Re: Naive DEFSYSTEM replacement
"Christopher Stacy (as cstacy at dtpq dot com)" <[email protected]>
| Newsgroups | gmane.lisp.lispworks.general |
|---|---|
| Message-ID | <[email protected]> |
On 7/28/25 10:42 AM, Robert Goldman (as rpgoldman at sift dot net) wrote: > > On 28 Jul 2025, at 6:15, Marco Antoniotti (as marco dot antoniotti at > unimib dot it) wrote: > > Yep. That is the crux of the problem. > > And as I show in my response to Yuri, this was an /intentional > decision/ on the part of the standardizers of the language. > Yes. Because people could not be convinced that all the hair of the Lisp Machine pathname system could be needed to deal with all the variant operating (file) systems. Turns out, it *was* necessary. But X3J13 just punted. CL has thereby suffered this from the beginning, with myriad implementations of partial solutions in many libraries and application all over the place. This is what you get when you won't accept the right answer because you dislike that the people who are offering it are so often correct, and you call them "arrogant". Personality clashes in standards bodies. I am sure there are examples of similar disasters in many other standards... The other "portable" languages don't have these issues, because they think "portable" means "everything is either Unix or Windows and we don't care about semantics".