that explains why Webmaker went into limbo. Now there's something that should be ripe for open-sourcing, since it grew out of CERN's original, but when I asked Harlequin about it, they said they had no plans to. :(

    Ob(Sub)Topic: I seem to remember hearing that AutoCAD is written in LISP. CACT?

    -Ed

  2. Re:Yawn on Is Usenet Dying? · · Score: 2
    Having been introduced to Usenet in 1981, I've seen quite a few more. The earliest I can remember are the 100th newsgroup, the net.motss flamewars, and "CAR FOR SALE IN NJ".

    In the latter case, AT&T in NJ was about 2/3 of the entire net, and their management was always threatening to stop forwarding news outside AT&T (thus making outside sites pay the long distance dialup costs). The whole community had to balance AT&T's free bandwidth against the stupid posts from AT&T people who didn't know what "Distribution:" meant, even after following the new users' guidelines that suggested that you read for three months before your first posting.

    Any other good death-of-the-net remembrances? Ahh, the glory days of being able to get a full feed, then pass it along to one or two downstream sites, all over a 2400bps modem.

    -Ed

  3. Three true stories on How To Write Unmaintainable Code · · Score: 1
    Long time listener, first time caller...

    I swear I am not making any of this up.

    1. A programmer I knew on a COBOL project in 1982 deliberately obfuscated his own code by making all his variables four letters of a keyboard diamond, e.g., ESXD. He developed a system where he could derive a variable's purpose based on which of the seven possible diamonds it belonged to, which corner of the diamond it started on, and whether it went around the diamond clockwise or counterclockwise.
    2. I worked on an embedded system which had been written in a weird macro-assembly-like language. When it came time to build the new version, going from an 8080 to a 68000 and from that assembly to C, our PHB was the only one who claimed he could meet the deadline, this by translating the code line-for-line. After all, the code works now, so a straight port will work without any of the engineers having to know how it really works (or how to program in C), right?
    3. Of course that same project went *WAY* behind schedule. A year and a half into it, the higher-ups decided to audit the project to see what was going wrong. Obviously, a big part of it was that the engineers didn't know how this decade-old patched-beyond-recognition code did its thing. Like good programmers, when we hit code we couldn't figure out, we commented it as such, e.g., "I don't know why this is here, but it breaks if we take it out," as a reminder to go back and decipher it someday. But then, to cover his backside in the audit, the PHB instructed a secretary to remove with prejudice all comments in the code that might indicate something we were unsure of. So she did. Good thing he didn't teach her how to hack SCCS files.

    That PHB got promoted three more times before he retired.