I find mid-development changes to the project name drastically
disruptive. Perhaps I am, indeed, doing it wrong. But I will not drag
another project through such a trial; from now on, the name must be
settled very early in development. I'm sorry to hear this precludes
guidance from this list. I have promised myself never again to trap
myself in mid-development name change hell.
On Wed, 2010-08-25 at 10:13 +0200, Aldo Calpini wrote:
> ... there is a fine line between "formally correct" and
> "obnoxious to read", and you crossed it :-)
My intent is to be respectful and clear at all times. I'm sorry; but I
cannot bridge the gap to a viewpoint where this might be obnoxious. When
I read, I strongly prefer precise, definite, highly detailed writing. I
care nothing for length so long as the words tell. When I write, I
strive to avoid all misunderstanding and offense. I take your criticism
most seriously and lament my failures on each count. You are not the
first to say this and I'm well aware that my only possible reaction, to
be yet more painstaking and polite, only increases the offense. But I
lack command of the register in which I might say, "Well then, screw
it," and somehow make it right. I must ask you to take my words at face
value; they do not hide supercilious condescension.
> I don't understand what you are trying to say.
> ("bring with us to the consultation"?
The project I described in OP generates naming suggestions. We authors
can include these suggestions in RFCs, instead of coming empty-handed. I
use the metaphor of an experimenter who walks into a room to consult
with colleagues.
> ...who are "those who guide"?
I have been led to think that this list is, in part, intended to guide
module authors in their choice of module names. If OP project is
successful, it still cannot replace helpful human insight.
Sorry; I simply don't have a way to apply "vaporous" and
"unsubstantiated" to what I've written. I would interpret the former as
"vague", yet I am criticized for being too detailed and specific. I
assert so very little that I don't see what substantiation I might
offer. If you feel I've made an objective statement ungrounded in fact,
please point it out and I will be happy to cite sources or retract.
The leap from solicitation of comments on a proposed module's name to
insistence on a "perfect" name is not one I would make. I would be
content to pick a name, following my own taste, and stick with it.
I lack clear direction here.
Make sure you choose an appropriate name for your module early
on.... You should contact modu...@perl.org to ask them about
your module name before publishing your module.
-- perlmodstyle
The very first keystrokes in starting any project comprise its name. But
it appears that both modules@ and this list, module-authors@, want to
hear about modules after considerable work is done. Is it not obvious to
all that this is the very worst time to decide upon the name? In the
current thread I see this thought expressed more than once: "Renaming a
project is a trivial matter and can be done easily, at any time." My
experience is so radically at odds with this that I cannot join the two
viewpoints anywhere. The fault may be mine but I recoil from the far
side of the abyss.
Naming your Perl packages well is one of the most important
things you can do.... The modu...@perl.org (the mailing list
for PAUSE admins) and module-authors@perl.org can help you
choose a good name.
-- PAUSE, On The Naming of Modules
I read this as something stronger than a mere invitation to ask for help
here. Perhaps more detail would be useful.
...once chosen, you rarely have the opportunity to change
[a module name] after people start using it. The name of the
module isn't for you; you don't need a name because you
created it and understand it.
-- ibid.
One-third of this is in agreement and two-thirds at variance with my
experience. Of course, the name cannot be changed without upset once any
public mention of it has been made. Links break, searches fail. The name
is just as important to the author as to anyone else; how can one think
about anything without naming it? It is not even technically *possible*
to begin a project without choosing at least a working name for it.
> Test::MOAR::MOAR::MOAR :-)
Please forgive my almost total inability to perceive the thrust of most
forms of humor. I have looked up the word 'moar'. I don't need the
smiley face to tell me that you are making a joke but I can't understand
what serious point you include, if any. I suspect you discourage me from
writing yet another Test:: module of any sort. I would be delighted to
discuss exactly what I expect to accomplish and how (or if) an existing
module meets these needs. But I will not do so without a direct
request.
Clearly, you want me to demonstrate a module before you comment on the
proper name for it. By now, I hope it's equally clear that I consider it
essential to fix the name before beginning work, let alone before
showing it in progress. Therefore, we are at an impasse.
Despite the sources I cite above, I have noticed that most module
authors seem to publish their work without consulting anyone on their
choice of name. I think it unlikely that I will run into serious trouble
if I follow their example. I apologize for attempting otherwise. I had
hoped to follow the marked path, with no intent to tread on the grass. I
offer my most sincere and serious regrets.
--
-Xiong
