> >On Tue, 3 Apr 2001 JArchibald at aol.com wrote:
> >> Second, Squeak has not been around for 20 years (though somewhat
> >> miraculously, a core [corps, corpse] of some of the original Smalltalkers are
> >> still around, disguised as Squeakers).
> 
> Tim Moore <moore at bricoworks.com> replied...
> >Wasn't early Squeak derived directly from a running Smalltalk-80 via
> >System Tracer?  If so, Squeak has been around for 20 years, in some sense.
> 
Dan Ingalls wrote:
> Indeed.
> The first running Squeak was a direct derivative of the APDA (Apple Programmers'
> and Developers' Association) release of Smalltalk-80, and contained many unchanged
> methods (most of what we did at first was rip out the toolbox and put in a simple
> file system, plus write the VM, object memory, and translator to C).  That in turn
> was a fairly direct derivative of the penultimate Xerox release of Smalltalk-80.
> I can't remember how those tapes (!) were designated, but the one that Apple had
> was the one before spelling correction was added.  It would be interesting to know
> how many methods in the system are unchanged (a) from the APDA Smalltalk, and (b)
> from the Xerox tape.  The latter would qualify for 20-year status.

I've wished a few times that the method timestamps had been added earlier in the history of the Smalltalk-80 descendents.  It'd be cool to see a timestamp like "di 9/15/1980 12:17" in Squeak.  (Then you'd know for sure that that was a well-written method. :) )

The earliest timestamp I've seen in Squeak was from 1996, I think.  I assume the timestamp feature was added shortly before then?  There are still quite a few methods in the base classes with no timestamp, that I assume are pre-1996.  (Unless there was a tragic mishap somewhere along the line in which timestamps were lost...)

- Doug Way
  dway at riskmetrics.com