Hi Jan,

On Thu, May 11, 2000 at 03:33:30PM -0700, demiourgos at smalltalk.org wrote:
> On Wed, 10 May 2000, Alan Kay wrote:
> 
> > 
> > Josh --
> > 
> > At 7:50 PM -0800 5/9/00, Joshua Gargus wrote:
> 
> 
> [snip]
> 
> > 
> > I hope so too! But I said that about Smalltalk 20 years ago -- in the hopes
> > that something much better would be invented by the next generation -- but
> > here it still is! 
> 
> [snip]
> 
> Alan, Josh,
> 
> Doesn't it depend upon the niche and the product 'though?  Like typewriters,
> "old style phones",  toilet tank mechanisms, and chopsticks have seen very
> little diversification in design until recently, the latter two still essentially
> the same as they were 100 or more years ago.  

I think that progress does depend on the relationship between the product
and the niche.  When the fit between Smalltalk-80 (in its various incarnations)
and the current commercial uses for general purpose OO languages is considered,
there is no clear impetus for radical change.

However, the niche is not necessarily static.  For example, Morphic provides
a new way of interacting with a world of objects that is qualitatively 
different from the 'familiar' way to do OO (ie: the way paved by Smalltalk
and more recently (sorta) followed by Java).  By demonstrating new ways that
people can interact with the system, the boundaries of the niche are enlarged.
As we experiment with the new possibilities, we will inevitably discover some
shortcomings of the changes we've made.  If we then redesign the system to
try to take these observations into account, we've just completed an iterative
cycle that can be repeated as long as we have the will to do so.

I think that this last phrase is the key.  Left to their own devices, 
corporate entities tend to focus on the bottom line.  In an environment where
there is a paucity of 'blue sky' research funding (such as ARPA in the 60's),
commercial interests will try to squeeze as much money as they can out of what
we already have, rather than attempt to make revolutionary technological 
breakthroughs.  Alan & company have been kind enough to find a corporate
sponsor willing to invest in research for research's sake (Thanks, Squeak
Central!), and progress on the 'blue plane' has resumed.

This is my own interpretation of things I've read... if anyone can help 
clarify my understanding of the situation, please do!

<snip>

> It's possible, I imagine, that the Smalltalk style is composable enough that
> any conceivable system could be easily made of it, and that might be
> its lasting quality.  I think FORTH had a similar quality, which is why it
> got so much use among people who had both to calculate and control
> various pointing and experimental mechanisms.

Although "any conceivable system could be easily made of it" is stronger
wording than I would use, I do agree that Smalltalk does lend itself to
the type of composition you allude to.  However, a paper I recently read
about aspect-oriented programming has me thinking (in a rudimentary way)
about how we might improve on the current way of doing things.  Please
don't ask me to articulate any of these thoughts ;-)

Joshua

> 
>  --jtg
> 
> 
> 
> __________________________________________________
>   Jan Theodore Galkowski          algebraist.com/ 
>   www.whysmalltalk.com/        www.smalltalk.org/
>   demiourgos at smalltalk.org       marssociety.org/
> **************************************************
>   PGP Key Fingerprint: 2757 F86D AA51 677D 38D7  
>                        964B 9A8D 7852 A494 3790
> **************************************************
>   Get my Public Key from my home page at:
>       http://algebraist.com/
> **************************************************
> 
> ______________________________________________
> 
> Get free e-mail at http://www.britannica.com

--