fference between z18 and z19, apart from scale considerations, is
non-existent. Apart from taking storage requirements into account, you
could well switch to z19 today.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
hedge fill.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
On 1-4-2012 3:40, Paul Norman wrote:
Just be careful to not accidentally upload the .osm that indicate the
problems.
You should modify it to have upload='false' in there. Then JOSM will
discourage you from uploading that file.
See http://josm.openstreetmap.de/ticket/4043
d. At least not
before the overwhelming majority of existing objects are compatible with
the new tagging, or after some extensive time has passed to allow this
tag change to propagate.
[1] Has now been done, but not deployed yet.
--
Lennard
___
talk ma
train=yes/no, bus=yes/no, subway=yes/no, etc etc is so dreadful (in the
context of the rendering chain) that every time I read #2798 I feel the
urge to run away screaming.
http://trac.openstreetmap.org/ticket/2798
--
Lennard
___
talk mailing list
hen you go to
http://coastline.openstreetmap.nl/
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
date. They
are regenerated more often. At an educated guess this happens every few
weeks.
The coastline error checker slippy, however, is a bit under the weather.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org
http://trac.openstreetmap.org/ticket/1835
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
y had this reported to me somehow, some time ago, but forgot
about it. There was no trac ticket created either. Would be helpful if
you could do that.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
und the deleted node, undeleted it, and resumed the upload.
Nuking 600k nodes and then reuploading them (creating new nodes in the
db) is not the most elegant way of doing things, either. :-/
--
Lennard
___
talk mailing list
talk@openstr
way. The chances of something like this happening rise
immensely with time and the amount of floating nodes you're accumulating.
[1] "~600k nodes, ~4k ways, 280 relations"
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
stuff.
Can't speak at all about Merkaartor. Potlatch doesn't do it, but it seems
it's a feature just waiting for a developer.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
| | Brandweg
53085949 | cycleway | | |
25806811 | path | | | Tunnelstraat
22903417 | unclassified | | | Brandenstraat
(9 rows)
Time: 7.328 ms
--
Lennard
___
talk mailing
hat: more of this! Nice to see wacky little ideas come to live
and be more than a plain text listing of objects. :)
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
icial subparts
of municipalities. Even those subparts can consist of more than one
dwelling, to make it even more confusing.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
unt of the municipality not giving us their boundaries
yet. There's another mapper tracing these from official documents (less
accurate than receiving the original geo data, but alas), but he doesn't
seem to have done this yet.
--
Lennard
___
ta
s that got dropped from the queue are forgotten. They are not rendered
later on, when the queue isn't full.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
On 21-3-2011 22:59, Andy Robinson wrote:
I'd place the coastline at the low water mark because you know then that its
always true.
Do you know what happens just before a tsunami hits land?
--
Lennard
___
talk mailing list
talk@openstreetma
ver, and it probably needs less than that.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
nces these dates.
Can we get this going again? What exactly is needed?
A server with sufficient storage and bandwidth to process a weekly
planet, daily diffs, and many hours of single task processing per
coastline run.
--
Lennard
___
talk mailing
ds true of user
collected and contributed information in OSM, to various degrees. Users
loose interest, or worse (some have even died), and thus areas fall out
of maintenance all the time. Waiting for someone else to come along and
pick it up. -If- someone comes along.
=-47.760662707134124&zoom=6
http://ant.dev.openstreetmap.org/bingimageanalyzer/?lat=61.51534917347587&lon=19.306271732481257&zoom=7
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
On 10-2-2011 23:37, ant wrote:
On 10.02.2011 20:25, Lennard wrote:
@ant: Would it be possible to have the editors collect and report* on
the available zoom levels, as users download Bing tiles while editing?
That's a brilliant idea, but I'm not involved in how editors handle Bing
ma
opt-in and/or anonymised
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
t nothing supports yet, but they're a nice idea).
> I don't know of anything that doesn't treat them as synonyms.
Oh, the discussion has been going on and off for years now.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
d* member ways WITH
> admin_level=6 aren't rendered until z11
Both the *way* and the *relation* render z11+.
Aren't overlapping* boundary renderings fun? :(
* See my previous reply.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
> renderer so all the rules here apply:
> http://wiki.openstreetmap.org/wiki/Relation:multipolygon#Usage
Right, so why even bother with type=boundary when type=multipolygon will
do fine? :)
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
and
the bbox of the tile being rendered could overlap even for tiles far inland.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ryone else rendering Ireland in their mapnik
renderer stacks.
Also, I presume your indexes are good?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ty good in limiting those situations where people
are deleting 'unused' nodes during your upload.
I also once requested the same sorting functionality in JOSM. Kind of
fell on deaf ears, as the ticket hasn't been acted upon.
https://josm.o
caveat here is that the CLC 2006 dataset is not publicly available
through the EEA website.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ompile/reimport. Also remove the natural=coastline 'delete' keyword
from default.style. You'll then have the raw natural=coastline ways in
planet_osm_line, and you can create a mapnik rule for those.
--
Lennard
___
talk mailing list
talk@o
capture in a fixed path,
and fixed paths is all that OSM can offer, currently.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
actual object, here is a newer
editor that puts them on the objects again?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
3 is subnational.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
rest.
[1] http://en.wikipedia.org/wiki/Toponymy
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
PI any day.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
d
me currently.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
that time is
again high, they might get dropped again.
So if you look at the same busy time every day, and nobody is looking at
that area during quieter times, that could cause this symptom.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
wiki.openstreetmap.org/wiki/Upload.py
(4)
http://svn.openstreetmap.org/applications/utils/import/bulkupload/smarter-sort.py
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
elpfully' delete those 'duplicated nodes'. I just wish those
scripts would only process nodes that have been idle for a few days, and
not very fresh ones.
