Archive of UserLand's first discussion group, started October 5, 1998.

How I Knew OpenDoc Would Bomb

Author:Dave Winer
Posted:5/9/1999; 5:59:14 PM
Topic:How I Knew OpenDoc Would Bomb
Msg #:5911
Prev/Next:5910 / 5912

Four years later It's been four years since I wrote The Subjectives of OpenDoc, and at the time I was prohibited from saying anything but what I thought, by a personal confidentiality agreement. But now, years later, the people I was talking to are long-gone from Apple, and the bosses they were protecting are long-gone too. I won't name any names, but I will tell you what I knew. Remember IBM and Novell I knew that Apple engineers were far from done with their implementation when OpenDoc became a political football between IBM and Novell and probably others. There was an alliance, it was a joint development project, but each of the companies already had APIs and a source code base and had done some amount of briefing developers on their technology. So the execs decided that instead of choosing one approach, one API, one codebase, they'd "merge" all three. These are the kinds of decisions execs make. The engineers screamed loudly, as only Apple people could. (This seems to have changed in Jobs' New Apple.) Basically out of Apple's hands, making any changes to OpenDoc was a lugubrious political process, and as Microsoft people can attest (they do, at length) the IBM people didn't "get" PC software, and they complained that the Novell people didn't understand the Mac. "It'll be ten years before there are any developer tools for OpenDoc," one friend said. "The docs will fill a bookshelf," said another. What do you do? Cut your losses ASAP. And if you're a responsible developer, responding to your users' enthusiasm, whipped up by the execs with their visionary speeches, tell your users No anyway. This is how you make trouble for yourself, but long-term it's the best practice if you want your word to mean anything. Honesty These kinds of channels exist, always, in the software business. Engineer-to-engineer there's a lot more honesty than exec-to-exec, or exec-to-engineer, and sometimes even engineer-to-exec. Information and judgment change hands, cross company lines, a consensus builds, decisions are made. If I see developers moving, as I see happening now with Linux for example, I am more likely to move with them. If I see no movement, I don't move, unless I'm inspired (I am proud that I was one of the very first people to ship a Mac app in 1984). And if insiders on the development team tell me it has no hope, I believe them. Anyway, that's the story. Hope this helps explain what was going on back then.



There are responses to this message:

This page was archived on 6/13/2001; 4:49:57 PM.

© Copyright 1998-2001 UserLand Software, Inc.