From ... From: Erik Naggum Subject: Re: Java is really convenient. Re: Sun thinks about switching Java to S-expression syntax: Lava Date: 1999/02/16 Message-ID: <3128197088724992@naggum.no>#1/1 X-Deja-AN: 445123304 References: <36C4A10F.164FC49F@IntelliMarket.Com> <7aattr$8e6@crl3.crl.com> <873e46hfyo.fsf_-_@2xtreme.net> <87soc6f7xn.fsf@2xtreme.net> mail-copies-to: never Organization: Naggum Software; +47 8800 8879; http://www.naggum.no Newsgroups: comp.lang.lisp * cbarry@2xtreme.net (Christopher R. Barry) | Common Lisp has a kinda oddball size for FIXNUMs. What is it doing | about 64-bit platforms? What will it do when we've got 128-bit ones? geez, why do you care? Common Lisp doesn't _have_ sizes of such things. implementations do. C and C++, however, _do_ have sizes of such things. C/C++ have the problem you allude to. Common Lisp just runs with a little less consing on a machine with bigger machine integers. | Why can't Lisp become more popular? why can't opera be more popular? why can't football be less popular? | What's stopping it? people like you. | You can't do sockets or threads in CL without a hell of a lot of #+ and | #- and have it work in any CL providing access to these features. just because you can't doesn't many other people can't. the simplest way to do this is to write a thin veneer, perhaps using compiler macros, to present a uniform, standard interface to the underlying implementation (which, by the way, differs between _socket_ implementations, too), stuff this in a file that is loaded only on the platform it applies to. porting your system is a matter of making a new copy of this file. this, by the way, is how intelligent people _implement_ real portability. only if the differences are very small and very localized does it make sense to use #- and #+. personally, I use #- and #+ only in configuration files and in DEFSYSTEMs. feel free to complain that this veneer isn't standardized. I fully expect you to, but also that you will complain about whatever you get, the same way people complain about the pathname abstraction prohibiting them from running the whole gamut of file system options and operations on their particular implementation. #:Erik