Re: [binarycloud-dev] A Request for "A Standard"

[Alex B. <en...@tu...>]

> Ummm, this discussion/interchange has been interesting for
> me.  Lordy, there are things I haven't thought about that
> BC is going to allow.  This bloody BC is as accommodating
> as a Bangkok cathouse.  I have to confess that I didn't
> even consider a 1st-loaded library class to serve for my,
> er, what shall I say, "global-but-not-global-scope" storage place.
> In a way, I have sinned by suggesting global scope,
> something that is a no-no in C.  Mea culpa.

I'm not entirely convinced yet that a core class isn't necessary, but so
_far_ I haven't been able to find any conditions in my head that would make
its existence necessary. We'll see.

> And, well, I am starting to think about the wonderful
> possibilities of calling $some_module->Init($options)
> from elsewhere (anywhere if loaded later) just to get
> different data from that module's Init(), that, with an
> option "directive", would store it in, say, a variable in a
> 1st-loaded library class variable.  Like you say,
> of course, these mothers can already talk globally
> and tete-a-tete.  I see now that I don't need a
> global BC_LUNCHBOX which now seems to appear to me as
> a kind of vulgar cut buffer. [NOW DEPRECATED]

hehe

Note that Init() no longer exists. it's the constructor now instead.

Also, you would probably want to do the "passing" with class variables in
the module:

$this->moo;
which would end up being something like

$Page->modules[x]->moo (or something, I didn't look)

> I can see/understand your "dilemma" about possibly
> establishing a kind of standard class/library that is
> loaded on initialization by BC.  What I hope is that
> the design will somehow be such that module writers
> all over the place will _tend_ to do data passing in
> a kind of "standard" way, that will allow a minimum
> of rewriting for local conditions.  This is the ultimate attractiviness
> of BC for someone, who, for example, needs some  M$Word guru
> somewhere, someplace, to contribute a WORD.DOC to
> HTML translation module.  I certainly could not do it,
> and therefore I am going to have to depend upon the
> kindness of strangers.

Word to html is pretty much covered, I have an applescript _now_ that cleans
xml+html+goober ms word html export.

That will become php, and part of something else, which will allow joe
schmoe to paste in a word doc html export and get clean code out the other
end.

It is doable! I can see the liiiiiiiiiiiiiight!

> Don't let me slow you down on unimportant matters right
> now, but, what is going to be the infrastructure
> or place for contributed modules?

I have considered a pear-ish infrastructure far into the future, but to be
honest it should be only a question of gathering developers to make
binarycloud more of a tool than brochureware. Installing modules would
really be like "tar -xzf foo.tgz ../binarycloud/user/mod/" and have a nice
day.

if you don't like the module:

rm -rf ../binarycloud/user/mod/foo/

there will be shell scripts for doing install/remove automatically (and
maybe even a browser interface or something)

For binarycloud, I imagine a directory of user/contributors, a cvs browser,
a bug tracker, and a module repository. probably forums, too.

> We fantisize a lot here about BC.  We were blue-sky-ing
> the other day about a kind of "Module Browser", and
> an application that could select modules and then
> build a site.  We could put this on a laptop and
> Marketing could prototype a site on the spot with
> a customer.  The PROBLEM is, shall we say, is that
> living here in the South, the customer would likely
> say, "Sheeeat, I ain't a-gonna pay you XX thousand fer
> that, ya jes built it in 5 minutes....".

Though, uh, you could actually do that :)

> That last paragraph was for levity only.

But, you could :)

_a


--
alex black, ceo
en...@tu...

the turing studio, inc.
http://www.turingstudio.com

vox+510.666.0074
fax+510.666.0093