According to Neal Richter: > My previous post with the proposed schedule could be restated with the > releases as 3.2beta4, 3.2beta5, 3.2beta6 etc. > > I guess it comes down to that I think the code is good enough now to > consider a release in the near-term without a raft of changes/improvements. > > The need for a release PDQ is a result of Gilles level of frustration > with 3.1.x.
Just to clarify my own position, these are the things I'm finding frustrating: 1) Having to repeatedly tell people not to use the 1 1/2 year old 3.2.0b3 release because it's too buggy. You can't blame them for doing this - it was the last actual release of 3.2. We need to get 3.2.0b4 out soon, if only to give b3 a proper burial. 2) Too many questions/complaints about database errors in 3.2 betas. We need something more solid, whether based on the newer mifluz or on a zlib compression retro-fit. This has to be the default behaviour - we can't put another beta out with the current buggy word db compression code. 3) My own lack of time in being able to get the 3.1.6 fixes/updates forward ported to 3.2. I'd be thrilled if someone else picked up the ball on this one, but since pretty much everyone sees me as the 3.1 guy (my own fault for that), I feel the expectation is that I should be the one to do this. Having said that, I also don't want to rush a new release out the door if it's going to mean a whole bunch of new bugs to deal with. But we have to get something happening. I don't want us to stop putting out solid releases for either the sake of ideology (as some members seem to be willing to do), nor for the sake of trying too many new things all at once. > I also think a case could be made for a release with some of the things > on your list along with the zlib-WordDB-compression and a improved > inverted index representation in the WordDB to cut out the excessive number of > rows in the WordDB. > > If we accomplish that, then it gets some of the pressure off to merge > with Mifluz 0.23 to fix bugs. The combination of the two would offset any > WordDB size increase penalty from using zlib page-compression. > > If a short-term need for a release isn't warranted, then as long as we > stagger some of these features into a schedule by priority... it sounds good. > > Let's just get a schedule of deliverables for either a sequence of > 3.2betaX or a sequence of releases. > > For task organization and morale this could be useful. > > So Gilles, is there a short-term need for a release without some of the > larger things on the TODO list? Well, I would dearly love to see 3.2.0b4 out the door in 2-3 months, but frankly I don't see that happening with the latest mifluz code merged in. I have concerns about its portability and dependence on yet another library (iconv). I think Neal's idea of the zlib-WordDB-compression retrofit has merit, if only to get an interim beta 4 out the door soon. I see it as a quicker solution to the reliability issue. The only other thing I see as essential for 3.2.0b4 is getting the 3.1.6 changes in there. Otherwise, there'll be too much confusion about features that have been in 3.1 for almost a year, but not in 3.2. Oh, and documentation updates, of course. Ideally, if we could get 3.1.7 and 3.2.0b4 released in close proximity of each other, and with all 3.1.7 fixes also in 3.2.0b4, then we could feel reasonably confident in saying 3.1.7 is the end of the line for 3.1, and 3.2 is getting solid enough for production use. With that, I think we'd probably cut a quarter to a third of the repeat questions on the lists, that I attribute to a lag in getting new releases out. Other side projects like defaults.xml are great, but this seems to be shaping up to be a much bigger task that originally envisioned, what with the idea of maintaining multiple translations. It's great, but it shouldn't hold up 3.2.0b4, nor the much needed corrections/additions to defaults.cc's documentation fields. -- Gilles R. Detillieux E-mail: <[EMAIL PROTECTED]> Spinal Cord Research Centre WWW: http://www.scrc.umanitoba.ca/ Dept. Physiology, U. of Manitoba Winnipeg, MB R3E 3J7 (Canada) ------------------------------------------------------- This sf.net email is sponsored by: viaVerio will pay you up to $1,000 for every account that you consolidate with us. http://ad.doubleclick.net/clk;4749864;7604308;v? http://www.viaverio.com/consolidator/osdn.cfm _______________________________________________ htdig-dev mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/htdig-dev
