SK53 has commented on the following diary entries
|Mapping of industrial entities||about 1 year ago||
Whereas I believe OSM should be a good platform for this type of information we have some way to go in developing robust approaches to capturing it. (In fact my recent stuff on shops and other retail outlets is really about looking at the 'value' chain for creating data for a sub-set of commercial entities). I think the same issues are likely to apply for industrial entities.
Broadly speaking I think the problems are:
Although there are many directories containing this type of info, many are copyright and most have huge amounts of cruft in them. Even things like company registrations (if available as Open Data) are likely to have poor encoding of industry sector (no one checks, its just for the statisticians, so there are no serious controls).
One starting point is to look at specific industry sectors which tend to have large, obvioius and well-known industrial or manufacturing plants. Off the top of my head: steel works, car factories, shipbuilding, chemical works, petroleum refineries). It's probably easier to get the OSM community to map one or more of these as a challenge.
|Fed up with abbreviations in tags||about 1 year ago||
You can add atm to your list (probably more frequently used than any of the above).
It's fine to peeve, but it works two ways, there are plenty of tags which are completely opaque or confusing to native english speakers too.
IIRC cycleway=asl (advanced stop line) was discussed, but probably on the osm-gb IRC channel and around 67% of current uses are in Great Britain.
Some uses of abbreviations are inevitable, such as ref:INSEE (not obvious if you're not French) or ref:vatin, particularly when these are not supported by editors. Needing to type in cycleway=advanced_stop_line is a good way to discourage their mapping. Your examples happen to all be documented on the wiki, which is a tad surprising. Adding editor support for such keys can hide the naughtiness of illicit use of abbreviations in the value.
Personally I find traffic_sign one of the more opaque keys.
|Massachusetts Lake/Pond Cleanup||about 1 year ago||
It's good to see this cleared up, but it's a shame that the MASSGIS import has been preserved ahead of objects which were mapped by OSM contributors with some degree of local knowledge.
I added Scargo Pond back in 2009 long before any water body data were imported. Back then the Upper Cape looked weird because some landscape features had been imported and others hadn't.
When I look again at this area I am struck that although there is a lot of detail of buildings, conservation areas and so on, there are next to no POIs, or sign of anyone human mapping in the area.
|Land Use = Open Space Buffers||about 1 year ago||
I, and many others use landuse=grass for this purpose. Unfortunately this latter tag is also often (mis-)used for other types of grassland: for instance grass pasturage on farms (clearly landuse=farmland) and grass on sports fields (leisure=recreation_ground). (In many case this has obviously been used becaused it gets rendered : "tagging for the renderer").
The more general term which is widely used is the catch-all "amenity grassland" which covers most types of publically-owned grassland in cities (including sports fields, many public parks as well as the random bits of green space).
In either case tags with dual semantics are a real pain, but I would suggest sticking with landuse=grass, perhaps with some kind of adjectival tag landuse:grass=open_space_buffer, which would allow future disambiguation of the different semantics.
|Land Use versus Residential Private Property||about 1 year ago||
@CloCkWeRX I still don't get exactly why it needs to be in OSM: are you wanting to free cadastral data from commercial constraints,? 'cos it doesn't sound like it when you mention commercial datasets. The type of application you describe sounds mainly of interest to organisations which can afford non-free data.
Most of the applications that use landuse which I am interested in: contesting major planning changes, pollution in watersheds, environmental (habitats/biotopes etc) value of particular areas, and so on - are things which single individuals, small non-commercial groups, and charities are likely to want to do.
@AdamMartin it's nothing to do with database weight (if I understand what you mean by this), although lots of buildings from cadastral data in France are regarded by many mappers as 'bloat' (largely because houses w/o addresses arent very interesting), it's to do with the pain of processing lots of polygons. The nature of OSM is that one does have polygons in polygons.
|Land Use versus Residential Private Property||about 1 year ago||
Broadly speaking don't use landuse for zoning areas.
Use the landuse tags for what is on the ground (which of course ought to correspond with local zoning). Many local administrations will do zoning on a long-term timescale of perhaps several decades. It is no use to anyone to mark areas earmarked for new sections as residential if they are still farmland.
