glenn@heaven.woodside.ca.us (Glenn Reid) (05/10/91)
Russ Nelson writes > In article <489@heaven.woodside.ca.us> glenn@heaven.woodside.ca.us (Glenn Reid) writes: > > Imagine two carpenters building a door. One builds the frame, the other > builds the door. Neither one is on-site, and neither has met the other. > They are both working from the "spec", or the blueprint. How likely > do you think it is that the door fits in the frame, that the latch lands > squarely in the hole in the door frame, and that the hinges are set > properly? I wouldn't give it 1 chance in 100,000, even with experienced > carpenters. > > I'm afraid that you're revealing your ignorance here, Glenn (don't > worry, I've done it *lots* of times myself). Not only *can* it be > done, it *has* been done. Sure it's been done. But on the whole, it doesn't work that way very often. Also, it's actually much easier to get timber framing right than it is to hang a door correctly. I was, of course, exaggerating beyond trying to make a point in suggesting the odds to be so distant as 100,000 to 1, but the underlying point, that I think is still good, was that it's difficult for two companies to write software against a spec and then expect it to work the first time. Someone else (I think Tom) made a much more valid point than suggesting my ignorance (ignorance of what, by the way?). That point was that the software industry is often *compelled* to do simultaneous work from a specification, and doesn't usually have the luxury of interconnected testing. Sun actually does a great job of this by putting on the NFS Connectathon every year, in which hundreds of vendors come and set up their equipment in a huge conference room and they build a network and test everything against everything else, and keep a matrix of what kinds of things worked correctly. It's intended to help vendors work the kinks out of their NFS software, and from what I've heard, it works great. -- Glenn Reid RightBrain Software glenn@heaven.woodside.ca.us NeXT/PostScript developers ..{adobe,next}!heaven!glenn 415-326-2974 (NeXTfax 326-2977)
nelson@sun.soe.clarkson.edu (Russ Nelson) (05/10/91)
In article <497@heaven.woodside.ca.us> glenn@heaven.woodside.ca.us (Glenn Reid) writes: Russ Nelson writes > In article <489@heaven.woodside.ca.us> glenn@heaven.woodside.ca.us (Glenn Reid) writes: > Imagine two carpenters building a door. One builds the frame, > the other builds the door. Neither one is on-site, and > neither has met the other. They are both working from the > "spec", or the blueprint. How likely do you think it is that > the door fits in the frame, that the latch lands squarely in > the hole in the door frame, and that the hinges are set > properly? I wouldn't give it 1 chance in 100,000, even with > experienced carpenters. > > I'm afraid that you're revealing your ignorance here, Glenn (don't > worry, I've done it *lots* of times myself). Not only *can* it be > done, it *has* been done. Sure it's been done. But on the whole, it doesn't work that way very often. Also, it's actually much easier to get timber framing right than it is to hang a door correctly. I was, of course, exaggerating beyond trying to make a point in suggesting the odds to be so distant as 100,000 to 1, but the underlying point, that I think is still good, was that it's difficult for two companies to write software against a spec and then expect it to work the first time. I agree, it is difficult. But I also added that we might be able to do it after we've had several centuries of experience doing it. Someone else (I think Tom) made a much more valid point than suggesting my ignorance (ignorance of what, by the way?). Ignorant of a counterfoil to your argument. Yes, timber framing is less exact than hanging a door. On the other hand, 300+ carpenters made timbers from the same spec that mostly all fit together. I think that's an achievement of equal difficulty to the one you proposed. That point was that the software industry is often *compelled* to do simultaneous work from a specification, and doesn't usually have the luxury of interconnected testing. The question is whether it's a luxury or whether it's necessary. I think we both agree that it's necessary. -- --russ <nelson@clutx.clarkson.edu> I'm proud to be a humble Quaker. Clear cutting is criminal, spiking trees is criminal, and using hyperbole of this magnitude in a serious discussion is criminal. -- Irv Chidsey