Martijn Brouwer wrote: > P.S. When will 1.4 arrive and what will it bring us?
* Not any time soon. * Lots. The last few releases have concentrated on cleaning up the LyX source code rather than on supporting a larger subset of LaTeX. Enhancements have occurred (notably the math editor supports more and more of AMS-LaTeX) but these improvements are a serendipitous by-product of the underlying clean-up. This basic process continues in the 1.4.x release. The difference however, is where this clean-up is now concentrated. Up to and including the release on 1.3.x, we moved code out of the core and into the periphery where it could be cleaned up with minimal impact on the core itself. The 1.4.x development cycle has continued to clean up these peripheries but has also taken the bull by the horns and started the process of cleaning up the core itself. The result will be a _much_ cleaner code base that may even be understandable. And extendable. Needless to say, however, changing the core of the program has introduced many regressions which must be fixed before 1.4.x can be released to the wider world. I believe that we are on the way back up to the surface now (it becomes quite hard to make LyX crash ;-) but we haven't yet started talking about a freeze. Past experience suggests that feature freezes last a considerable length of time (1-2 months) whilst we indulge in a bug-squashing frenzy. So, in summary, there _are_ several new pieces of candy on offer but most of the work continues to be behind the scenes. Is this a good developent strategy? Yes. Experience suggests that the development of open source software ususally founders once the principal developer moves on. Only if the code can be understood by relative newcomers does it have any chance of a long life. (Witness the huge number of dead projects on sourceforge.) We have lost one major developer in the 1.4.x series but have gained two or three new ones. Indeed, I would suggest that most of the feature enhancements this time around can be creditted to them. In turn, this suggests that the basic design of the software is good; the learning curve is not too steep Phew! I've run out of steam. You'll have to wait for another time before finding out what the candy is ;-) -- Angus
