[squeak-dev] Smalltalk v. .mcz Objects
overcomer.man at gmail.com
Mon Sep 12 12:41:41 UTC 2011
>> Along the way I noticed the .mcz files contained a lot of other junk
besides Smalltalk code, and all the code for several FFI sub-projects is
under the same filename inside the different packages. Using .mcz instead
of .zip is unnecessary complexity. The extra packaged files are not
Smalltalk. Why not just use Smalltalk?
> Why use text when you can use objects? What you call "junk" is in fact
just serialized objects. Loading them is a lot more efficient than compiling
text. The stored text is just a fallback. When loading the text, MC actually
just recreates the objects (MCDefinition instances), and then proceeds
exactly the same way as if it had loaded the serialized definitions in the
> - Bert -
The obvious error with that is that other versions of Smalltalk exist.
People with completely different Smalltalks which do not all have
Monticello, as well as different forks of Squeak like Cuis should be given
equal access to libraries of Smalltalk code. Second there is the completely
unnecessary use of the .mcz instead of .zip extension, making it more
inaccessible until one is given the simple conversion. Third, it is not
obvious that a person with an older version of Squeak, with years of
personal new code added, will have compatibility with a binary object, or be
able to update the code if needed, from a binary form. Compile speed is
inconsequential to the speed of locating a potentially valuable project in
the library. Thus Monticello as currently defined is a hindrance to the
greater Smalltalk community and unnecessary to Squeak.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Squeak-dev