[OSM-talk] The new OpenStreetMap.org design

SomeoneElse lists at mail.atownsend.org.uk
Sun Dec 8 23:37:01 UTC 2013


On 04/12/2013 00:52, John Firebaugh wrote:
> This past weekend, the OpenStreetMap.org front page launched with a
> new design.

First of all - thanks for posting here.  As I mentioned in the other 
thread it's always helpful to put a human face on some of the design 
decisions to try and understand the reasons for change, especially when 
the reactions to the redesign have been "mixed".

> Concretely, here are the improvements we implemented:

Like Christoph Hormann, I'd be interested to see how you are planning to 
measure these improvements, both for new mappers and for existing ones.  
What and when are you planning to do in this area?

> - A better experience for veterans.

This is where most of the complaints seem to be targeted (not 
unnaturally, since new users by definition won't know if they're getting 
a better or worse experience than they would have had previously).  For 
example, the new design has less space devoted to information and more 
devoted to a map that may or may not be completely irrelevant.

https://github.com/openstreetmap/openstreetmap-website/issues/564

is one example of an "irrelevant map", but there are others.

The "less space devoted to browse object information" results in the 
effect that I measured in

https://lists.openstreetmap.org/pipermail/talk/2013-December/068779.html

- roughly 50% more mouse clicks and keypresses needed to do the same job 
compared to previously.  It is now more of a "memory test", since 
different bits of information that need to be compared are now far less 
likely to be on the same screen.  I'm currently working around this 
latter problem by killing the css on browse pages when I need to compare 
stuff - it doesn't look pretty but at least you can read it.

In the future, how do you propose to address this?  Is there a reason 
why the left-hand-column size and map picture size are fixed?

> There's no longer a needless distinction between "browsing" a feature
> and "viewing it on the map".

I'm not convinced that that was a "needless distinction", actually. When 
sending messages to new mappers trying to help them with stuff I used to 
use either a "browse" page, a "history" page or a "view on map" page 
depending on what it was that I wanted to highlight - it's no longer 
possible to do this with the OSM website.

> And navigating between features and
> changesets is fluid, fast, and preserves more context.
I'm also less than convinced here (see github issues 568, 576, 581, 585, 
588, 589, 591, 619 and 629 at least).

In other cases where functionality has been removed (github issue 623 is 
an example) something that was never intended as an end-user feature 
(but end-users did find useful) has gone, so it's not in a sense a 
"bug", but it does decrease the usability of the site.


> - Bug fixes and usability improvements. Most notably, the site works
> much better on mobile devices. For other fixes, see the linked issues
> at end of the pull request ([5]).

Some of those seem to be "there was an issue with feature X; we've 
changed how it looked so we must have fixed the bug".  In some cases 
that simply isn't the case.  Take

https://github.com/openstreetmap/openstreetmap-website/issues/211

which was closed because the page the bug referred to no longer exists 
in that form (correct - it doesn't).  Unfortunately the replacement page 
has this problem:

https://github.com/openstreetmap/openstreetmap-website/issues/564

which is a more serious issue than the previous just "needing a better 
style".


It may be that what we need is the current osm.org website for, er, 
whoever it is for and something else for people who actually need to 
work with the data.  It's something that's covered in detail in the 
description of

https://github.com/openstreetmap/openstreetmap-website/issues/631

While here Frederik says a second website won't come any time soon we 
already have many external tools and collections of tools already 
available (I'm thinking of Ian Dees' "deep diff" tools and the osmhv 
ones among others)


One other question - OpenStreetMap prides itself on creating a map 
that's inclusive (Wheelmap, 
http://wiki.openstreetmap.org/wiki/OSM_for_the_blind, 
http://blog.openstreetmap.org/2013/12/03/disability-mapping-openstreetmap/ 
etc.).  What work was done to make sure that the new website is more 
rather than less accessible than the previous one?  I'm thinking 
particularly of

https://github.com/openstreetmap/openstreetmap-website/issues/588

here.  I have fairly good eyesight but the white-to-cream colour change 
in the changset list and on the map is difficult to see.

Cheers,

Andy

PS: It was interesting to read the links at the end of your post to the 
July "Upgraded map controls" mails which include 
https://lists.openstreetmap.org/pipermail/talk/2013-July/067632.html and 
the immediately preceding one (which notes the inactivity on the 
"design" ML).  What I'm not seeing is any discussion about design (how 
do we design something to make it more useable?), compared to lots and 
lots of discussion about implementation.  Is there a link anywhere to 
the actual _design_ discussion that lead to Saman's slides at SOTM-US, 
and who was that discussion with?




More information about the talk mailing list