[squeak-dev] election and my views (part 2)

Jecel Assumpcao Jr jecel at merlintec.com
Mon Feb 23 17:56:19 UTC 2009


An issue that I didn't mention in my first email but about which the
community is very divided is how much effort should be spent on playing
nicely with the rest of the computing world. On one extreme you have
SqueakNOS which tries not to depend on anything about the hardware, at
the other extreme you have Huemul Smalltalk (I have never understood its
relation, if any, to Squeak) where the focus is using as much of other
people's work as possible.

http://sourceforge.net/projects/squeaknos/
http://www.guillermomolina.com.ar/huemul/

Besides Squeak, I also enjoy using many other Smalltalks (though I
haven't done so in many years) such as VisualWorks, Smalltalk X,
Dolphin, Smalltalk MT, GNU Smalltalk, Little Smalltalk, Self, Smalltalk
Express (free version of Smalltalk V/Win) and so on. Some of the nicest
ones owe a lot to their tight integration with the underlying platform
(Windows, in the three cases I am talking about) and it is obvious to me
while many people long for the same thing in Squeak. You get to leverage
the work of a lot more people and get advantages, such as full use of
hardware acceleration, "for free".

But to me the most valueable characteristic of Squeak is its "runs
everywhere" feature. So my support for being friendly with other
software systems goes only as far possible without sacrificing
universality. In other words, any improvement to Squeak which would kill
off SqueakNOS won't get my vote. Full disclosure: my project is even
more radical than SqueakNOS (I don't even feel the need to play nice
with popular hardware) and I do intend to make money on it. But my vote
in this regard would not be on my own behalf since I already have to
deal with a lot of stuff like this and adding a few more wouldn't be a
big deal for me.

In my previous email, I mentioned that having all those old images which
still run allows us enough backwards compatibility that we can move
boldly forward. That is in part due to how isolated Squeak has been.
Most other applications I still have from the same time frames no longer
run on my current Linux installation.

Speaking of moving boldly forward, this doesn't mean we should ignore
other Smalltalks. I am participating in the new ANSI (no longer ANSI...)
standard effort and would like to see it become easier to move code
around.

-- Jecel




More information about the Squeak-dev mailing list