[PLEASE VOTE] so what should we be working on

[Robert Helling]

Hi Dirk, hi others!

> On 28.05.2016, at 23:37, Dirk Hohndel <dirk at hohndel.org> wrote:
> 
> I have been focused on other things for a while (and the why and the what will become public fairly soon) and decided to use that chance to figure out what would happen if I just stopped paying attention here for a while.

I can understand your frustration but I would not paint it as black as you seem to do. I agree that it’s sad that a large number of people that have contributed a lot in the past seem to have decoupled for a least the better part of a year and except for you (and the kirigami people) pushing the mobile version.

My personal excuse for doing so little recently (but as I tried to explain a while ago on IRC I would not say I do nothing but I have only very little to show) is that all my free time got soaked up by other things in particular job and family but I still hope that better times are still ahead. In no way I have given up contributing and strongly plan to do so again in the (hopefully near) future.

> 
> Here's what I think I learned (correct me if I'm wrong)
> 
> a) no one cares about Subsurface-mobile. No bug reports, no code, no suggestions, nothing. So I guess we should call this a failed attempt and abandon it - screw the 500 or so people who use it.
> 
> b) Miika continues to deal with any import bug that shows up. Awesome. Your patches have been pushed
> 
> c) there continues to be a slow crawl of features, ideas, patches for the planner. Those have also been pushed (someone poke me if I got something wrong there)
> 
> d) no one cares about the dive site mess that we have. The current version in Subsurface 4.5.x is pretty much unusable and makes no bleeping sense.  No patches, some suggestions for improvement, no discussion, nothing.
> 
> 
> In summary, this project is pretty much dead. I guess it does everything the formerly active developers wanted? Or everyone has moved on to more fun things? Or had children, changed jobs, built a house, lost their job, or one of a number of other life events?

My analysis would be: In the current version, subsurface desktop in its current incarnation does many things well (and often much better than other programs that I have seen). So there is very little pressure to fix things (and as  far as I can tell responses to bug reports on the mailing list are still quick and stuff that has a chance to be fixed usually is within 24h). The location thing might be a mess, but it works for me so I feel little itch there. But I should also say that I did not find much time for diving either (I haven’t been in the water in 2016, yet) so I did not have a lot of chances for real world use of subsurface.

The mobile version is a little different, but only a little: First of all, for what it is designed for (read access to your log book on the road) it already works very well. It might have some visual glitches but unfortunately, my experience of attacking those has been quite frustrating. I guess that this is mainly due to the fact, that I have long breaks between having goes at the mobile version which implies that just to get started an get the thing to compile and deploy on my device already takes significant time (before any line of code is added) and then (as you know) debugging QML is let’s put it like that, challenging.

Last time I tracked down one of the problems on my IPhone 4s (overlapping labels) I found a way to resolve that but that might have other implications so I asked on IRC (on May 9th) but never heard from anybody. I could probably still send a patch (but would have to remind myself what the actual issue was).

Another thing I tried was to address some of the many many warnings/error messages that I get on the command line when running subsurface-mobile. But I never got very far since I always ended up deep in kirigami code without proper understanding what was going on and what what was supposed to go on. Mainly the latter, my lack of understanding of “how things should work together” is what I blame on not being able to make any progress. But that on the other hand requires more investment into reading and understanding stuff which is only possible with larger chunks of time available (which was not the case in the past but is hopefully in the future). And of course, these things get much easier when there are other people around doing the same where one can get answers to questions or just steal ideas from code. I have not given up here, it’s just that progress is slow.

Still, I would see the mobile version to be quite close to a complete V 1.0. To me there is no obvious thing that is missing from the design goal. That is to say: Dirk (and others), you already did an amazing job! And I wish I had contributed more.

There is of course the issue with reading out dive computers using the mobile version. But (as I said in the past), I thing with the help of a CHIP (or other for that matter), we are not very far from that goal (but of course still farer than the rest of subsurface mobile) but I will address that in a separate mail. With that, in my understanding, it would double the usefulness of the mobile version.

