home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!uunet!not-for-mail
- From: guy@auspex.com (Guy Harris)
- Newsgroups: comp.std.unix
- Subject: Re: Unix Wars Over (Query)
- Date: 31 Aug 1992 02:33:48 -0700
- Organization: Auspex Systems, Santa Clara
- Lines: 25
- Sender: sef@ftp.UU.NET
- Approved: sef@ftp.uucp (Moderator, Sean Eric Fagan)
- Message-ID: <17sp1sINN9jo@ftp.UU.NET>
- References: <17jq00INN7ei@ftp.UU.NET> <17m2e9INNsmh@ftp.UU.NET>
- NNTP-Posting-Host: ftp.uu.net
- X-Submissions: std-unix@uunet.uu.net
-
- Submitted-by: guy@auspex.com (Guy Harris)
-
- >Thus the same programming
- >environment will exist on either system regardless of whether you developed to
- >SVID or AES.
-
- Well, the SVID and AES environments will exist on both systems; however,
- an application may well want to use interfaces *not* in the SVID or AES.
-
- For example, SVR4's run-time dynamic linking interface ("dlopen()",
- "dlsym()", etc.) wasn't in the SVID, last time I looked (and yes, I
- *did* look in the Third Edition, the SVR4 SVID). Dunno if OSF/1's
- run-time dynamic linking interface is there or not (I believe it has
- one).
-
- (NOTE: "run-time dynamic linking interface", with the emphasis on
- "run-time". This doesn't say anything about SVR4's or OSF/1's shared
- library mechanisms, just about SVR4's and OSF/1's mechanisms to allow an
- application to make calls to attach a lump of code, by file name, at run
- time, with the file name possibly generated at run time, and to find
- routines in that attached lump of code by name, with the routine name
- possibly generated at run time.)
-
-
- Volume-Number: Volume 29, Number 18
-