[Scummvm-devel] Bring testers to the team?

[Travis Howell]

Walter van Niftrik wrote:
> Travis Howell wrote:
>> Unsupported games (ie SCI games) are a different issue, as we currently 
>> don't accept bug reports until games are officially supported. In the 
>> case of SCI games, it might be worth adding sticking topic on forums, 
>> for people to mention regressions noticed, until we accept bug reports 
>> for SCI games.
>>   
> 
> Well, the point is that a lot of work in this area could be done by 
> people without programming skills. And instead of taking a passive 
> approach like waiting for reports to come in, we could have people on 
> the team that are actively monitoring stability and compatibility. These 
> people could do testing with the event recorder, monitor user input on 
> the forums and in bug reports. They could also help us track down the 
> exact revision for regressions, etc. I think this could increase the 
> quality of the software and at the same time lighten the burden on our 
> developers so that they can spend more time developing.
> 
> Maybe we can't find any people willing to help out, but I think it's 
> worth a shot...

It sounds like you don't know exactly what you do want. If you want
feedback on the SCI game engine, they provide a means for users to
reports problems.

Monitoring stability and compatibility? the event recorder still isn't
complete, so that means manual play testing of games at this point. Game
engines aren't always in a good state to have frequent play testing
done. Frequent play testing of games, would definitely make people get
tried of playing the same old games again and again too.

Monitor user input? how exactly will that help? the people with bugs or
issues still need to report their specific details, and bugs still need
to actually be looked into and fixed by someone with more knowledge
about the relevant code.

Even tracking down regressions would at least require more technical
knowledge too. At least how to use SVN (to check out code), compile
source code, and compare changes between revisions. Someone with some
knowledge about programming, would usually be required to track down the
exact code segment causing the regression.

>> It already it possible to join the ScummVM project, to help with many 
>> non-programming tasks.
> 
> It is? Is this mentioned anywhere? I'm not talking about people getting 
> wiki accounts, I'm talking about full team members.

Yes, although it isn't mentioned anywhere (is there really a need to? 
with an open source project).  If someone offers to work on a specific 
task for ScummVM, or starts working on specific task for ScummVM, they 
are usually accepted into the project. Especially if work is provided, 
to show they are genuinely interested in that task.