Namespaces (was: Re: [ANN]A plan for 3.8/4.0...(insertdrumrollhere))

Lothar Schenk lothar.schenk at gmx.de
Sun Apr 4 08:50:54 UTC 2004


stéphane ducasse wrote:

> ...the idea was you have multiple Smalltalk
> environments (difficult to express again)....

Mmm, that's something I've been idly speculating about. I don't know if this 
holds up to the scrutiny of experts, but has anybody ever considered the 
possibility of splitting up a Smalltalk system at the image level? It is, a 
Smalltalk system should be comprised of multiple images, not just one.

You already have that case (but in an unintegrated manner) when you have a 
networked situation with two machines each running a Smalltalk system and 
communicating. So, why not have the same situation in your local environment? 
You could have several VMs with different images running at the same time, or 
even one (reentrant) VM executing several images at once. Not all images need 
to incorporate the same components as every other image. I think this is 
essentially what classboxes are intended to achieve, to separate one 
modification from another.

For example, you could move all of Morphic into one dedicated image and have 
all the other images (sans Morphic)  communicating with this dedicated 
Morphic image for their user interaction (essentially similar to what the 
X-Window system does). Or you could have two different Morphics in different 
images (maybe one of them a modified version of the other one) and have the 
other images communicating with either one or both, if desirable.

I am aware that this raises the question of how to coordinate all these 
different images into one coherent whole, which I am not, unfortunately, 
qualified to answer. It only seems to me, that this question just might 
become more tractable by moving it outside of the scope of any one particular 
image or development environment.

As indicated, I am very much a dilettante, here, and my idle ramblings are 
supplied for what they're worth: free of charge. :)

Regards, Lothar



More information about the Squeak-dev mailing list