On Thu, 04 Jan 2018 11:49:03 +0000 Andrew Williams <[email protected]> said:
Originally T5301 work was intended to be done a long time ago. But there was little interest in doing it. Then some interest, then that interest vanished. Now it's back. I have no idea if it'll vanish again. I won't be betting any timelines on it. As I mentioned - We can do the timed releases orthogonally to this EO work. A version number is an arbitrary goal. We could do an efl 1.21 "now" (well take a few weeks out to do the stability work/testing/QA etc.), irrespective of the T5301 work (other than that work having to go on hold for a release cycle). The version to ship the T5301 work has slipped before and will likely slip again (we talked about 1.20 and I think 1.19 too). I don't see any value in tying it down to a specific version number, as it'll almost definitely be an incorrect piece of information. 1.21 is a goal. An aspiration. Not a "make or break" deal. It may end up in 1.22 or 1.23 ... depending how many interim releases go out. If we should put out such releases in between is a separate discussion that has more to do with there being enough fixes and improvements to roll out such a release (e.g. in the wayland work). Your pessimistic extrapolation of time to go I think is incorrect as more people became involved in T5301 in the past ~3-4 months compared to before, with the optimistic one being closer to reality (well we'd all love to believe that too :)), thus there is a lot of work underway at the moment. If people stay involved - as above, is unknown. So I don't think it's defined if 1.21 will or will not include "Efl interfaces stable/usable". It may come out if there is enough call for it to come out to justify the pause. 1.21 doesn't have to release any interface work at all... it can be in the tree but not "released and stable". I.e. behind BETA defines etc. etc. Why does the release plan document not reflect this? Ask Stefan. But would it make a huge difference if it did or did not? In what way? To who EXACTLY? Since if we were to do a 1.21 soon it certainly would have zero eo/interfaces work available (without a #define)... it'd just be a delayed regular release. So I think I might modify your questions to this: 1. Should we doe a 1.21 anyway (and have no Eo/Interfaces stable in it)? (is it worth the delay?) 2. We should reflect the decision to 31 as an update to the release plan. ? > Hi, > > Our release page (https://phab.enlightenment.org/w/release_roadmap/) still > says that we are doing time based releases > > "We aim for a 3 months release cycle (with a release on the first Monday in > February, May, August and November)" > and "Release 1.21 is planned for late 2017". > > If we are now looking at a feature based release then > *) Why does our release summary document not reflect this > *) Does this mean there will be no 1.21 until that > https://phab.enlightenment.org/T5301 is completed? > > If we are waiting for T5301 then here are some statistics: > > * The parent ticket was created 9 months ago and has 92 child tickets > * 23 of the child tickets are closed (that's 25%) > * On average 2 new tickets are added every month > * 24 of the tickets are lower than "High" priority > > Assuming that high and above is what is required then we've completed 1/3 > of the tickets. > A very crude calculation would show an average 2 added and 2.5 closed per > month could extrapolate to many years of work remaining. > If instead we assume that most tickets are partly done and that the new > tickets are smaller it still points to 6 - 18 months work remaining > > With this in mind I think that 1.21 probably needs to be either a subset of > the interface work or somehow revised. > I think whatever we do intend to do we should update the release page so > that the intentions are clear. > I pulled this together to help understand our current plan, not to complain > or point fingers - I hope these thoughts are taken in that spirit. > > Let's figure what makes sense for all the expected recipients of our next > major release :) > > Andrew > > On Thu, 4 Jan 2018 at 06:32 Carsten Haitzler <[email protected]> wrote: > > > On Wed, 3 Jan 2018 21:09:17 -0500 Michael Blumenkrantz > > <[email protected]> said: > > > > > On Sat, 23 Dec 2017 19:37:31 +0900 > > > Carsten Haitzler <[email protected]> wrote: > > > > > > > On Fri, 22 Dec 2017 19:54:02 +0000 Andrew Williams < > > [email protected]> > > > > said: > > > > > > > > > Hi Vincent, > > > > > > > > > > I would really love this too - in fact I have been pushing for it. I > > > > > suggest regularly that it would be helpful to have a roadmap. On my > > most > > > > > recent request I was told bluntly “too much effort, you are the only > > > > > person that wants it”. > > > > > > > > https://phab.enlightenment.org/T5301. you keep telling me "that's not > > a > > > > roadmap". > > > > > > > > comparison table: > > > > > > > > | | T5301 | GTK+ | > > > > | Has title | X | X | > > > > | Has short description | X | X | > > > > | Has longer (multi-parapgraph) descriptions | Some | Some | > > > > | Has assignment of who is doing it | X | X | > > > > | Has status/priority | X | X | > > > > | Are a changing "document" | X | X | > > > > | Has dependencies | X | | > > > > | Has discussion thread | X | | > > > > > > > > OH look there. it has everything your supposed roadmap has and even > > more! > > > > What were you telling me that this is not a roadmap? Please indicate > > > > clearly how it is not as I have clearly shown above that it has > > everything > > > > you claim a roadmap has (you would want that document linked to) and > > even > > > > then some more. > > > > > > > > > If we can get more support for such a document I would be far > > happier to > > > > > push forward again and see it is pulles together. > > > > > > > > The document exists. Stop saying it doesn't. You just want someone to > > write > > > > it up