[Talk-GB] [Spam] oslVosm, OpenData Locator 2010 data and not:name

Tim Francois sk1ppy14 at yahoo.co.uk
Mon Oct 4 16:46:34 BST 2010


It would indeed be a shame if we/they just ignored the not:name data - I
grep'd the current Great Britain OSM dataset and there's 1054 instances of
k="not:name", which is not insignificant. (If anyone wants to see the
results download it in zipped txt format http://tiiiim.com/osm/<goog_1770857475>
not_name.zip - I got grep to also provide the line immediately before and
after the k="not:name" instance for some context.)

Tim

On Mon, Oct 4, 2010 at 10:14 AM, Peter Miller <peter.miller at itoworld.com>wrote:

> great news.
>
> Regarding the not:name tag at the OS, I think we need to be a bit patient.
> The OS is a large organisation and do take time to change. They are actually
> changing fast at present and have expressed enthusiasm for the not:name
> technique and I think it is just a matter of us using it for some time while
> it gets into their systems. We are going to do some more analysis on it at
> some point and will be working with the OS of the processes. It would be
> shame if we gave up using it before they got into listening! In the mean
> time it is a useful way of stopping people checking a conflict that someone
> else has already determined is a error on their part.
>
>
>
> Regards,
>
>
> Peter
>
>
> On 3 October 2010 17:47, Tim Francois <sk1ppy14 at yahoo.co.uk> wrote:
>
>> All,
>>
>>
>>    1. If there are any heroes left using oslVosm<http://wiki.openstreetmap.org/wiki/OslVosm> [1]
>>    to compare OSM data with OS Locator data, I've just updated the script a
>>    little so that it now honours the not:name tags. Any highways which include
>>    a not:name tag are automatically assumed to be 100% correct in OSM, whatever
>>    OSL says, and are ignored in any further comparisons. This assumption may
>>    have to be reviewed at a later date...
>>    2. Anyone else using/noticed that the Locator data has been stealthily
>>    'updated'? I remember Robert writing something about it a while back, but
>>    can't remember the conversation. In Bristol, a load of roads seem to have
>>    disappeared, whilst some more have been added. All in all, there were about
>>    10 more roads than in the '2009' release. Anyone else notice differences?
>>    3. not:name. Are many people using it? Is it working? And are these
>>    being sent back to OS? The reason I added it to oslVosm is that it seems to
>>    be being used here in Bristol, so it is useful for 'accurate' numbers.
>>
>> For those that may not be aware, oslVosm is a script which compares OSM
>> data against OS Locator data, and can output a gpx, kml or wiki file of any
>> discrepancies. It also tries to do some spell checking if it finds similar
>> names. It is primarily aimed at people with a programming tilt, as it only
>> works from the command line (and probably only in Linux). For those who want
>> an easy to use interface, use Robert's excellent Musical Chairs [2] web
>> interface or ITOs slippy map layer in Potlatch or JOSM [3].
>>
>> Cheers
>> Tim
>>
>> [1] http://wiki.openstreetmap.org/wiki/OslVosm
>> [2] http://ris.dev.openstreetmap.org/oslmusicalchairs/map and
>> http://wiki.openstreetmap.org/wiki/Using_OS_Locator_files#Browseable_OS_Locator_to_OSM_comparison_with_fuzzy_matching
>> [3]
>> http://wiki.openstreetmap.org/wiki/Using_OS_Locator_files#OSM_and_OSL_Differences_as_Background_Tiles
>>
>> _______________________________________________
>> Talk-GB mailing list
>> Talk-GB at openstreetmap.org
>> http://lists.openstreetmap.org/listinfo/talk-gb
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-gb/attachments/20101004/6202f543/attachment-0001.html>


More information about the Talk-GB mailing list