| PLEX86 | ||
the new math: old battle of the lovees was: PDP1 3585the new math: old battle of the lovees was: PDP1 3586 On Wed, 07 Jun 06 11:19:44 GMT in alt.folklore.computers, Comment in a book I read recently about OO being successful because you... Sigh! This is fine and necessary but all it will do is tell you if you've reinvented an old bug. It never shows you the new stuff. And the whole OS development biz is about the new stuff. It takes a rigorous field test when all those pesky users ;-) do their normal, everyday, bizarre, and diabolical computer work. Once again, this is about a timesharing OS not the other kinds.
Yes, I know about those. It is always the usage by normal everyday people that show up the abnormal bugs. You cannot regression test random user interactions with both, the computer and each other. There may be. It all depends on what the new version is supposed to do this time around. The guys would often long to rewrite code because they had learned from their mistakes. However, the mechanics of doing all the work required to ship that code usually would have extended the project by another year. When your core business is delivering hardware with the software to drive it, you do not delay ships. People here can confuse the OS development biz with application biz. They are very different. The differences have been hidden because an OS is no longer tied to a piece of hardware delivery these days; it's done by two companies and cannot be a package. This at the heart of why Microsoft's stuff is Micpoo. You cannot look at Unix for information because the arithmetic sign of AT&Ts quarterly net profit did not depend on whether the new device driver shipped on time. BAH
|
||||
the new math: old battle of the lovees was: PDP1 3586 Alt Folklore Computers from Newsgroups The #1 Usenet Provider on the Internet
|
||||