[python-advocacy] How programming language webpages should be designed

Paul Boddie paul at boddie.org.uk
Wed Nov 11 00:35:12 CET 2009


On Tuesday 10 November 2009 20:59:27 Carl Karsten wrote:
> On Tue, Nov 10, 2009 at 12:28 PM, Aahz <aahz at pythoncraft.com> wrote:
> >
> > Far more important for whom?  If you agree that testimonials are
> > important, saying that code examples are "far more important" implies
> > that you think that code examples are more important than everything
> > else on that page, including the left navbar links and the news items.
> >
> > I have to strongly disagree with that.

Let's agree to disagree, then. Example code isn't more important for people 
who already use Python, but then they're probably not going to the front page 
all the time. I hardly ever visit the front page, myself; I go directly to 
docs.python.org, wiki.python.org, and occasionally pypi.python.org.

News is important because it gives the impression of vitality, but then you 
can argue that blogs and "tweets" are important, too, and many sites actually 
incorporate that stuff on the front page as well. And the navigation menus on 
python.org are, shall we say, less than optimal, although I'm not going to 
deny that they have a purpose.

It's interesting that in a recent python-dev thread, one of the contributors 
said that he'd rather try out a bunch of languages than use Python 3 (for 
many, the only one they will have recognised would probably be Haskell). Upon 
looking at the site for one of those languages, you'd probably end up 
screaming "Just show me some code!" while reading the lengthy description of 
how it relates to Self and Smalltalk - maybe it's aimed at people who've 
already decided that they like Smalltalk.

Many people would run out of patience on such a site when many would argue 
that how a language appears is an important factor in whether people will use 
it or not. I still think that it's weird that the subject - a language - is 
surely worthy of its own place amongst the prose in another (natural) 
language, and yet is kept hidden from the reader. It's like reviewing a 
painting but not actually showing it.

> > But assuming you're simply engaging in hyperbole, I repeat what I said
> > earlier: go ahead and create a sample front page replacing the
> > testimonials with code samples.  Until someone is willing to invest the
> > effort into writing HTML, I think we're arguing in a vacuum.

I actually wrote up some thoughts on this three years ago when GvR suggested 
that Python should be marketed. You can still find them here:

http://wiki.python.org/moin/MarketingPython

Testimonials say something like "Python is a safe choice and 
trusted/innovative companies/organisations really like it". To an extent this 
is a great differentiator in Python advocacy because people have collected a 
lot of this kind of material, but it only really persuades one group of 
decision-makers (the managers), and as PHP and Rails have shown, the other 
group (the people doing the work) have grown more influential over recent 
years (or their influence has been recognised).

> I am willing to invest the effort if it was likely it would be used,
> which right now it seems more unlikely given the conflicting goals.
> but I think that satisfies your requirement, so there is a point in
> the discussion (I don't think it is an argument until 2 people have
> strong conflicting opinions, and so far I am not seeing that.)
>
> I do agree with the group that thinks there should be code on the
> opening page.  I am not sure what the rest of that group thinks, but I
> am in favor of replacing quite a bit of what is currently the 'first
> view' (center, above the fold, what people are going to read before
> they click away.)

I agree with this, and I'm quite happy to put forward actual designs for 
criticism.

> But I also think I have a different view of the goal of the first view
> than others, especially whoever did  the current layout.
>
> Given that, I think we need to talk about goals before bothering with
> implementation, like "put code on the page."

I'm quite happy to work with whatever we have on the Wiki at this point, 
although I don't think we're short of material.

> I see 4 groups of people that should be considered: (list is in
> alphabetic order to obscure any sense of importance, which I am
> undecided on)
>
> 1. potential python developers - need to write something, get to pick
> the tools, wondering what python has to offer, maybe considering a
> career choice.
> 2. python developers (old and new, but the choice has been made)
> 3. python users (I have an python app, I need help making it run)
> 4. technical management that gets to pick tools for a team

This is like the Marketing Python page I wrote, except that I more or less 
ignore 2 and 3, since the topic was marketing to people who don't use Python 
already.

> I have no problem catering to all 4 groups, but each groups needs are
> fairly different and much like pycon talk selection, there are more
> options than there is room for, so we have to figure out what to cut.
>
> Personally, I think we should give as much as we can to 1 and 4, so 2
> and 3  get one link each: "developers" and something like "help me
> make my python app run" or "my app broke" (thats the only reason I can
> see #3 coming to the site.)

In fact, a lot of people in group #3 will never make it to python.org unless 
they downloaded the application from a site they found via python.org, and 
even then they're more likely to just go to where they found it (or complain 
to their distribution vendor).

[...]

> > Side note: the webmaster team is always looking for volunteers.....

I don't mind drafting concrete suggestions, but I feel that my volunteer 
budget is already spent on a number of other areas, I'm afraid. There's only 
so much time I can devote to all of this.

Paul


More information about the Advocacy mailing list