Keith Hodges wrote:
> So you dont like traits much?

Well, I don't know really. I had much hopes for it but the results do 
not seem to match my expectations. BTW, just in case you didn't know: 
this project started as a discussion between Nathanael and myself, and 
to me, although Nathanael solved the harder problem he also solved the 
less interesting problem. I'm still looking for more real world examples 
but to me it seems as if the biggest problem with traits is that they're 
the logical equivalent of C's #include statement. This may sound harsh 
but when you think about it, it is very, very similar. And of course 
that has some uses for sure, but it often creates more problems than it 
solves. In *particular* when it is applied just because it seems like a 
good idea in general without having specific reasons for using it. 
That's mostly why I reacted to your email, because my instant reaction 
was hell, no, you wouldn't want to use traits instead of delegation in 
such a situation unless you have to (same goes for subclassing btw).

> a small question though...
> 
> How come I get a longer and quicker reply in a rant as a result of one 
> sentence in an email to squeak-dev, than I got from two specific 
> requests on Tweak mailing list which was dead for months. The first 
> dated (26 May) asking for a current update and strategic direction 
> progress information, and the second asking what I could do that would 
> be of assistance in porting morphic tools to tweak which received no 
> replies.

There was a period earlier this year (and I think it was the period you 
are describing) where I didn't get email from the Tweak list. Not sure 
why this happened but when I sent a test email (let me check ... yeah 
that was on June 26th) everything seemed to work (again?). Before the 
last message I have is from february. So there were probably some emails 
inbetween that I didn't see.

Cheers,
   - Andreas