[zile-devel] Zile 2.2alpha1 imminent, 2.2 less so; new project Zee

[Reuben Thomas]

As you will have noticed, I have already made a 2.1.0 development 
release of Zile. For some reason I have been almost addicted to working 
on Zile since Christmas, and I have progressed faster than I hoped. I I 
am now nearly feature complete for a 2.2 alpha release, barring only a 
simple garbage collector for the new Lisp interpreter. I hope to make a 
2.2alpha1 release in the next couple of days.

However, I have then decided that I want to diverge from Zile's aims; in 
particular, I want to remove a lot of features that I don't use, and 
gradually stop trying to emulate Emacs. Don't panic! I have started a 
new SourceForge project, Zee. I hope to take the Zile code base in new 
and interesting directions, and use it for experiments in programming as 
well as in editing.

This means that I will not be spending much more time on Zile, and in 
particular, that I do not currently plan to work towards a 2.2 release. 
If there is interest in the 2.2alpha1 release, then I will probably be 
happy to fix bugs and make further releases, culminating in a stable 
release if I'm happy with the quality. Of course, if someone else would 
like to work on Zile, I'm happy to let them do so. In particular, the 
2.2 code base is in some ways rather dirty: in particular, there is 
duplication between Zile's own data structures and routines and those in 
the new Lisp interpreter. I will be working hard on fixing this in Zee, 
but I don't see any way that I can feed this work back into Zile without 
doubling the amount of effort I have to put in.

In fact, having a gap in stable releases now might not be a bad thing. 
Zile 2.0 will be in Debian sarge, and I suspect that many of Zile's 
users are Debian users, and there won't be an opportunity for another 
stable release to get into Debian for a while anyway. Secondly, Zile's 
last stable release was two years ago, but I don't think that is a big 
problem. Features like UTF-8 support will eventually be demanded, but a 
lot of users still don't need this.

I've had a great time with Zile. In particular I have learnt that I can 
program much better when I don't have to start from scratch, and I have 
learnt how to proceed in really small steps, as encouraged by the 
principles of Extreme Programming. I have really valued working with a 
code base that is already useful and that I use all the time, so that 
when I make changes, I quickly find out if they work, and whether they 
are useful. It was this combination of factors that gave me the idea of 
using the code base to pursue more experimental ideas, knowing that I 
will be using the results at every stage, rather than having to spend a 
long time before I have something I can actually use.

In conclusion, I'm not saying goodbye, even if I'm scaling back my 
involvement in Zile. In particular, I'm happy to continue bug-fixing 2.0 
(and once it comes out, 2.2) for now. Most of the bugs found will, at 
least to start with, also be bugs in Zee, after all! I hope you will 
take a look at Zee as it develops away from Zile, and tell me what you 
think. Most of the ideas that I outlined for Zile versions after 2.2 
(or, in my plans as I sent them to the mailing list, after 2.1) are 
still on my wish list; indeed, the current Zile Feature Request Tracker 
will be important to initial Zee development.

thanks once again,

Reuben

--
http://rrt.sc3d.org/ | wit, n.  educated insolence (Aristotle)