Michael Gerz wrote: > I looked up the meaning of "endorse" and I must confess that I am sorry > that you cannot endorse a product on which you have spent sooo much > spare time. > > Apart from the Armenian support and the recent InsetListings discussion, > what exactly is missing to make LyX 1.5.0 a great product - better than > any other release we have published before?
It will no doubt be a great product, but not better than any other release. 1.3.7 was the best release I know. Of course I know that a .0 release can never compete with a .7 release WRT stability, so we should not assert that. 1.5.0 will also transport a certain message to the users: "Hey, we ripped everything apart before 1.4.0, put together some of the most important things again for 1.4.0, put together some more for 1.5.0, added many new and exciting features for 1.5.0, but the rest will stay ripped apart forever and nobody cares". I don't want to be connected to such a message in any way, and therefore I wrote that I cannot endorse 1.5.0. Don't get me wrong, I agree that this ripping apart was necessary, but IMO it is also necessary to put things back together again. And since until now nobody was interested in doing so the chance is very low that somebody will become interested after 1.5.0. > And a software that eases > _your_ life? I don't think I will switch to 1.5 for my thesis. I somehow adjusted to the quirks of 1.4, put some additional stuff that eases my life in my personal branch, and I prefer living with the known problems over the risk of unknown ones. Some of them (math speed problems) are fixed by brute force: I have a fast machine. By the time I would switch (1.5.2 or so) I am hopefully finished. > Could you point us to some bug reports that you consider > critical for the success of LyX 1.5.0? I guess, the whole LyX community > is eager to make you happy! The bugs that annoy me most are - unusable scrollbar for long insets - the math regressions WRT 1.3: wrong font for unknown commands, unlogical cursor movement, wrong color for some symbols - 1973 - 2555 - 2556 (although I got used to the workaround with the space) - 3291 There might be some more that I don't remember currently, but that means that they are probably not too important. And then look at the very recent marginal bug: AFAIK (correct me if I am wrong) this happened because of missing review of the insetName -> name change. A simple and safe looking change caused a serious bug. I am convinced that more bugs of this sort will be found after 1.5.0 is released, because the practice of committing without review is followed since some time. > As you know, I promised to give some gifts to the most busy bees of the > LyX 1.5.0 development cycle. No matter happens, a prizes will go to you > and Abdel (third prize is still vacant). Thank you for that. Georg
