begin quoting Andrew Lentvorski as of Mon, Dec 17, 2007 at 12:21:32PM -0800: > Ralph Shumaker wrote: > >Well, then, I guess winmodems were on the right track after all, eh? > > Um, the problem with winmodems wasn't that they ran in software.
I recall someone saying that winmodems were just a fast A/D & D/A converter, and if they'd only gone that extra bit and made 'em totally generic, it would have been a wonderful standard feature. > It was the fact that they were completely undocumented and had no way to > write a non-Windows driver for them. I wonder how much MS had to do with that. > Even worse was that we had a perfectly fine de facto standard (the Hayes > Smartmodem command set) that those stupid winmodems displaced so that > the manufacturers could save a couple pennies (which it's not clear they > saved given the problems and support costs those winmodems caused) It worked for the first couple of quarters, presumably, which is all the board of directors cared about. -- In theory, market forces would prevent such things. In practice, they encourage. Stewart Stremler -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
