You do realize that you're allowed to use the "leave a comment" on your own diary entries too, right? No need to post a new diary entry or send a private message to respond to comments.

IMO, name postboxes by what anyone calls them, rather than what the RM has them down as (if nobody calls it anything more than "the postbox", then don't add a name). In the cases of boxes at POs or former POs, then I would think simply adding the ref to the amenity=post_office node will be enough (unless it's particularly large and has multiple boxes) in the case of an open PO, and adding the PO name to the amenity=post_box in the case of a defunct PO.

Since you seem to make a big fuss about the licence you signed up to, allow me to remind you of s.7(b) of said licence (emphasis added):

"Subject to the above terms and conditions, the license granted here is perpetual (for the duration of the applicable copyright in the Work). Notwithstanding the above, Licensor reserves the right to release the Work under different license terms or to stop distributing the Work at any time; provided, however that any such election will not serve to withdraw this License (or any other license that has been, or is required to be, granted under the terms of this License), and this License will continue in full force and effect unless terminated as stated above [i.e. by material breach of the licence]."

Of course, it may be that you have submitted your traces under a compatible (and therefore equally perpetual and interminable) licence, or, if the traces had been marked as being in the public domain or that you have otherwise waived some rights in it, you are estopped from exercising any rights you may hold.

Just thought you may want to know, since it makes you llook like a whining hypocrite to bitch about the respecting licence terms you agreed to and then not respect them.

I don't see how the process outlined in clause 3 can be used to remove all the source tags and the text on the credits page. While those objects are still tagged, our requirement to the OS is (AIUI) satisfied. If a third party creates a produced work and fails to mention the OS, then that is their problem, not ours.

I don't see why that's relevant. The requirement is for attribution. It is for anyone wanting to make a change which removes the attribution, or creates a work which needs to be attributed, to ensure they are obeying the rules. It is not for us to enforce other people's obligations to the OS.

Indeed - you shouldn't be adding an ncn_ref to the way, but including it in the two relevant relations. Assuming they are both National routes. Sustrans are doing some crazy things with the Regional routes with a view to abolishing or renumbering them.

I have worked on that assumption for the Welsh points, and not on that assumption for the Scottish baseline (the points fit better as arcseconds than decimal minutes). However, I realize I may have accidentally used the decimal minute conversion for the Northern Ireland limits, so will probably need to reimport those (and redraw the lines).

I think someone is misunderstanding the licence when they construct that scenario. "Share-Alike: If you publicly use any adapted version of this database, or works produced from an adapted database, you must also offer that adapted database under the ODbL." Nothing seems to suggest that ODbL data is not compatible with other ODbL data - such a scenario would be absurd. Clearly the ODbL grants explicit permission in section 3, and the rights it grant certainly looks broad enough for "any act restricted by copyright", so you would not be violating the CT. "Explicit" in law doesn't refer to something specific, but to something which is clearly stated - the opposite being "implied", from something which is assumed but not clearly stated. The permissions in the licence are clearly stated, therefore they grant explicit permission to all licensees. The only problematic licence involved is CC-BY-SA, which has implicit virality. Anything else (e.g. CC-BY) is generally not a problem, since the requirement is that it's attributed, and we have the source= tag for individual items and the master attribution on the wiki for the collection.

Having gone and converted them from DMS to decimal, and then from OSGB36 to WGS84 *twice* and now got different results, I now see from another order that actually the numbers may not have been DMS, but degrees and decimal minutes. *sigh*

Bizarrely, they seem to have felt the need to get this from what looks like a third-party supplier under licence, so they wouldn't be in a position to release. It also looks like most of the smaller discrepancies are to do with various islands that we're missing, or just needing a robot redrawing.

