[squeak-dev] The Trunk: Monticello-dtl.685.mcz

Chris Muller asqueaker at gmail.com
Sat Nov 10 01:50:21 UTC 2018

> > YO!!  Two versions for one feature?  Before I could even get my
> > feedback submitted?
> >
> > Somebody got a itchy commit finger there!    :(
> I don't understand what you mean.
> I put this in the inbox last Tuesday. Eliot said it looked nice.

"Last Tuesday?"  Is that supposed to mean "it's been a long time?"
Because, to me, it's not nearly enough time to properly think about
what we're really wanting to do, what we want to guard against, and
the best way to accomplish it.

But apparently a new button has now already been added to our dev tools...

> Chris said "you read my mind", which I took to be positive feedback.

That was positive feedback about the _idea_ of being able to reparent,
but I did not fathom that this implementation was your final

> Bert sounded cautious and asked for my used case, which I provided.

Bert had concerns about safety, but this implementation completely
ignores them.

> Levente replied with another used case that seemed to make sense.

Dave, he presented a use-case in which a Reparenting function is
useful.  We all agree it is useful function, but you didn't allow
adequate time for feedback and refinements.

> What other feedback should I have been waiting for?

Actual usage feedback.  How many times has it actually been used?

Do you actually think this raw power should be there with no warnings?
 <---- Please don't take that as me advocating that a Warning is the
way to make it safer...   e.g., What do you think of my suggestion in
my other email about selecting one from the existing ancestry?  What
you have (as well as Adopt) is selecting one from the _Repository_.
That's unnecessarily dangerous for the use-cases we're solving.

> Or are you saying that the original Monticello-dtl.685 version
> should have been deleted, rather than merged and retained in trunk?
> That does not sound right to me.

Yeah, Eliots commit was a couple of days ago,so it would be nice if
you could've merged it before committing -- and, hey -- you actually
passed on the chance to try out the very feature you were implementing
(for the very first time?) by leaving dtl.685 behind and Reparenting to eem-685.
I guess I'm fine to just move forward and make the refinements it from
here, but I hope you're convinced to let some things of this caliber
to gestate longer, and until you are we are all _certain_ that we have
our _final product_, the diamond we want to commit to trunk.

Either that or liberal use of a new Squash feature proposed by Bert...

 - Chris

More information about the Squeak-dev mailing list