Le 06/11/2015 14:55, Gilles a écrit : > Hi Luc. > > On Fri, 06 Nov 2015 10:04:23 +0100, luc wrote: >> Le 2015-11-06 02:34, Gilles a écrit : >>> On Thu, 5 Nov 2015 15:41:57 -0700, Phil Steitz wrote: >>>> On 11/5/15 1:58 PM, Luc Maisonobe wrote: >>>>> Le 05/11/2015 12:25, Gilles a écrit : >>>>>> Hello. >>>>>> On Wed, 04 Nov 2015 10:13:00 +0100, luc wrote: >>>>>>> Hi all, >>>>>>> I would like to release 3.6 in the upcoming weeks. >>>>>>> There have been a bunch of bug fixes and a few evolutions that are >>>>>>> important to me. >>>>>> s/3.6/4.0/ >>>>>> And the statement is still true. >>>>>> >>>>>>> [...] >>>>>>> Of course, once 3.6 is out the MATH_3_X branch will remain alive and >>>>>>> we could also release other versions later on in the 3.x series. >>>>>> Should we worry that the next major release will be endlessly >>>>>> delayed? >>>>> I think we are really quite far from releasing 4.0 as in many packages >>>>> we did not even start refactoring API. Optimization is clearly the >>>>> most >>>>> well known example, but there are also other things waiting in the >>>>> pipe >>>>> for geometry and ode. >>> Is there any specific target for 4.0? >> >> Yes, at least having changed public API. > > There are two parts to that task: delete the deprecated code, and provide > the same (or better) functionality with new code.
Not only that. When an interface intended to be implemented by users is changed (say FirstOrderDifferentialEquations for example, then as soon as a user update to the version after the change, all its implementations have to be changed at once. > I thought that the "delete" updates had been performed by Thomas. > Is there anything else that must go, but still is in master, and is not > deprecated in MATH_3_X? Yes, at least in ode (interface changes) and geometry (semantic changes). > If so, a JIRA "task" issue should indicate what to do, and this should be > resolved before release (IMHO). OK for a JIRA task, but we cannot deprecate the user interfaces involved. > >>> Could we judge how far we are from releasing a major release with the >>> same arguments which you used for 3.6 (many additions, bug fixes, >>> things someone would like to use, etc.)? >>> 4.0 does not need to be perfect. [3.0 was supposed to be perfect ;-).] >>> 4.1 will be! Or 4.2, or 5.0... >> >> 4.x for x > 1 will have exactly the same problems as 4.0 API-wise since >> once 4.0 is out the API will be fixed. > > Yes, but I mean that > 1. we should list all agreed changes even if they are not implemented > (like for "optim"), > 2. then, the scene is set for 4.0 (and the release could indeed miss > many bits, for which users will need 3.6), > 2. we can add classes and methods in 4.x (x > 0), and where not > possible (Java "interface"), then we'll go on and create 5.x. > [As someone said, plenty of integers left...] With Phil proposal to use an even/odd scheme, things become simpler. > > "Release early, release often". > We still do neither. > Despite several developers agreeing on the principle, I'm being put in > the position to argue against "Release later"... > >>> Let's not forget that we agreed to work towards 4.0, and that the 3.x >>> branch was an afterthought. >> >> I agree and was slightly reluctant to continue on 3.X at that time. >> Deciding >> to still keep this branch was indeed a good idea. I properly address >> the problem >> that we did not find time to work on 4.0 as we wanted. > > I've mixed feelings about MATH_3_X. I now realize that I had unconsciously > imagined that it would be short-lived (so the few new features introduced > during that period could indeed be routinely back-ported). > > Providing longer-term bug-fix support is a good thing. > Back-porting everything is IMO definitely a bad idea. > >>> Since we now effectively maintain two versions of CM, it makes sense >>> to allow people to benefit from the extra work. >> >> Yes, but our own overzealous rule about compatibility (I can take the >> blame here, I am guilty for this) induces that we change API only when >> major versions are published. We do have the opportunity to do this for >> 4.0, lets use it at least and not again postpone needed changes. Our 3.X >> API sucks in many places and we know it. > > 4.0 will be pristine (w.r.t our current view on the design): no ugly API > since all that suck should have been deleted! Yes, but as explained above, we need to provide also support for the new user interfaces. Typically if FirstOrderDifferentialEquations as well as two or three other user interfaces change in ode, all our ode code should support all these changes as soon as 4.0. > >>> My