in a table on a wiki page (which is what the GTK+ one is) > > instead of > > > > as tickets. If you want that - then you do it. And maintain it. > > > > > > Unless I've misread the previous mail, Andy has just said that if people > > are > > > in favor of such a page that he is willing to do the work. I'm confused > > by > > > the apparently hostile demeanor of your reply considering that someone > > has > > > just offered to do the work which was requested? > > > > Andy spent several hours in a private IRC /msg session telling me a roadmap > > document doesn't exist, and I kept pointing him to this, sayning that > > maybe it > > can be fleshed out a little more, but all of the essentials are there, the > > him > > repeating it's not a roadmap, then I see these emails pointing to > > something as > > "oh that's a roadmap". and it is in essence exactly what I spent hours > > pointing > > him to. And the next day or so was this. I was entirely irked. > > > > He never offered to write it until this mail. So you missed the bit where > > he > > kept me up until like 3am or whatever telling me we have no roadmap, never > > actually volunteering to write it. > > > > > That task in phab (which cannot be found from the phab wiki, the main > > site > > > wiki, or a google search) does not constitute a project roadmap of the > > sort > > > that this thread is discussing. A roadmap should be visible and easy to > > > locate, it should be readable by those who are not project insiders, and > > it > > > > That I do agree with - it's kind of hidden without being publicized. > > > > > should be a usable document for people working on related projects to > > judge > > > whether they can/should contribute to the major tasks or if their > > projects > > > will use any of the current/upcoming work items. I'm not interested in > > > nitpicking or being pedantic; no, I didn't reference a dictionary or > > > wikipedia or technical journals to arrive at my definition, this is just > > my > > > expectation upon seeing the roadmap for any project. The cited task > > fulfills > > > none of these criteria in my view. > > > > T5301 does to me represent a roadmap. As much as the GTK one does. It > > doesn't > > have timelines, but then again neither does what is being pointed to as a > > shining example. It has a list of tasks with their status - only thing > > missing > > on the T5301 page is the short 1 or 2 line description. > > > > > Furthermore, although I appreciate the time that you put into creating > > this > > > ascii comparison table to demonstrate the possibility that our ticket is > > > better than their actual roadmap, this table is itself irrelevant > > because the > > > phabricator task has no correlation to release planning and thus is not a > > > roadmap such as the one linked in the original mail. > > > > Actually it is the same thing. As much as what is being referenced. That > > ticket > > was originally intended for "EFl 1.20" and has since been pushed to 1.21. > > it's > > for a release like the "GTK+ 4 checklist" is for a release. Perhaps it > > needs to > > mention that in the top of the ticket. > > > > > I'm in full support of having a roadmap for EFL to try coordinating > > releases > > > around. This would make our supposed time-based release schedule more > > > sensible and less "I think we're maybe still waiting for some feature > > but I > > > don't know what it is or who's working on it?". > > > > Right now our time base releases are not really functioning because we're > > trying to do a FEATURE based release... which is precisely what that > > ticket is > > and what a "GTK+ 4 roadmap" is. Releases are waiting for this. > > > > Time and feature based releases are different dimensions. We could just do > > another time based release now, but none of the EO stuff is usable yet and > > still will be switched off. So they are orthogonal. > > > > > > > Thanks, > > > > > Andy > > > > > On Fri, 22 Dec 2017 at 16:57, Vincent Torri <[email protected] > > > > > > > > wrote: > > > > > > hello > > > > > > > > > > > > i've recently seen this gtk roadmap : > > > > > > https://wiki.gnome.org/Projects/GTK%2B/Roadmap/GTK4 > > > > > > > > > > > > is there the same for EFL ? > > > > > > > > > > > > Vincent > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------------ > > > > > > Check out the vibrant tech community on one of the world's most > > > > > > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > > > > > > _______________________________________________ > > > > > > enlightenment-devel mailing list > > > > > > [email protected] > > > > > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > > > > > > > -- > > > > > http://andywilliams.me > > > > > http://ajwillia.ms > > > > > > > ------------------------------------------------------------------------------ > > > > > Check out the vibrant tech community on one of the world's most > > > > > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > > > > > _______________________________________________ > > > > > enlightenment-devel mailing list > > > > > [email protected] > > > > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > > > > > > > > > > > > > -- > > ------------- Codito, ergo sum - "I code, therefore I am" -------------- > > Carsten Haitzler - [email protected] > > > > > > > > ------------------------------------------------------------------------------ > > Check out the vibrant tech community on one of the world's most > > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > > _______________________________________________ > > enlightenment-devel mailing list > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > -- > http://andywilliams.me > http://ajwillia.ms > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > enlightenment-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- Carsten Haitzler - [email protected] ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
