[Scummvm-devel] What is happening to the ScummVM team?

[Eugene Sandulenko]

OK, now it's my turn. I've addressed some issues in other e-mail, see
the rest below.

On Fri, Feb 13, 2009 at 3:25 AM, Johannes Schickel <lordhoto at gmail.com> wrote:
> First of all I want to take a look at our releases.
[...]
> Now we are nearing our next release: 0.13.0. Seeing this time we know
> already that MM v2 is not completable and some games are again not even
> tested, this includes ITE. Tagging is also near, to be precise in 6
> days. Everyone following -devel also notices there had been silence
> about release talk since some days.
>
> This leads me to the question: Do we learn anything from our experience
> with past released? Actually I am surprised we so far do not have set
> the bugs preventing MM v2 from being completable as release critical
> bugs. Another thing is, seeing how many games have still not been
> tested, that I would personally propose to delay our release a bit to
> get more testing results to sort out possibly other games not being
> completable.
First. We really do learn from our past mistakes. Particularly, 6
months release schedule was announced beforehand as some porters had
big concerns with that. I.e. we had (and unfortunately still have)
some ports delaying badly at each release.

I was hoping to plug that issue with such announcement, but no awail.
Moreover extending release schedule to almost 2 months did not help
either.

As of the quality, unfortunately reality shows that with each release
since 0.8.0, when we introduced testing, we get lesser and lesser
people willing to test. For instance, still nobody tested all Kyra
games, and that engine definitely had big changes recently. Johannes,
you could contribute to the testing too, no offense.

As of MM v2 bug, as I stated in my release status e-mail, I am not
convinced that that is the real issue. AFAIK there were no changes to
the scripts engine of SCUMM, and nobody confirmed that those bugs
occur on English version. Moreover, MM v1 is completable, and it uses
practically same codebase at least for the scripts.

Still, I am really don't like the way you point this out. As I
announce with each release schedule, I do not put it in stone. A
simple request with sound reasons will easily make me push release
date or even cancel it altogether. It is not that I am dictating it,
although certain degree of dictatorship is required from the one
wearing a release manager hat ;).

> Of course we now have a special date for ports to create prerelease
> binaries. That is a step in the right direction. I am currently not sure
> in how far our ports adapted that though.
I saw a delighting reaction from our porters on that. Some of them
took a step ahead and started to publish pre-release builds even
earlier. Thank you very much, guys.

> I also know some might say, "but we should really have a release soon"
> or "we always release every 6 months", but in my opinion we should
> rather have a quality release than a release in a fixed period, which
> has too many annoying bugs. Thus my vote for this release is: wait for
> more testing results! (means delay it)
Although I wholeheartedly agree with you on this, unfortunately
experience shows that in this case it will NEVER happen. The hardcore
users just refuse to replay even beloved games 130-th time. And the
newcomers just don't care.

Thus I would rather release 0.13.0 and then 0.13.1 soon when wider
audience will test it. The reasons why 0.12.1 was released is (a) lack
of time (b) I did not see many complains about ITE not working
(unfortunately).

For the sake of this, and hoping that we will participate in GSoC'09,
I am looking into automated testing task which I recently added to
http://wiki.scummvm.org/index.php/OpenTasks#Implement_recorded_play
Otherwise I see no bright light in the way of quality release testing.

> I'm happy to hear any comments from you about it, might it be: "just
> bullshit, our release testing is great, just test the games yourself" or
> "we might really want to think again about it". I would not be happy to
> hear the usual silence, though. Also please note ITE was purely taken as
> an example here, which sadly had some problems in the last releases.
I simply lack of time these days. When in the past I easily spent
20-30 hours per week on ScummVM when I worked from home, today I get
15 hours per week wasted on my way to the workplace, so I am really
happy when I get even 15 hours to work on my beloved project. And in
that time I have to squeeze lots of e-mailing, work scheduling, commit
reviews and sometimes bugfixing.

So if you like to help the team and step in as a release manager, at
least I will not be against it.

> The next point I wanted to talk about is our engine additions.
The current policy was that as long as there are people interested in
working on the engine, and as long as it is a 2d advenure (preferably
point and click), it just gets a green light. I saw quite rare
interest from people working on engine FOO in the development of
engine BAR. Sometimes I was not even aware of certain engine
development, consider Tucker which just arrived completable.

I personally see no problem with that, since the policy mentioned
above proved to be working in most cases. I care when some engine
demands our core changes, thus touching more people than just the
engine authors. Then I have to diligently evaluate the needed changes.
In other cases I keep in mind that after all, an engine could be just
compiled out.

> I would have really loved to hear more about  what is going
> on with ScummVM development.
Quite often there is another factor which should be considered. Price
of the ScummVM-supported games. As you remember, I tried to do my best
and talked privately to the team members when I was aware that certain
games are about to get supported, so they could obtain copies for them
on eBay at cheaper prices.

However as I mentioned, sometime even I do not know about some engine
coming our way.

> Now another issue I came across today in our IRC channel by chance:
> FreeSCI merge. It seems that the FreeSCI team is preparing for an merge
> into ScummVM. I am pretty astonished that so far nobody of our team
> members involved in the FreeSCI discussion about this and/or planning
> the merge together with them had said any word about it. Actually I
> thought our policy was to post mails to -devel to discuss such changes.
There were NO changes. Everyone would certainly be warned about it.
And the manner of that message would be something like: "Hey! FreeSCI
folks finally agreed to merge, hurray!", not asking opinion or
whatever. Not because I ignore opinions of the team, but because
everyone was informed that SCI engine is on the way, and everyone
greated it. Being it SCI1 or FreeSCI doesn't change anything.

I gave several reasons for not announcing it till now in my another
e-mail. Let me stress that currently I consider all of these
accusations as premature. We did not silently dumped anything into our
repository. Neither we had plans to do so.

> * Are we still one team? Or is it nowadays rather that "the ScummVM
> team" as an umbrella team and various single engine teams?
We always were ScummVM project consisting of several subteams:
  - Core
  - Engine-specific
  - Port-specific
  - Other, such as Wiki editors

Name it as you like. Such structure comes from 2001 when Simon engine
was merged.

I hope that my e-mail was constructive, and in case I offended anyone
by saying or not saying anything, it was unintentional, and I
apologize.


Eugene