Tim Rowledge certainly wrote :

> >         1/-  34 classes with 0 instances  (so shouldn't be in the
image):
> completely wrong. Just for one example, if you want to have integers you

    Now I realise that the superclass chain analysis is completely broken.
    So those classes with 0 instances may still be needed.
    This must be fixed.

    But, there may still be un-needed stuff in there ?

    Cheers,

    PhiHo.

----- Original Message -----
From: "Tim Rowledge" <tim at sumeru.stanford.edu>
To: <squeak-dev at lists.squeakfoundation.org>
Sent: Tuesday, February 11, 2003 3:58 PM
Subject: Re: Image Builder (Re: Mythical small kernel images?)


> "PhiHo Hoang" <phiho.hoang at rogers.com> appears to have written:
>
> >         1/-  34 classes with 0 instances  (so shouldn't be in the
image):
> completely wrong. Just for one example, if you want to have integers you
> need the integer branch of the hierarchy available SmallInteger
> (probably Large*Integer as well), Number, Magnitude, Object. Oh and you
> need Class and thus Metaclass, ClassDescription, Behavior etc. in
> order to have methods to run (so you need CompiledMethod,
> MethodDictionary - and Dictionary, Set, Collection). Then again if you
> want to have almost any control structure avaialble you need True and
> False and thus Boolean. MethodContexts and BlockContexts are rather
> useful in actually running code, so ContextPart has to come along.
> Unless one proposes a single threaded system you might like Process (and
> thus Link and ProcessorScheduler and Semaphore - bringing in LinkedList
> and SequencableCollection)
>
> It takes a village, of sorts, to make a child image.
>
> tim
>
>
> --
> +================================================================+
> |Though a nation watched her falling, yet a world could only cry |
> |As they passed from us to Glory , riding fire in the sky.       |
> |(Jordin Kare, Fire In The Sky)                                  |
> |Farewell Columbia.                                              |
> +================================================================+
> Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
>
>