Need to do something

Chris Muller chris at funkyobjects.org
Fri Oct 14 16:15:22 UTC 2005


>> Watch me.  I *will* rewrite it over the top of your head and submit it to
>> SqueakMap with a "[ANN] New And Improved Chess Game" to the list.  And the
>> community will love it so much that it will become part of base Squeak. 
What
>> will you do about it?
>
> Go for it. Just as long as you don't expect me to maintain that code. 
> But *if* you want me to maintain that code then I do expect you to 
> respect my ownership of the code and to work with me instead of against me.

Well, of course I wouldn't expect you to.  That was my whole point, every
individual (or unified group) has the freedom to walk whatever code-path they
wish.  They are free to change another owners code, and that owner is free to
support them or not.

The minor divergence of our ideas is I prefer to empower the individual (or
unified group) via the deep-collaboration tool.

Hernan Tylim articulated it beautifully:

> This schema means that Bob will stop sharing things with Peter ? No. If 
> Bob wants the goodies of Peter, and viceversa, they will both work to 
> adapt they images and goodies to be shareable. Here all the work and 
> what needs to be done will be done out of necessity, and by the people 
> who need it.

That last sentence is, in my view, the primary characteristic of the scalable,
self-sustaining harvesting process.

You seem to prefer empowerment of the *owner* so that there will be fewer
"flavors" of a particular framework floating around and instead one single
unified version made only of code that was channeled through the owner.  Energy
must be spent managing "owner-turnover" which is energy that doesn't get spent
on development.

Honestly though, at this time, I concede that your focus is probably more
realistic.  We don't have the deep-collaboration tool I dream of.  We need good
quality right now, not 31 flavors, and unification will help that.

>> This is an example of ownership becoming bottleneck.
>
> Only if you can show that there was actually anything that needed to be 
> done. I'm still waiting for that...

Well, isn't Marcus still waiting for you to "concede" as promised, since he
posted that fix to bring Chess up to work with the new Wide strings?  Sounds
like an indisputable improvement to me that shouldn't have been held up by
owner bottleneck..

I don't recall reading your apology after that, but I am looking forward to it!
  ;-)

>> With Tweak, you make changes to base Smalltalk over the head of Alan and
Dan,
>> right? 
>
> Wrong. So very wrong, you wouldn't believe it. Dan (and John) are the 
> reason why Tweak started from scratch.

I was thinking of the method-annotations for the events.  IIRC, Dan had
questioned you about making the syntax more complex.  So it was a change, at
least initially, for which won his support later..

If I am wrong I apologize, even though I see no harm in changing existing
systems.  Again, it's just a question of whether it affects the owners support
for what you're doiing and if you care about that (which you did, in Tweaks
case).

> As for Alan ... *grinning wildly* you should try to do things over 
> Alan's head sometime. You really should, I'm sure it would be a unique 
> experience ;-)

I sincerely hope to meet the man some day..

Cheers,
  Chris



More information about the Squeak-dev mailing list