Front page | perl.perl6.internals |
Postings from May 2001
Master-Apprentice and a "sneak peek"
Thread Next
From:
A. C. Yardley
Date:
May 21, 2001 16:28
Subject:
Master-Apprentice and a "sneak peek"
Message ID:
157145788109.20010521183640@tanet.net
Ok, the idea of a master-apprentice program, where the wannabes,
while working with masters, sort-of cut their teeth on the docs, and
then move up in skills over time, has been mentioned a few times on
this list and discussed rather extensively on perl6-meta. Now, I'm
wondering, has a consensus on this emerged among the members
perl6-internals? Is this something which can now be implemented?
If the answer is yes, then I'd be happy to try to set up something,
*IF* someone will point me in the right direction as to what he or
she believes is an appropriate approach. E.g., a mailing list? A
web page on, say, on perl.org, delineating the respective internals
subgroups, including the names of the masters within the subgroups
and their respective apprentices? Both? Something else which I've
not considered?
Correlative of the above is, is there any actual code for us, the
mere mortals (but, perhaps, I am only speaking of myself here) in
the bunch, to review? A "sneak peek," so-to-speak.
I know, from prior posts, some have expressed concerns about the
idea of periodic releases, especially of works-in-progress. The
principal concern, of course, is the distraction--from all of the
questions and speculation that is sure to follow any such periodic
release--to those who are doing the hard bits. On this, however, I
have a suggestion.
Perhaps, you, Dan, Simon, and others, could adopt a two-tier
approach: Where you (plural, you)
(1) state expressly you will *not* entertain or respond to any
questions about the periodic release, consequently leaving any
discussion (and the concomitant speculation about the same) to the
members of the list, and
(2) state, either expressly or otherwise, you will respond to
questions (or ask for suggestions, etc.) about the periodic
release.[1]
Is such a scheme feasible?
Please note, with the above suggestion, I am *not* trying to "paint
[anyone] into a corner." Rather, I am only trying to come up with a
solution to address one of my concerns:
Like everyone else, I have many responsibilities, and my time is
divided among those responsibilities. Unfortunately, while I
*wish* I had both the time (and the skills) now, today, to be more
involved with the design and the development of Perl6, the fact is,
right now, today, I don't. I, again, like everyone else, must
allocate my time and effort within rather fixed constraints. What
is more, this is unlikely to change in the next six (6) months to a
year, or so. (But one can always hope. :-)
Consequently, in the absence of periodic releases, there is a very
real possibility, imo, the Perl6 internals code base will grow, by
the time of the first release, to such an extent that it will prove
rather difficult, for some of us (read me), to get on top and to
stay on top of the development track.[2] With periodic releases,
however, even of works-in-progress, I (and others) will have a greater
opportunity to get on top and stay on top of things.
In addition to the above, I suspect some will want to *experiment*, at
the same time as the initial code base is under development, with
other language implementations. For example, someone may well want
to develop everything (or just bits and pieces), starting from the
first periodic release, in Perl or C++ or Java or ..., even as the
initial C code base undergoes development.[3]
Hopefully, the above is clear and reasonably persuasive. Opinions?
Remarks?
Regardless, I do apologize if I have overstepped in this instance.
/acy
-----------------------
[1] I'm thinking of something like Chip's approach with Topaz.
[2] Again, I'm thinking of Chip's approach with Topaz.
For example, Chip first released topaz-0.09, and then
subsequently began to release, topaz-0.10, topaz-0.11, and this is
where the track stopped, IIRC. But because topaz-0.09 constituted a
substantial base of the work-in-progress, it was rather difficult to
get a handle on what he was doing in topaz-0.09 *before* he released
the subsequent releases. Especially, again, given my own personal
circumstances.
If, however, I'd have had the opportunity to start closer to
the ground floor, e.g., a middling release, somewhere around
topaz-0.05 or so, I believe I'd have had a much easier time of it
all.
To be fair, I should note that Chip was writing Topaz in C++
and employing, imho, some C++'s more obscure features, for
efficiency reasons, and this was also an impediment in *my* getting
up to speed on topaz' code base.
[3] Some such, of course, has been mentioned on this list in prior
posts.
Thread Next
-
Master-Apprentice and a "sneak peek"
by A. C. Yardley