> 
> 
> So what should we be working on...? Should we...
> 
> (1) just shut this down and move on?
> (2) move things to maintenance mode, abandon Subsurface-mobile, fix bugs in Subsurface whenever we find time but otherwise declare victory? It won't be too painful to track libdicecomputer and keep making 4.5.x releases for a while, I guess...
> (3) focus on Subsurface-mobile, release the iOS version, update the Android version and see if there is a single person besides me who is willing to invest time into that?
> (4) focus on Subsurface 4.6, fix the dive site management and create a list of prioritized features that we want to get in place?
> (5) focus on Subsurface 5.0, write a completely new UI and abandon what we have in 4.5?
> 
> If your vote is for 3, 4, or 5 I assume that you are volunteering to carry some of the work that is needed to get there.
> If you don't vote, I will count that as a vote for 1.


There is little point to answer here, since the vast majority of the 150(?) people on this list are lurkers. So 1) has immediately won.

But seriously.n Given how feature complete the desktop version is, I see no need to rush things. Yes, we are progressing at much slower pace than before, but I would not see that as zero. I would keep going and hope for the fact that we will gain more momentum again (in the summer perhaps). Until then, it would be a mixture between (2) and (4) (and possibly (3)). Just keep going. I don’t know what “abandoning” would actually mean. Not accepting patches certainly not. Maybe no promise of support. Which would be sad.

A new UI is certainly very low on my list of priorities but that should not stop others I would say. But if anybody asks me where they should spend their time it wouldn’t be there.

I must say, regarding dive site management, I am not even sure what the problem is. Could somebody please spell that out? Not in a “it would be nice to have feature X” or “we should be doing stuff completely different” but rather as “currently Y does not work as it is supposed to”. And if somebody has more energy to do these things they could have more ambitious goals. at our local hacker space we have a simple rule “wer macht hat Recht” - who does (implements things) is right. So rather discussing competing designs to death people should send patches. And then, criticism should again come in form of patches rather than “I don’t like this”. I would expect that would be much more productive. And if somebody wants to attack really big goals, I would love a way to share dive site information with others to build up something like “the database of inland dive sites in southern Germany” for example.

Please don’t give up on the mobile version. I currently spend my time on getting the CHIP thing to work and consider that part of the mobile version.

I just had a look at the trac (sorted by issue number). Many of the bugs of the mobile version are not obviously bugs in the subsurface code. Are you aware of other things there that need to be done (which might be shallower entry points) and add those to trac?


I realize that “just keep going” and keeping the status quo also has a lot of costs (not only financially) mainly on Dirk’s side. Off the top of my head are

1) Running the git server, accepting and pushing patches

2) Running (and keeping up to date) the build infrastructure

3) Running and maintaining the web site (and Facebook, google groups etc)

4) Running and maintaining the cloud server (including password resets etc)

5) Running other infrastructure like trac, mailing list etc etc

Maybe it makes your life easier if you share some of those burdens (like sysadmin things) with others? Are there other ways to help you keep going?

Finally a note to many others: “I cannot code” is not an excuse. When I came here, I had not written a line of C code in several years, had never coded C++, had not written any GUI stuff since the days of my Atari ST (in the 1980s). None of these things are rocket science. Many improvements can be simply done by investing a bit of time, looking at the existing code, doing some googling for documentation and stack overflow pages and then chasing a few lines of code (along the lines of what is already there. Please please give it a try!

And the others: Don’t give up. Please accept a slower pace and don’t confuse it with 0.

Best
Robert

--
.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oOo.oO
Robert C. Helling     Elite Master Course Theoretical and Mathematical Physics
                      Scientific Coordinator
                      Ludwig Maximilians Universitaet Muenchen, Dept. Physik
                      Phone: +49 89 2180-4523  Theresienstr. 39, rm. B339
                      http://www.atdotde.de

Enhance your privacy, use cryptography! My PGP keys have fingerprints
A9D1 A01D 13A5 31FA 6515  BB44 0820 367C 36BC 0C1D    and
DCED 37B6 251C 7861 270D  5613 95C7 9D32 9A8D 9B8F





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20160530/1e21f3da/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20160530/1e21f3da/attachment-0001.sig>