OpenStreetMap

robbieonsea has commented on the following diary entries

Post When Comment
Water in OpenStreetMap about 1 month ago

An excellent MapRoulette challenge - thanks.

[Although one could argue for the case of large elevation change (e.g +200m) or ways from the sea - could be automatically fixed by a bot].

It would be nice if the link to edit with Id used the main OSM.org's Id rather than the OSM.us version!

1000th Edit 3 months ago

Congratulations.

Here's to the next 1000 edits!

All the best.

Another new MapRoulette challenge: Suspected Missing / Wrong One Way Streets 6 months ago

Are you using Telenav GPS traces (so presumably only driving traces) or OSM traces?

Otherwise for OSM traces one would need to filter out traces from walking which presumably can go in any direction, for example perhaps just discarding traces with slow average speeds when performing vehicular type analysis.

Can you also enlighten us on the 'confidence level' methodology? (or mention in your SOTM talks)

Thanks for all the improvements to MapRoulette.

Am I the Only One... 7 months ago

Yes definitely.

I've taken to using my SatNav all the time when driving/being driven - mainly for recording random things seen whilst out and about - rather than for routing instructions (generally I know where I'm going!). Alternatively sometimes I video it instead but I don't have a high def device only a cheap £40 one.

It helped when I learnt how to actually create a waypoint on the SatNav - for a Garmin Nuvi255 - just press and hold down on the screen saves a point where you are and name it.

Presently I seem particularly addicted to spotting historic milestones. http://wiki.openstreetmap.org/wiki/Tag:historic%3Dmilestone

Unfortunately they're hard to spot (especially at night!) and on drive (or cycle) by you never have time to see the description on them.

Finding the sections that speed limits apply to is also one of my aims. Often - if I remember - I'll check the speed limit map of where I'm going - http://www.itoworld.com/map/124?lon=-0.91523&lat=51.21377&zoom=12 to see if it needs improvement.

Elevation 8 months ago

Cool map styles.

It's always interesting to see new cartography efforts,

cleanup monitoring_station/measurement_station 11 months ago

Seems generally sensible to me.

The proposal itself mentions amenity=clock, so one observation is why use display:analog=* instead of display=analog (and of course likewise for a digital display)?

NB for a clock the display is always implicitly at least yes, otherwise why map the clock at all :)

Is it a precedent already set that should be followed?

SOTM13 about 1 year ago

Me too - even if I said I was going to do this merge for my locality ages ago.

Although I still have my own collected data (of various other things) to process first...

GPS Traces about 1 year ago

Note that GPS_Editor applies a constant time diff to each point. The time difference between points is kept - thus calculating speeds between points is still viable - and could point to where one has exceeded the speed limit.

However post calculating 'instantaneous' speeds from GPX tracks is not reliable since the standard only has an accuracy rounded to the nearest second. Often my GPX tracks from a Garmin Nuvi saving every second means I can get speeds over 100mph, when otherwise going slowly (or a spurious point)!!

Averaged over a several trackpoints should give a more robust speed value - and hard to deny that would be due to spurious points (especially if a median or modal average is used).

I don't know if the OSM Traces API will allow all trackpoints having the same timestamp - as that would be the only way to remove speed inferences. According to http://wiki.openstreetmap.org/wiki/API_v0.6#GPS_traces - it says "trackpoints must have a valid timestamp", thus setting all points to 01/01/1900 could be feasible.

Unfortunately it will not help people trying to process these files to gather speed statistics ;)

NB2 that GPS_Editor is Windows only.

Goes off to consider adding some kind of Anonymize Time function to Viking (http://sourceforge.net/projects/viking/) which I maintain...

Just tried the new iD OSM Editor... over 1 year ago

Me2!

Especially don't try panning the view. It makes Potlatch2 seem blazingly fast.

I don't remember the early alpha/beta versions of iD being sooo slow.

I think I might leave it a while before attempting to use iD on my mobile phone...

Is it possible to have access to statistics about the application of tags to GPS traces? over 1 year ago

Good question - I don't know for sure, but I don't think there's any easy way (not without resorting to importing the full GPX data into your own Database and running queries against it -- eek! -- see http://github.com/iandees/planet-gpx-dump for the gory details)

Perhaps you'll have more luck asking this question on the help site:

http://help.openstreetmap.org/

impassable gate? over 1 year ago

Not quite sure what you mean by 'impassable (tall)'! Height restriction?

There is the opening_hours tagging to specify timing properties:

http://wiki.openstreetmap.org/wiki/Opening_hours

Light up my way! almost 2 years ago

+1 for cycling and walking.

IMHO lighting for cars is fairly irrelevant - you have powerful lights to see by and a metal box to keep you safe.

Which route I would take in a car would hardly ever be effected by the lighting. Horses for courses I suppose.

Have the lights been completely removed from the above mentioned places? Parts of the A3(M) around north Portsmouth are on an energy saving mode so they turn off around midnight to about 6am.

I've often thought about blanket setting all roads in Portsmouth to lit=yes, but I haven't been bothered. Cycleways/service roads/footpaths vary greatly and I've tended to tag those.

To my mind, in principal a router should automatically assume roads in a (UK) city/large town are lit. I would have thought not being lit is exception. Small towns/villages are much more variable.

changing mapping interests almost 2 years ago

Initially I into mapping bridleways and footpaths, but most of those around my neck of the woods is very complete.

ATM I'm addicted to postboxes and max speed tagging, but also adding missing road names, adding shops.

Then there's things no other maps have like recycling facilities, lighting, surfaces, etc...

The more involved you get the more you realise there is to map.

Mapathon - Operation Cowboy almost 2 years ago

Maybe this could be tied in with a new Map Roulette challenge for Arm Chair mappers?

https://oegeo.wordpress.com/2012/10/29/new-maproulette-challenge-connectivity-bugs/

I suspect the current connectivity errors could already be all fixed by the end of November.

OSM Server Virtual Machine Available Now. about 2 years ago

Good stuff!

I've always thought about doing something similar, but other things have taken priority.

I thought maybe the setup for the VM (as far as possible) should be configured itself (and as such held under source control eg in github) rather than a collection of [Wiki] instructions.

Vagrant seems a good open system for defining Virtual Box images, see: http://vagrantup.com/

but I can't say I've actually used it.

Royal Mail Pinpoint about 2 years ago

Good work.

I'm not entirely convinced how accurate the collection times are anyway, the other day I posted a postcard and just in time caught the post person emptying the box at 7pm. The box's listed time was around 5:30pm.

IHMO It's just a guide as too whether it's likely to be an early or late final collection, and to know where the definitive late post box time is for the area.

I think collecting the post box type & royal cypher is vaguely interesting (on a historical slant) as the Royal Mail themselves may not have this info.

Royal Mail Pinpoint about 2 years ago

'I rather hope that they build a dataset of their postboxes...'

They already have.

See:

http://www.whatdotheyknow.com/request/location_of_every_post_box_that

http://www.whatdotheyknow.com/request/last_collection_times

And to go about mapping it:

http://www.dracos.co.uk/play/locating-postboxes/

http://robert.mathmos.net/osm/postboxes/