[csw-pkgsubmissions] newpkgs py_webpy

[Philip Brown]

Thanks for taking the time to look at each of those.
Let me preface this email by summarizing where I'm at.
As I am neither the python maintainer, nor even a python user, I dont
have a particular "personal" stake in all this. You're the python
maintainer - it's pretty much your call.
As the release manager, I just want to make sure that we have a
carefully considered policy in place, so that down the road, we dont
have a situation of, "oh,hmmm.. didnt think of that. We should maybe
rename some stuff back the other way now".

renames are very disruptive to the end user experience.
(both on a "urg this is annoying" level, but also gives a bad
impression of opencsw: "arg, these guys have no idea what they're
doing, cant even keep names straight!")

So, to continue...

On Sat, Jan 22, 2011 at 5:21 AM, Maciej (Matchek) Blizinski
<maciej at opencsw.org> wrote:
> .....
>> As the python maintainer, is it your stance that ALL of these should
>> be renamed to py_xxx?
>>
>> I would ask you to make your decision not merely on "yes because thats
>> what the policy says now", but, "is that what is going to make the
>> most sense to our users"?
>> I would also ask you to carefully consider each one as an individual,
>> rather than just making a blanket decision without careful inspection.
>
> Sure.  What criteria should be applied?  What kinds of cases should be
> special enough to break the rules?

I'm not entirely sure :-/ My thoughts are mostly, "if something has
strong name recognition".
Where unfortunately, "strong" is completely subjective :-/
Based on what I previously posted, along the lines of "webpy.org", it
would see like webpy does have exceptionally "strong" recognition,
even by fairly objective standards though.

Since "webpy" would be currently the "only" one in the [not
standalone] python related category, I can understand the tendency to
go with "oh lets just make it like everything else".
So that's why I asked you to consider the other stragglers, to see if
this is really a "unique" case, or whether there are other "strong
name recognition" candidates like it.

Even if it was completely unique in our catalog, my personal view of
this, is that "webpy" really seems to have uniquely high name
recognition. Because of that, it really seems to me that a relatively
fresh solaris admin who wanted to install it (either for themselves,
or for a user request), is most likely going to try first

# [pkginstaller] install webpy

So, that's my view. but I'll leave the final call up to you.