The worst part is that we know where the non-MLW baseline is (it's right there in the legislation - for example http://www.statutelaw.gov.uk/content.aspx?LegType=All+Secondary&title=Territorial+Sea+%28Amendment%29+&searchEnacted=0&extentMatchOnly=0&confersPower=0&blanketAmendment=0&sortAlpha=0&TYPE=QS&PageNumber=1&NavFrom=0&parentActiveTextDocId=2786697&ActiveTextDocId=2786697&filesize=12974 has the bit around the west of Scotland). It's a matter of whether plotting points at those co-ordinates and drawing the lines scheduled in the various Orders would amount to infringement. If so, we'd have a bigger problem, because the appropriate body to approach for release would IIUC be Parliament itself (or, worse, OPSI). Then there's the issue of counties, etc. There are some conflicting provisions - traditionally, counties extend to MLW, but at least one statute relating to Wales places the boundaries of the principal areas at baseline+3nm.

"CC-by-SA is only compatible with later versions or country specific versions, it is not compatible with ODBL."

I don't see why this is a problem. After all, it's a *RE*-licensing exercise. Complete compatibility between old and new licences is not a necessary requirement. Imports from public domain sources are not an issue (they are unencumbered, so no issue). Many of the other imports are also of little issue - for instance, the OS OpenData stuff requires attribution, we provide it on the master page and each individual item derived from it is (or should be) individually tagged as being from OS sources, so requirements met. It would only be a problem if anyone were stupid enough to get people to agree to CC-BY-SA imports in the full knowledge that this change has been in process since 2005.

There is no question of declining the new terms, only accepting them. If you don't want to accept them, then don't. But don't make that decision based on something asinine like the ability to import data - *any* licence will restrict our ability to import data. The current licence prevents us from importing any source with restrictions on commercial use, for instance.

FWIW, something released under CC-BY (or a similar attribution only licence) or ODbL *does* come with explicit permission for re-use. "Explicit permission" doesn't mean a personalized statement identifying the specific act you want to perform - it means that permission has been given and that it has been made explicit, e.g. a written statement that says that you can do anything as long as you attribute the source is "explicit permission".

Of course, it doesn't help that CC-BY-SA is worthless for pure data in pretty much every jurisdiction based on English common law because it's a pure copyright licence which does not address database right issues (database right is not copyright, and copyright does not cover databases or facts).

I suspect the main reason for the "success" in Haiti was that we were gifted tons of information - high-res aerials, public domain street-level maps, etc. The fact that the area was flooded makes that a problem, since lots of features aren't visible.

Indeed - I would suggest the real fix here is to make the rendering of the wetland/tidalflat combo more appropriate.

I would think that tagging it with something that makes it look brown on the map is clear-cut tagging for the renderer. natural=beach seems OK to me - whether it's a beach people will want to go to is something for the tourism= tag family. If the rendering is wrong, fix the renderer.

No such luck in my area. After 5-3-1 another three were added later, numbered 5A-3A-1A.

Google appears to take its naming mostly from OS data (as I can see from streets which are misspelled in Google that are spelled the same way in OS Locator). Don't forget that there may be other possible explanations if there are places where names change, etc. which may be missed if someone has been less that thorough with their surveying, or if the names have only been observed from crossing roads.

FGW do some odd things in and out of the Bear. LHR stopping services are sometimes shown as "Hayes & Harlington" to encourage people to use the more expensive Distress Express services.

As for KX having a Platform 0, they're a bit behind the times - there's been one in Cardiff for years now. What else would you call the platform beyond platform 1?

If you can still put mail in it, tag it as usual, using the ref of the aperture that's still open. If you're only supposed to post franked mail in it, mark it as a box for franked mail.

You can always survey them anyway. GPS traces are generally better than OSSV, particularly if you have a decent GPS unit.

Map trace: osm.org/go/euHqmkeTv-
GPS trace: osm.org/go/euNMGzaG7-

Any chance of providing a key for the maxspeed map? Seeing lots of random colours, mostly dashed red. These are now gone, since they were in UK.

Can we trace, and if so how do we set that up? A little more comfortable than eyeballing it - especially for astigmatics such as myself. (Not that it's stopped me eyeballing features before ...)