browser metaphor (Re: squeak program delivery, etc)

Alan Kay alank at wdi.disney.com
Mon Dec 7 19:51:58 UTC 1998


Michal --

At 5:03 PM -0000 12/7/98, michal starke wrote:
>> objects handled correctly already carry their important behaviors
>> with them and thus don't need to be interpreted by a "browser".
>> ... distributes everything except how the user's screen is allocated.
>> One's "browser" has now shrunk to almost nothing except the management
>> of screen real estate.
>
>an underTheHood-ignorant question: why does squeak itself not take this
>approach? More precisely: why not free the objects from the 'image',
>allowing them to live anywhere that is reachable by url/pointer and have a
>separate 'browsing' object to achieve the effect of the current image?
>
>[this would trvially address the original poster's question of how to
>produce 'independent apps'! also, as a side effect this might allow the
>'extreme multiplatform transportability' you mentionned, without
>compromising integration as much as it is now]
>
>[i can imagine the historical reasons why that approach was not taken (if
>st was developped by and large as its own OS/desktop). but those reasons do
>not hold anymore in a world where just about every squeak is embedded in a
>larger system (inside an OS/desktop or more generally inside infoBahn)]

You are quite right on all counts -- and we are oozing
("Object-Oriented-Zeno-ING") towards just such a model by gradually
reformulating the old model in such a way that we can make progress on
other fronts as well ... There are a few "minor details" having to do with
caching, versioning, etc., that have to be done really well ...

Cheers,

Alan





More information about the Squeak-dev mailing list