[News.eclipse.foundation] Re: E4 / SWT 4.0

["Boris Bokowski" <Boris_Bokowski@xxxxxxxxxx>]

Comments below...

"Eric Rizzo" <eclipse-news@xxxxxxxxxxxx> wrote in message 
news:g2bvif$ctu$1@xxxxxxxxxxxxxxxxxxxx
> Boris Bokowski wrote:
>> But honestly, things are changing. Looking at the list of active 
>> committers for the Platform project, and more specifically Platform UI 
>> (JFace, Workbench, IDE), there are six active committers who work for IBM 
>> and two who don't work for IBM. With Francis joining our team, we will 
>> have six vs. three, or 66% IBMers and 33% non-IBMers, which I think is 
>> quite good given that we are talking about a mature code base that has 
>> been developed by only IBMers over many years.
>>
>> Just to be clear, this is real participation in a core piece of Eclipse. 
>> We are certainly not as diverse as for example CDT, but if you compare 
>> with other projects at Eclipse that started with a contribution from a 
>> single company, I believe that we are opening this up pretty 
>> successfully.  (At least that is how I like to think about it.)
>
> That is indeed very good and encouraging news. However, is it just an 
> anomaly (as Ed's diligence in fixing EMF bug reports is), or is there more 
> evidence of that happening across the board in other Eclipse projects?

I am seeing it in other projects under the Eclipse top level project, 
certainly in PDE for example. I don't know much about how other projects at 
Eclipse deal with their contributors.  I have filed bugs against other 
Eclipse projects myself and was surprised how little reaction I saw for what 
I thought were "easy win" kind of bugs.

My hope is that other projects at least follow the Platform laggards ;-) 
when it comes to inviting participation and becoming more open for outside 
contributors and committers.

>> So in a nutshell, participation is possible and real, and it might even 
>> be rewarding. Talk to e.g. Tom Schindl, Matthew Hall, or Francis Upton if 
>> you want to find out more about how they became committers and why they 
>> are doing it.
>
> Can we extrapolate the behaviors, actions, and results of the truly 
> exceptional to the general population? I think not. I think Eclipse 
> projects should make it easier for the average community member to 
> contribute more, even for those who are not as dedicated, talented, and 
> tenacious as Tom, Matthew, and Francis.

I think I argued on my blog that at least for code contributions, we really 
need talented contributors because there are just so many things to consider 
and so many rules to follow.  But I don't think this is a sign that there is 
something fundamentally wrong with Eclipse - I am sure that to become a 
committer on a core part of Apache, or Firefox, or Linux, you need to be 
just as dedicated, talented, and tenacious.

Could it be good enough to attract people like Tom, Matthew, and Francis? We 
also had Dave and Brad working on Platform UI, and they were pretty good 
too. I don't think these are the only five in the world who would be up to 
the task. Maybe what we need is better ways to attract and reward the really 
good ones as opposed to the "general population"?  I am skeptical if we will 
ever be able to use a Wikipedia-like model where everybody can contribute 
what they want and somehow the wheat is separated from the chaff over time.

Now coming back to e4, one of the goals of it is to get rid of all the 
accidental complexity so that it will become easier to contribute to it. 
Just don't expect that this will get rid of all the complexity we have. 
Making things perform well, report progress accurately, deal with error 
cases properly, avoid deadlocks, or making it easy to use and accessible to 
all etc. will always be hard. Not to mention the rules about IP cleanliness, 
process considerations, and API compatibility, each of which add another 
dimension to the problem of making it easy to contribute.

Boris