[pydotorg-www] project plan

Jesse Noller jnoller at gmail.com
Fri Apr 23 05:38:24 CEST 2010

On Thu, Apr 22, 2010 at 9:33 PM,  <skip at pobox.com> wrote:
>    Jesse> Thank you for adding a link. Now how do we solve this:
>    Jesse> http://jessenoller.com/2010/04/22/why-arent-you-contributing-to-python/#comment-46111214
> As I indicated earlier in this thread, it seems to me the best way to report
> problems about the website is to send email to webmaster at python.org.  It can
> be triaged there and sent to the right place.  Perhaps every page should
> contain a mailto: link with the subject parameter being the page's URL.
>    5> Why isn't there a code example on the front page?
>    >>
>    >> Is that also coworker's request? How would that have helped the
>    >> coworker?
>    Jesse> He's given me feedback in the past; ergo, I asked him to restate
>    Jesse> it.  It's been a long time since he, or any of us were
>    Jesse> newbies. He was pointing out that when he started, and hit the
>    Jesse> python site the first thing that struck him was a code example.
> There is limited real estate on the front page which no amount of redesign
> is going to make less limited.  Perhaps it could feature a link to a page
> full of "Look how easy it is to do X with Python" examples.  Given the
> current existing content I don't see where you'd actually put a snippet of
> code though.
>    Jesse> I know, and I'm sorry for being short; but I just gave you links
>    Jesse> to two massive conversations wherein people who are not me are
>    Jesse> discussing this very topic. Most of what I have said here has
>    Jesse> been echoed, independently, by people not influenced *by me*. I
>    Jesse> hoped that that would be sufficient evidence.
> I think you hyperbolize a bit with the phrase "massive conversations" but
> I'll give you the benefit of the doubt.  That fellow's eight-bullet response
> would be solved if there was a "Report a problem with this page" link which
> linked to a mailto: URL.

I don't think I hyperbolize when the post has been live since 10:44am
and as of 11:19pm 5,016 unique hits, and:


As well as the 70 or so comments, not including twitter or private
emails. Seemingly, it's a sore subject for many people. Not only do
they feel that "getting changes into core" is difficult, but many have
cited poor layout, lack of information, difficulty with logins, etc.

> Or is he referring to problems submitting a bug in Python itself?  It's not
> clear.  If what he wants is to report a bug with Python itself, then add a
> "Report a problem with Python" link as well which either leads to
> http://bugs.python.org/ or to mailto:report at bugs.python.org.

And it could also be solved with a web form which looked nice, had
some nice simple instructions which told the user "Submit a python
problem" with some basic guidelines, rather than just popping up an
email window. Little things like this help with user experience, small
enhancements that look nice, are clean and simple.

Not everyone likes the command line, emacs, or vim (yay vim). Not
everyone is comfortable with big svn checkouts, or complex bug
tracking interfaces. So, no, I don't think an 8 bullet point comment
can be boiled down to "just slap an email link here and there". Sure,
that's a great start, but it should not be the final solution.

> I don't know how much time you spend perusing new bug reports, but there are
> some pretty basic "bug reports" such as this old chestnut:
>    Python's floating point is wrong.  See:
>    >>> .1
>    0.10000000000000001
> which pops up every few months.  I take that to mean that there are plenty
> of people who have extremely limited facility with the language who manage
> to figure out how to submit a bug report.  Then there is always Google:
>    http://lmgtfy.com/?q=python+bug
> If that's not enough bug reports, just wait a few hours and watch Victor
> Stinner's reports come flowing in.  Realistically, we don't have too few bug
> reports.  We have too few people processing the ones which are submitted.
> You seem to have cherry-picked the responses to your blog post a bit.  I
> kind of got the impression most of the respondents hadn't considered
> contributing (in any way) to Python's development and maintenance.  "It
> works well enough for me.  I do my job and move on."  For those people no
> amount of website redesign is going to change them.

I'm perfectly familiar with the bug reports that come in. And yes,
plenty of people who seem to be able to "figure it out" - that doesn't
mean the process is easy, nor that it is empowering or low friction to
the users. My passion is for lowering the bar, simplicity, making it
rewarding, etc.

As for bugs; I don't just want bug reports. I want patches; I want
suggestions on new features. I want improved documentation and most of
all I want people to feel involved and invested in something which
they might not otherwise be. Programming languages aren't sexy to a
lot of people, but I think people can also have a greater attachment
to Python than as just another tool in a toolbox.

As for cherry picking - I can say the same to you. In no way did
"most" of the respondents say "works for me I don't care". Many of
them (most of them, even) pointed out that they perceive getting
involved as time-consuming, bureaucratic and high learning curve. The
website, and the tools and content involved can all go a long way to
help changing this perception problem. Small changes to wording,
organizing content more consistently and clearly, making important
things prominent instead of just jumbled in with the rest of the
information can really help.

It's not as if I'm suggesting we all go out back and euthanize
ourselves for horrible failures. I'm trying to advocate that there are
things we can do drastically better, and that we should. Rich, and
others are working on a proposal to help redesign, reorganize and
modernize the site, the tools, content, etc. Especially given the
feedback I feel I managed to garner in a short amount of time, I don't
think our response should be "works for me"!


More information about the pydotorg-www mailing list