[RANT] Come on people! ;)

goran.krampe at bluefish.se goran.krampe at bluefish.se
Fri Dec 17 07:04:50 UTC 2004


Hi!

(trying to get back from rant mode to reason mode)

=?ISO-8859-1?Q?St=E9phane_Rollandin?= <lecteur at zogotounga.net> wrote:
> goran.krampe at bluefish.se wrote:
> 
> > The illusion that Squeak is moving fast is also due to all the cool
> > packages that are moving fast - but those are built on top of Squeak and
> > aren't causing things to break.
> 
> what I call Squeak includes a lot of packages. my own application builds 
> on top of a large code base.

Yes, I agree with that too. I just meant that different parts move in
different speed.
And my own packages (for example) will move in the speed I decide etc.

> > So... it is not that people don't have the time to "keep up" - the fact
> > is rather that we are dragging along a lot of "dead code". Code that has
> > *no* active maintainer. Nor very many users.
> 
> this may very well be your impression. for example to me wonderland is 
> not dead code, and I use it: I developed a 3D game with it that is not 

It is more than an impression. With "dead" I didn't mean "not used" (I
wrote not very many users) but primarily "not maintained". And such code
tend to rot, simple as that. I can give you numerous examples - but that
seems pointless - you must be aware of examples too.

> finished yet and that I left aside for one year, being busy with 
> something else. I definitely intend to finish it, but I don't know when, 
> as I'm a slow programmer :)

Not sure what your point was.

> >>we may have a real problem here 
> >>if there is not some sort of centralized direction for Squeak 
> >>development, or at least common goals and strong guidelines. 
> > 
> > I strongly disagree with the analysis. 
> [...]
> > What we DO need though is better structure and better tools. 
> 
> ... "common goals"

...no. But sure, common goals are fine. It just seems to me that such
"steering" doesn't work very well.
Sure, it is not a black and white thing - as many things we can agree on
the better. But it is dangerous to set up goals and then try to "force"
the community to reach them. As I said - people mainly scratch their own
itches. But things tend to turn out fine somehow anyway.

> Easier to
> > see who is doing what and how things affect each other. Better ways of
> > communicating and codeveloping. etc
> 
> ... "strong guidelines"
> 
> so apart for vocabulary, where do you disagree with me ?

"strong guidelines" sounded like "a bunch of words we agree on". I meant
tools and mechanisms. And social processes.

> now let me state that the ideas I expressed in my previous post are not 
> theoretical: I speak from experience. Squeak has been my main 
> development environment for several years now, along with Emacs, and 
> there is definitely this feeling I get with Squeak that things can 
> change under my feet at any time that I do not have at all with Emacs. 
>   it's very, very uncomfortable. one may wonder if the system is 
> reliable: don't you think this is a worthwhile concern ?

It is a worthwhile concern - but Emacs isn't trying to go anywhere is
it? Squeak is trying to evolve.
And with that comes change. Inevitably. Things *will* change under your
feet. But hopefully in such a fashion so that you can adapt. For
example, if you are trying to live in 3.9alpha - then of course things
are changing - you are then living in development land.

What we of course should (and *are doing* IMHO) try to do is to make the
experience of change as comfortable as possible.

> regards,
> 
> Stef

regards, Göran



More information about the Squeak-dev mailing list