viewpoint is that we can have releases from both branches, making >>> it clear what is old/deprecated/broken in 3.x and what is still >>> missing in 4.y. >> >> If it were only missing features, that can be added, I would agree. >> However >> some of the changes are really modifications of existing interfaces. > > Fine, if this is in the pipe... > But I'd insist that we clearly separate the issues of "complete" > refactoring > from other modifications. > E.g. "optim" must be refactored, and that can be done in 4.x (x > 0) > because we (re)start from scratch. With even/odd scheme, yes. > >>>> I agree on this. One thing I forgot to mention above is I think we >>>> may have a few places in 3.x optimization where every way to do >>>> something is deprecated. I suggest we take a careful look and >>>> undeprecate where necessary to make 3.6 usable without warnings. I >>>> may be wrong since I don't use that code myself; but I think its >>>> worth taking a careful look and considering removal of some >>>> deprecations. >>> I'm against this. And is why I started the sure-to-be-controversial >>> discussion on 4.0. >> >> I also don't really like the idea of undeprecating these things. >> >>> We agreed that things were (relatively) broken, and we agreed on a >>> way forward (fluent API, refactoring of the "optim" classes); yet >>> things don't move, because there is no urge to fix them since new >>> features can be back-ported indefinitely to the 3.x series. >> >> No, things don't move because I didn't find time. I am really, >> really busy doing lots of different stuff. I am also really, really >> aware this API should be improved and fluent API is still a way I would >> like to explore for this. > > I did not blame you, and certainly would not want to. > I see it as a "community" problem. Really. > We all decide on something but > * there are areas in the code repository where it is extremely difficult > to make disruptive changes (e.g. "stat", "exception") > * there are areas where only one person performs very thorough changes > (e.g. "geometry", "neuralnet") > * there are areas left in disarray waiting for one volunteer to do all > the work whereas the solution was promoted by a majority (e.g. "optim"). > >> And no, I am not sure this will work and >> 4.x will see the end of these problems. > > We cannot be sure, nor should that fact prevent us from going forward > (i.e. "release early"). Agreed. > >>> Undeprecating what we agreed should be deprecated would only >>> reinforce that feeling, and certainly won't attract attention that >>> we need help to make progress. >>> [And, in addition, 3.x is tied to old Java5 (known tune)...] >>> In summary, I think that new features should only go to the master >>> branch, while only bug-fixes would be back-ported to MATH_3_X. >>> Thus everybody can have the best of both, while reducing the >>> amount of work for the developers. Continuing in this way, and >>> we'll soon have to also "forward-port" bugs reported against the >>> 3.x series. :-/ >> >> Hey, I already do that! > > Right, indeed. > I was stressing that it'll not be getting any better, unless we > stop feeding the MATH_3_X branch. > > People who want _new_ features should go forward with us. Users that > are content with existing features will remain happy even if the 3.x > series only contains bug-fixes. I do need some new features for libraries that do use Apache Commons Math. However, since I need them now, I implement them in both 3.X which can be available in a short time frame and in 4.0 for later. I cannot wait for 4.0 for some of these features. > > The current state of matters did not bring any new contributors, only > more work for those already here. Hey, we have Otmar here now. > >> The following one-liner is my new >> favorite: >> >> git diff -p MATH_3_X~1 MATH_3_X | sed 's,math3,math4,g' | patch -p1 >> >> Yes, it is cumbersome. > > It's as simple as it can be (thanks for the tip!) but, multiplied by the > number of commits, it's nevertheless becoming a severe burden. > > To summarize, I'm not actually asking that the 3.6 and 4.0 releases be > synchronized! > I propose that > 1. we start thinking of each as providing a different service (bug-fixes > vs features) to the users of CM, and > 2. that we work together on a roadmap that will set a target date for the > 4.0 release. Let's see if we can go for the even/odd scheme. best regards, Luc > > > Best regards, > Gilles > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > For additional commands, e-mail: dev-h...@commons.apache.org > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
