Re: slrn 1.0 [2007-08-25]?
[email protected] (Bernd Nawothnig)
| Newsgroups | gmane.network.slrn.user |
|---|---|
| Message-ID | <20070720133904.GB6586@miraculix> |
On Wed, Jul 18, 2007 at 11:27:14PM +0100, Peter J Ross wrote: >>> Does it make sense to make iconv a "must" dependency and to remove >>> all "have no iconv" code? >> Definitely yes. That would ease maintenance. >> Better to drop the possibility to port slrn to architectures which >> don't come along with iconv (and which are not that very important >> for the existing slrn-community in my eyes) and gain stability and >> active developement in exchange for that. > This would be a big change from the history of slrn (and s-lang). One > of slrn's distinctive features is its exceptional portability. It > would be sad if VMS support were lost. Sure. But how many users would be affected and how much work will it be to maintain that branch for non-iconv-systems in the future? The amount of time, the maintainers and developers have, is limited. That is a simple fact. >> There are only few developers actively involved in slrn and one >> should face that simple fact together with it's implications. The >> iconv thing together with real utf-8 support *is* what is urgently >> needed in the stable branch. And Developement would be easier without >> the need to look at 1001 #ifdef possibilities which only a few people >> in the world do really need or use. > I certainly can't see any reason to support platforms that are no > longer being developed. For that the support for those platforms (MS-DOS, BEOS, etc) should end. >> Version 1.0 is a good opportunity to drop slang 1.* and non-iconv >> support. The mourning would surely come but it would be short ;-) > There's no reason not to use slang 2 if it's available, and as far as > I know it's available to everybody. Yes. > Currently, my efforts to suggest improvements to the documentation > have been delayed by my ignorance of SGML, but I'm working on it. > Meanwhile, here's yet another thought: can somebody be found to > compile a Windows binary? Cygwin should be sufficient and more easy. Native Windows support is not necessary (nor desirable) since Cygwin is available for everybody. Bernd ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/