[pydotorg-www] [Psf-redesign] Fwd: [Pydotorg] Fwd: pydotorg 'easy' github issues

[anatoly techtonik]

Sorry, I have to leave for a day or two, so I don't have time to strip this
long stream of words. It is a little bit harsh, too verbose, may be cynical and
expressive, but looks ok.

On Sat, Aug 9, 2014 at 1:07 AM, Steve Holden <steve at holdenweb.com> wrote:
> [BCc'ing RFP Redesign to end-thread them]
>
> On Aug 8, 2014, at 2:23 PM, anatoly techtonik <techtonik at gmail.com> wrote:
>
>> Hi,
>>
>> Judging from the fact that there was no response from pydotorg inner circle
>> where all admins are supposed to be, and no response from the so-called
>> 'community of pydotorg-www' I propose reorganization to make things clear
>> and obvious. In result:
>>
> Didn't you propose the last reorganization and naming?

Right. If you look at https://mail.python.org/pipermail/pydotorg-www/
it all started with my mail to kill pydotorg-redesign list. Note that this list
and pydotorg-redesign were both dead at that time.

Note also that there was this move few days later, which is in my opinion
good (moving from private to public if I understand correctly) and bad
(giving private list a more privileged short name than to public one)
https://mail.python.org/pipermail/pydotorg-www/2010-April/000094.html
and list got a lot of new a very exciting talks (sometimes even
overexciting).

> What is this one supposed to achieve?

This list is dead. Take a look at the stats. Traffic is mostly
requests for adding
people to wiki to promote courses and events after wiki editing was closed. List
for a open source open community collaborative site with over million downloads
and I don't know how many daily views gets wiki closed for edits, because
nobody can write a piece of code in Python with several simple checks and
moderation queue to eat spam? Yea, we don't have time for that. And all of this
at a time when US invests millions of dollars in education programs to teach
children how to code. All this knowing that there are thousands students who
are learning Python and would be glad to do something useful instead of
boring and predefined tasks. Damn.. Even companies are interested to keep
communication with upstream communities and give apprentices task useful
for upstream (if only upstream had resources to review them).

I don't know who are involved in all recent python.org activity, but apparently
even they are not here if nobody can answer the question who updates teh site.
Don't get me wrong, but IMHO that's an indicator that this list is dead if all
shiny pydotorg stuff happens elsewhere.

So, this reorganization is supposed to achieve only one ML for all pydotorg
email traffic. There is also IRC, tracker, and I suppose Viber, Hangouts and
friends for more tight contacts (Trello, Basecamp anybody?), so I guess that
even this won't help (Facebook, Forums, G+). But at least we need to stop
confusing people with a multitude of lists. Call this a rebranding (for which
it would be nice to get a designer with a few JS gurus on decorating
Mailman pages).

https://en.wikipedia.org/wiki/Rebranding
"Rebranding is a marketing strategy in which a new name, term, symbol,
design, or combination thereof is created for an established brand
with the intention of developing a new, differentiated identity in the
minds of consumers, investors, and competitors.[1][2] Often, this
involves radical changes to a brand's logo, name, image, marketing
strategy, and advertising themes. Such changes typically aim to
reposition the brand/company, occasionally to distance itself from
negative connotations of the previous branding, or to move the brand
upmarket; they may also communicate a new message a new board of
directors wishes to communicate."

I am not proposing any radical changes knowing how hard it might be to
be both critical and constructive. pydotorg name it easier to use and attract
people to. There is a reasons why Google calls their tools Webmaster tools
and not WWWtools.

>> - pydotorg-www@ stops to function, redirects to pydotorg
>
> Where it was before yo suggested renaming it to pydotorg, if I remember correctly (which I admit I may not).

I never suggested renaming public list to pydotorg-www. There was a time when
I suggested the same thing - make pydotorg public, but people objected that this
will reveal keys, password and sensitive private talks. So this time I
am proposing
a rename.

>>  - pydotorg@ is closed and renamed, new pydotorg is created, which
>>    becomes open (public) point of contact with public list of subscribers (with
>>    admin/activity/committer to project marks), searchable archives and link to
>>    web interfaces for posting (nabble or Mailman3)
>
>>  - infrastructure@ is the list where all sensitive information is
>> handled - closed
>
> Of course.
>
>>    or open - depends on admins professionalism, who do not rely on security
>>    by obscurity schemes
>
> I doubt the infrastructure admins would be happy sharing architectural information with people they were not intimately professionally and sometimes personally known to them.

If they are comfortable with this - https://github.com/python/psf-chef
(which is the right thing), then I don't see why they should feel
unsafe discussing it in public. In the end you can't update the site
without knowing who to ask for. Closing this information from public
doesn't work and doesn't help really. ">>> 2. How can I check who the
project admins are?"

More effective ways to mess with security by installing rogue WiFi
points on Python conferences or even passively eavesdropping insecure
traffic that comes to bugs.python.org and PyPI websites.

>>  - security@ is the list to accept mail from lurking hackers, as usual
>>
>> If that goes well, I promise to give a further boon to resolve user experience
>> problems for new contributors, and I guess that we need some fresh blood who
>> knows about HTML5, Angular, d3 and all that fancy CSS generators that are
>> extensively used by Ruby folks.
>>
>> That's is deadly wrong that we don't have an inclusive playground for new
>> talents and instead of learning new things and adapting out habits, force them
>> to use our old and often very awkward practices.
>>
> The web site is an open source project. It isn't the sense of what's needed that's absent, it's a busy crew of volunteers swarming all over the site code (including HTNML and CSS) making it better. It's an active team of developers anxious to make python.org the best open source language web site in the world. Without that, little else is going to help.

Nothing personal, but compare the site of wannabemost popular language
of the world:
https://github.com/python/pythondotorg with 30 contributors
with some not well known tool written in this language for managing
the aforementioned infrastructure:
https://github.com/ansible/ansible with 833 contributors.

And know that among those 30 there are only 10 really busy "volunteers"
https://github.com/python/pythondotorg/graphs/contributors
"volunteers" in brackets, because if I remember correctly, the initial
version of the
site (which is probably reflected in Github history) was made by paid
contract from
PFS.

> Given the lack of response to my post, what difference do you think these changes would make?

Depends on motivation and desire of people to build an open process with proper
crediting, high-score, some fun, experiments and review process. It
will also need
a lot of coordination maybe with the help from outreach initiative
Python web also
needs to be attractive place for people with web skills who want to
learn Python,
and a lot of dynamic stuff and integration stuff with all Python
services out there.
Yes, I said hiscore https://twistedmatrix.com/highscores/

Site are made for users, and the most important task is user experience. So,
I think that being a part of one [pydotorg] team is more pleasant
experience than
being a busy volunteer of underprivileged [pydotorg-www]. Maybe that's a lot of
buzz from nothing, but in having fun that's a big difference.

Also, [pydotorg] need a lot of people who can draw and make visualizations,
because decisions in web should be based on data and not on intuition of a few.
Just do a corridor testing and ask people which name do they prefer.
If I had any
money I could bet $100 that over 3/4 would chose to strip -www suffix.
I am not a
social scientist, so betting is my way of proving things.