contributing/self contained system

[Noah Evans]

Aaaah... okay. I made progress on the plan 9 port today. I manually
de-ANSI'ed interp.c and got a useless vm compiling with all stubbed
out functions. This was more work than expected because the squeak vm
is so heavily tied to ANSI and includes ANSI code as strings in some
situations. I manually edited out the #if's and #defined away printf.
In the end a real plan9 port will require subclassing Interpreter,
VMMaker and friends or parameterizing their included headers and
changing how liberally they use the preprocessor.

That said I'm curious to see how well the thread(2), draw(2), mouse(2)
and keyboard(2) apis map to the squeak model.

http://plan9.bell-labs.com/magic/man2html/2/thread
http://plan9.bell-labs.com/magic/man2html/2/draw
http://plan9.bell-labs.com/magic/man2html/2/mouse
http://plan9.bell-labs.com/magic/man2html/2/keyboard

Any advice?

Noah


On Fri, Feb 8, 2013 at 9:13 PM, Craig Latta <craig at netjam.org> wrote:
>
> Hi Noah--
>
>> Is there a protocol for contributing to spoon?
>>
>> ...
>>
>> I've used metacello to get packages so far. Is this the canonical way
>> to get object packages on spoon?
>
>      At the moment contributing to Spoon means taking code away (not
> adding it), or debugging Naiad, an all-live-objects module system.
>
>      The current focus is on testing the releases and documenting
> desired use cases. We're using http://bugs.squeak.org for that. As for
> code contributions, Spoon is rather unusual, given the goals of a
> minimal system and new modularity technology. Part of the work is
> defining the long-term contribution protocol you seek. While we're in
> this bootstrapping phase, I'm coordinating contributions manually.
>
>> Has there been any work done to make the system self contained? (ie.
>> containing scamper, celeste, irc etc... so the user never has to
>> leave the system)
>
>      No, currently we're developing a minimal system with which you can
> load and unload code modules. Creating "distros" from that is a later
> goal. We can prepare for that, though, by finding the boundaries and
> dependencies between the code we have now in existing object memories
> (e.g., previous releases of Squeak, Pharo, and the other Smalltalks). We
> can use the existing change management tools to do that (e.g., Monticello).
>
>      Please see the details about Naiad on my blog[1].
>
>
>      thanks!
>
> -C
>
> [1] http://thiscontext.wordpress.com/a-detailed-naiad-description
>
> --
> Craig Latta
> www.netjam.org/resume
> +31 6 2757 7177 (SMS ok)
> + 1 415  287 3547 (no SMS)
>
> _______________________________________________
> Spoon mailing list
> Spoon at lists.squeakfoundation.org
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/spoon