Secondly, dealing with many hundreds of landuse polygons will make landuse data from OSM more or less unusable. Because there is no enforcement that given types of polygons must not overlap, a lot of post-processing is needed on OSM data already. Increasing the number of objects (by perhaps a hundred- to a thousand-fold if you map individual house plots as residential and individual shops as retail. Additionally complex algorithms are then needed to define landuse beyond property boundaries.
I've yet to see a convincing use-case for mapping property boundary details in OSM: enough people have done it to show it's possible, but I'm not aware of anyone with a need to get this data from OSM. Incidentally the vast bulk of use-cases for landuse (including cartographic products) require reasonably generalised polygons.
|Think bout boundary of INDIA||about 1 year ago||
I suspect that you think that the boundary of India should include all of the pre-partition States of Jammu and Kashmir. However it is policy for OpenStreetMap to map what is on the ground. This effectively means that we try and map the Line of Control accurately, rather than areas claimed, but not controlled by India or Pakistan (and indeed China).
The reasons for this are straightforward, OSM is used in countries on both sides of border disputes and therefore it is impossible to adhere to both views of where notional borders might be. Furthermore OSM, unlike, say the maps of the Survey of India, does not serve political purposes. OSM is not a platform for fantasy maps, whether those created by individuals or mandated by legislation by governments.
I realise this may create problems for people in India where it may be an offence to represent the boundaries of India other than according to the states wishes, rather than any accordance with reality. For that purpose I would suggest using an external source for the boundary of India: moving the boundary away from the LoC will be treated as vandalism.
If on the other hand the boundary needs minor tweaks (which I'm sure it does) and you can map them based on personal observation or sources which are open, usable in OSM, and readily available for others to check that such mapping is accurate then do go ahead.
I have also answered this on [OSM Help](https://help.openstreetmap.org/questions/23042/how-could-we-edit-the-boundary-of-india-and-pakistan_.
|Adding addresses: NG9||about 1 year ago||
This is a great piece of work: and an extremely thoughtful insight into the complexities of address mapping.
|tagging a green roof?||over 1 year ago||
Green roofs often have a wide range of different plants depending on a whole host of factors. Relatively few will be seeded with grass.
More often they will have relatively drought tolerant plants, certainly local ones have lots of sedums, and a range of Carophyllaceae (Dianthus, Petrorhagia and others).
For this reason I have used the green_roof=yes tag, see http://www.openstreetmap.org/browse/way/37015658.
|OSM datasize in PostGIS||over 1 year ago||
In my experience its often faster to use osmosis to write a pgsql dump & just use copy to do the import (--pgd option in osmosis).
I think you'll find most of the storage overhead is in the geometry columns
|Mapping Croydon||over 1 year ago||
The Food Hygiene ratings have a large number of shops, pubs, schools, hospitals (and a few oddities) with OGL address data including postcodes. Their lists exactly fulfil your idea of hunting down POIs from a list. The data for Croydon LB is here: http://ratings.food.gov.uk/OpenDataFiles%5CFHRS507en-GB.xml (warning XML). You can read this with Excel or I could send you a CSV extract with the key info. The lat/lon are actually the postcode centroid.
|Addresses from an old survey||over 1 year ago||
OGL is fine. Even Ordnance Survey OGL is OK, but tedious from an attribution viewpoint. CodePoint Open is not OK, but we can substitute the Office for National Statistics postcode data set which is to all intents and purposes identical.
In addition Food Hygiene, Land Registry and other available data sets (such as Planning Applications) contain a huge amount of useful address information. See my blog Maps Matter for far more about this.
We have huge amounts of OS data under their extended OGL, and as OGL is in effect a more liberal licence there is no problem with OGL.
|Would there be any use for a builder=* tag?||over 1 year ago||
In the UK the phenomenon of a housing estate being known locally after the company which built/developed the land is pretty common. The area of Silverdale in Nottingham was (and may still be) known as the Wimpey Estate (a national firm of house builders); and the Hofton Estate (a local firm of builders) is to the west of Wollaton Road.
The development off Cranbrook Drive in Maidenhead is still called the Sterling Homes estate some 30 years after it was built.
