[OSM-talk] Thoughts on OSM design, and looking forward and back

SteveC steve at asklater.com
Wed Feb 24 14:06:36 GMT 2010

On Feb 24, 2010, at 2:27 AM, Andy Allan wrote:

> On Tue, Feb 23, 2010 at 11:42 PM, SteveC <steve at asklater.com> wrote:
>> Thanks for your thoughtful comments...
> And I hope you've taken some of them on board.
>> You don't seem to make any realistic suggestions for moving forward and just, instead, suggest potlatch is fine as is the front page. That doesn't seem to be in touch with the reality of every newbie who encounters the project, does it?
> That's not what I said, and it's not what I believe either. Please
> don't misrepresent me. You should know both from the time you employed
> me to work on these issues, and discussions we've had since, that your
> accusation here is false. But it's a nice sidestep of the issue I was
> discussing.
>> Sure, you don't like the way I communicate sometimes
> Hmm. It sounds to me like you don't think you've done anything wrong?
> I notice you don't write "Sure, I'm grossly offensive without due
> cause sometimes", but instead infer that the issue isn't the way you
> behave, instead it's that other people dislike your behaviour.

Andy all I'm doing is repeating what I get from newbies all the time, and adding another sentence that you ignore because below you just want to evolve things, and I think it needs a step change.

Sentence 1: the UI/UX is crap

I think we all know it is

Then the Sentence I add is: Let's fix it.

>> but do you have any ideas at all on improving things other than the status quo and pissing on someone for doing anything?
> Yes I do. You should know, since I've discussed them with you before.
> But since you disagree with them, it's probably easier for you to
> accuse me of suggesting that everything is fine, and then portray me
> as a malcontent holding the project back.
> Alternatively:
> * Steve goes and reads "Art of Community"

Andy, building community and moving things on is not just "be nice to everyone all the time". Sometimes it's also about being honest and saying, this is wrong, we can fix it.

> * Steve apologies - without caveats - regarding his demeaning of
> existing developers work

You're not seriously suggesting the PL user experience is non-crap and the codebase is non-crap right? Because I think that's all I said.

> * Steve re-reads
> http://wiki.openstreetmap.org/wiki/Developer_community , which he
> originally wrote
> * We all work constructively on motivating our developers and
> nurturing new ones.

Have you tried getting a flash developer to work on PL? It's very, very hard.

> * We acknowledge that the developers actually know what they are talking about.
> On the user experience:
> * Work continues on Potlatch2
> * We make available the rails_port in a DVCS (e.g. git), to encourage
> experimentation
> * We implement the ideas on branches and see what works:
> ** this redesign,
> ** the previous redesign from CloudMade
> ** OSB integration,
> ** the "minor UI improvements" list
> ** Integrating Walking Papers into the main site
> ** Integration of routing
> ** More layers on the map
> ** and other suggestions
> * Add a javascript click-tracker to see what people actually press
> * Focus redesigns on making things easier to understand, rather than
> removing options
> * Run multiple hack weekends, with the aims of encouraging new
> rails_port contributors, and existing developers to work on these
> topics

All of those are slightly evolutionary, will take forever, and won't improve things a whole lot. I think it's much more useful to step change things.

Some of them, of course, are super good things like git and continuing PL2.. but 'adding more layers' doesn't do anything for a new user. A better UX, less is more, and a very clear up front help page and a clear feedback tab will help hugely. That's a step change we can just get done very easily.

> If you want to spend money on achieving the above, then sponsoring

I tried and tried with Richard. I tried hiring him, sponsoring him, sponsoring him to learn AS3 and all kinds of other things.

> existing developers to do the work, or sponsor someone to aid new
> developers (through documentation and engagement etc) would be my
> recommendation. Or hire a venue for hack weekends, buy pizza, fund
> travel and so on. These are all things that have worked great before.
> If you want to spend time on the above, then engaging with the
> existing developers and finding out what they want help with would be
> my recommendation.

Andy, the point is the existing developers are holding things up and aren't listening to the thousands of newbies who throw themselves at the site every day, then give up. You can pour cold water over my $70 design all you want, but the fact is it's the only major step in ages and you might not like it, but just go and read from all the newbies who *do*. The point is you're not the intended audience, and the intended audience is coming up with all kinds of cool stuff on uservoice while you're saying here "let's take 12 months and fix a few bugs" when they want something entirely different.

On of the nicest ideas I saw was splitting openstreetmap.com and .org - what you think of that? Have a nice interface on .com for newbies and then the community hub etc on .org

Yours &c.


More information about the talk mailing list