Re: [Axiom-developer] Ping: file removals

[Waldek Hebisch]

> > I agree. In general there should not be any pdf files in the source
> > code repository.
> 
> Eh? Methinks you've missed the whole point of aim of the axiom project.
>

Tim, first I must question your monopoly on stating Axiom goals. 
In open source project all members choose goals.  If members
goals are fundamentally incompatible, then there is time to fork.
However, (as I belive is in Axiom case) if goals differ slightly,
it should be possible to work togehter (compromising on lesser
goals if needed).

Secondly, I read many times what you wrote about Axiom goals and
I find it still not clear -- there are nice dreams (which I share
with you) and implementation strategy (which I doubt will work).
 
> I must say I'm completely confused by the current directions.
> Who invented the idea that we need to recreate everything from source?
> And who invented the idea that we don't keep exact copies of published
> documentation (with ISBN numbers) in the archive?
> 

I think here is crucial misconception: what I (and other people) propose
is separation of SCM and Axiom distribution.  Typical user will
fetch _distribution_.  Exact content of distribution may be discussed
separately, but the point is that distribution should be usable
"immediatly" (or maybe with minimal effort).  Axiom servers should
archive past distribitions -- from may point of view it is not
very important how (dated tarballs, blobs in datbase or version
contrion could do that).  SCM should contain just source.

If sombody thinks that we should distribute a Git pack containing
full Axiom archive (including history), I have nothing against
it (as long as int in not the only distribition format).
 
> > > One extra remark:  It make sense to put generated documentation
> > > into distribution tarball, but IMHO generated files (with the 
> > > exception of files needed to bootstrap) should be removed from
> > > source archive.
> > 
> > Yes, I agree completely.
> 
> Again, methinks you've missed the whole point of the axiom project.
> 
> There is NO distinction between documentation and "source code". That is
> oldthink. In this brave new world there are only literate programs.
> Creating human readable output from human oriented input is THE soul
> of the project. We need to completely entwine the research with the old
> idea of source code and conflate the idea of compiling with the idea of
> latexing the code. This is THE PRIMARY DESIGN GOAL of the axiom project.
>

Tim, I am affraid that the newspeak does not allow you to say clearly
what you want.  For me compilation is basicaly freezing some aspects
of source and caching intermediate low-level data to gain execution
efficiency.  Now, human oriented input means that we really _do not
want_ latexing as extra step: Axiom content should be presented in
optimal way without any extra effort.  Similarly with compilation:
the user should see human oriented input and the final result, while
the intermediate data should be hidden.  

Whith this in mind, can you sincerely say that machine generated
Lisp is a "human oriented output"?  We may wish to show some
specimens of machine generated Lisp to illustrate how the compiler
works, but the bulk of it belongs to a special ghetto.
 
-- 
                              Waldek Hebisch
address@hidden