The up-market house developer Berkeley Homes always puts a distinctive marker of properties which it develops including the year of completion.
Recent controversy about 1880s housing in Liverpool (the Welsh Streets) reminds me that there are many examples of 19th century builders leaving their mark in street names, such as these streets which spell out the names of the builders, Owen and Williams.
In other words there is plenty of evidence here that such information tends to stick around in the public memory.
Another reason, particularly for national builders, is that individual building firms often have a particular house-style of building. Wimpey houses of the 1970s are very recognisable, but I'm sure to this is true of other builders. I know architectural historians can often date old buildings with surprising accuracy because certain features reflect specific time-bound trends. I am sure that there are plenty of clues as to the builder in a similar way.
|Activity on rail||over 1 year ago||
Be careful UPRN's belong to someone.
|Lycian Way - Turkey||over 1 year ago||
Many parts of the Lycian Way are mapped between Fethiye and Antalya, see http://www.openstreetmap.org/?relation=51855 (might be a bit slow to load) for an idea of what is there.
Relations aren't readily downloadable as a GPS track/trail although it is possible.
If you are using a Garmin device there are many ready made files containing walking routes, see http://wiki.openstreetmap.org/wiki/OSM_Map_On_Garmin/Download. Extremecarver offers downloads specifically made for walkers and mountain bikers at Openmtbmap, with useful optional contours as well.
|Activity 1||over 1 year ago||
I think that you don't need to do this.
You are probably not aware of relations which are used to add things like the European Road network. See this page on the wiki.
Here is an example of a short one E 24.
Many of these relations are very unwieldy (take a long time to save when edited), so the int_ref technique might still be preferable, or the relations could be broken up which was done with the US Interstate system
|Updating housing site no 5...||over 1 year ago||
You have an issues with projections: you can't just grab OSM data and hope it to look like OSGB data.
I'd do this with Quantum GIS, set up a project using the OSGB projection (EPSG: 27700) with conversion on the fly checked. You can load some OS OpenData into this (for instance the 250k raster or the Meridian 2 vector data are both very suitable).
Then download your required OSM data (this should now be re-projected automatically into OSGB). Do whatever styling is required.
Now create a composer which enables you to set the bounds of the area (I presume you mean 6 1km squares) & export this as SVG.
You could also download your OSM data as Shapefiles and use ogr2ogr to do the conversion, but this is what QGIS is doing and it hides some of the gory detail.
The SVG files can be huge: I've not bothered with them myself, preferring PDFs.
|Airport Runway Names||about 3 years ago||
Hmm, how do we know these exist except from Google Maps. If that is how we found out about them then we're deriving this data from Google. Doesn't seem like a good idea for me. Let alone using Google Maps to name the runway.
GNIS does have names for islands in this part of the Gulf. According to Birding UAE both islands are private without access. One of them may still be a breeding site for the Socotra Cormorant.
|Rae Lakes loop||about 3 years ago||
fee=yes/no is a more or less universal issue in UK (particularly railway (train) stations which are more or less 100% fee. Toilets are increasingly only open with restricted hours, once the decision has been taken to close them the opening time is determined by staff hours. See Gail Knight's blog for a comprehensive introduction to the issues.
There is a good tradition about obsessing about mapping toilets and sanitation facilities on OpenStreetMap and often for very good reasons: How to map open defecation areas".
|Naming passes seems difficult...||about 3 years ago||
The usual way to mark a pass is with a node tagged mountain_pass=yes with a name=* and ele=*. This could be a node on the way marking the trail or a single standalone node.
For part of a longer trail which both have names follow the suggestion of FK270673 (a relation for the longer trail and its name, the way for naming a section of the trail).
There is a single use of a tag for trailhead with a value of start. I'm sure marking trailheads has been discussed on one of the mailing lists in the past. Feel free to use the existing tag or invent one of your own. The bulk of mapping footpaths in OSM is European-based where most footpaths form a dense interconnected network, so the notion of a trailhead is less significant. Signposts (finger posts), waymarkers and information boards are much more likely to be mapped.
Check out Lonvia's Hiking Map to see which trails have been mapped worldwide: at large scales this shows a lot of hiking related detail.