The change to the data alert dialog is a regression. The original design is preferable and makes much more sense in usability terms.

A single button, essentially implement “go ahead” is correct, because it provides a single option to change state from the default, i.e. data not showing. The new design is ambiguous, because one of the options actually does nothing - it retains the current state; its actual effect is to dismiss the dialog. It is also worse in usability terms because there is now much more to read, where as a single action “Load data” is a very clear “do something”.

That is why the current design is correct: there are essentially two options: change state (“Load data”) or dismiss the dialog (X button). I would recommend you revert this.

However, I would agree that moving to a floating dialog rather than misusing the left panel system is a sensible change. But it should be a single button with X.

Martin

We would agree that smoothness is subjective, and would prefer to see fuller mapping of surface and tracktype.

We would definitely recommend inclusion of these on both paths and bridleways - there is a big difference certainly around the UK between different qualities of bridleway, and far too many don’t have any surface indication tagged.

Martin, CycleStreets

Why couldn’t Nominatim priorize results from the bounding box or surrounding?

Turns out this isn’t always what you want. On the CycleStreets website we try to take this into account, but it’s difficult as you have to take into account context.

Suppose you are on the train from Cambridge to London, and you type in ‘Downing Street’ (which is a famous street in London). It comes up with Downing Street near where the train happens to be, rather than the more famous Downing Street. Which did you want? It’s hard for an app to know.

Your blog gives some assumptions about update times from the proprietary map providers, e.g. “probably in a week or two” and “as little as three months from now”. Out of interest, what are these based on - are there any published times anywhere or evidence for these?

(And yes, we do a fresh import every few days - thanks for the mention!)

Do include paths for walking/cycling-only if possible during your surveying. The CycleStreets routing engine takes account of lighting (and much more):

http://www.cyclestreets.net/blog/2012/05/20/more-osm-tags-supported/

If you see signage on the ground, you can give it the lcn=yes (local cycle network) tag.

Try http://m.cyclestreets.net/journey/ …

We’ll be interested to see the effect of the data improvements. CycleStreets currently has (unadvertised and untuned) routing for Denmark and we’d be interested hear from local cyclists how well it copes.

http://www.cyclestreets.dk/ (plus iOS/Android mobile apps and a mobile web version)

which will take account of signed cycle routes, path types, surface quality, cycle lane widths, barriers, the presence of traffic calming, etc. if present in the data.

Actually there is an API that can be used if wished while developing your project: http://www.cyclestreets.dk/api/journey/

Martin

Have you seen this? http://www.cyclestreets.net/blog/2012/09/03/england-cycling-data-project/

It would be great if you could help out with this project!

Welcome to OpenStreetMap! Great to have you on board.

Do have a look at CycleStreets: http://croydon.cyclestreets.net/ . The routing automatically will take hills into account automatically.

Do consider adding things like surface quality to existing OSM data - that can considerably improve routing in some cases.

I'm one of the two guys that created CycleStreets: http://www.cyclestreets.net/

I'm strongly with @Chaos99 and @AndyAllan on this one. What's most important is that you add as much objective information about the situation as possible.

For instance, numbers of lanes, the speed limit, surface type, and whether there is a cycle lane. (And is the width of multiple lanes possible in OSM yet?)

In coming months the engine will be supporting these kinds of attributes. A 70mph road in a dense area with multiple lanes and no cycle infrastructure should be a good indication that it's generally not very nice for cycling.

Martin