[1] http://josm.openstreetmap.de/ticket/4299
--
Lennard
___
talk mailing lis
their edits may be better than yours, as they're based on
survey.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
if you are the only one editing a certain area, it is still useful
to use changeset comments. If someone else starts to work in your area,
or vice versa, it's still useful to know what happened.
--
Lennard
___
talk mailing list
talk@openstreetma
Selection feature, it can already do everything
you describe above.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
osm.xml font section is attached below
What are those fontset sections doing in your osm.xml? Are you using an
old version of the stylesheet?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
t person who has run into this and
came here asking what was going on, and I fear you won't be the last
either.
Normally I would wholeheartedly recommend the Geofabrik extracts, but
alas, they don't seem to carry them for North America.
--
Lennard
__
ting your own stylesheet to use with the OSM input plugin. The
standard osm.xml stylesheet is tuned for postgis and will not work with
the OSM plugin.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
appropriate water tag, and you're done. It *is* water, surely. That it
is a marina, is like the distinction between a national park and the
fact it has wood, meadows, heath, etc within its boundaries.
--
Lennard
___
talk mailing list
tal
? It hadn't been rerendered as a living street yet. I just did
that, and it looks fine now.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
sylvain letuffe wrote:
> - you reach easily the API max members limit
There is no such limit (yet).
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
nterest in them.
>
> +1
Paging General Dreedle ...
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Nic Roets wrote:
> Is it just me or did the earthquake create some new land ?
> http://osm.org/go/ZNIhF
It's an error in the coastlines that was present during the last time
the coastline shapefiles were generated. We'll get some new ones after
this week's planet dump
R you generate with 20 km overlap. I always believed that
was the reason you weren't using the hypercube processed_p shapefiles.
The need for the larger overlap will soon be gone, with the release of
mapnik 0.7.0 and the gamma parameter.
--
Lennard
_
Lennard wrote:
> I will however deploy newer coastlines to http://haiti.openstreetmap.nl/
These have now been deployed, based on planet-100119.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
in OSM, and none can be created. Unless
avar finds a way around that too, actually. :)
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
y fixed, and it's still processing now.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
n van Oosterhout, as he's the author of the coastline error
checker.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Daniel Neugebauer wrote:
> I will look into Mapnik as well if there's also nothing in progress already.
There is no ticket, open or closed, for mapnik regarding veterinaries.
--
Lennard
___
talk mailing list
talk@openstreetmap.
look at the multitude of other POI
rules in osm.xml and its include files.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
verlays, they handle specific route rendering, and do not result
in the all-in-one tiles that OCM produces, but you're welcome to have a
look at them, and see what we've done.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ld
Most of the time, there really isn't.
> render, otherwise you get the names of postcodes and names rendering
> in the middle of no where.
Indeed, which is currently the case.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
. Putting it as 2 means it's 2 meters +- 1 m.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
n render farm, and start experimenting and
working out all the kinks and bugs, and keep it performant at the same
time. Only when you've achieved that, can we think about migrating this
to the main mapnik map. Or not, depending on then curr
main listing, but not any of the subdirs or the README
[03:35] does that still work through NFS?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
nik
> afaik, but that doesn't make it less useful.
Good luck getting the Cycle Map stylesheets.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Dave F. wrote:
> barrier=* means that any barrier tag should be rendered, yes?
>
> It's listed that mapnik should pick this up, but it doesn't.
It does, but only for ways, not areas. Steve's list doesn't make that
distinction.
--
Lennard
___
;s getting slightly annoying that this thread is on 3 mailing
lists, with 3 sets of answers, and an OP that doesn't appear to read the
answers.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ered the output to not show 'length'.
That's not a key that's pulled from OSM. It's generated internally. Same
goes for 'point', 'way_area' and 'z_order'
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
nature or by man.
* When hypercube isn't down or otherwise slow, which is quite often
these days.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ne glory.
I nominate this for the DUH! moment of the day awards. :D
http://en.wikipedia.org/wiki/Painter%27s_algorithm
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
already answered your own question. Having the node with
the barrier in the c-d road would make it also be a stile that is
blocking travel in that road.
I've used your 2nd tagging, with the node with the stile a small
distance away from the co
gives you a list of only way ids, 1 per row, with the first 'w'
character stripped. You still need some way to join the member role in
the results. I didn't get that far when I wrote the unnest function, as
I didn't need the role at the time.
--
Lennard
__
se or just treat them as synonyms for building=yes."
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ss the main landuse=agricultural
tag, and not try to group various agricultural landuse types into 1.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Matthias Versen wrote:
> A border-way should contain admin_level=X where X is the highest number
> of the border it represents and an boundary=administrative Tag.
Lowest
--
Lennard
___
talk mailing list
talk@openstreetmap.or
irection to take
for the A29. Clear and to the point.
"Take the next exit, follow the slip road A58-A29*, then follow the A29".
Exactly how would this be better? "A58-A29" won't be on signs, and it
tells me non-essential information that could
these should be rendered on the main maps, and
they should not be made up by us if there is no real ref on the ground.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
maxspeed, surface,
etc. We're already breaking it up for any number of reasons, at a border.
But, a bit of preprocessing could also be involved, so that we don't do
'tagging for the renderer', if that makes people feel better. :P
--
Lennard
__
but we have to consider
the non-relation variants as well.
> If it's just ref numbers changing that shouldn't need anything more
> than to split the way would it?
Indeed, but I'm thinking ahead here, to a situation where the style of
the road itself wou
Because these sections are so short, it's not
noticeable on a map, and you don't have "style bleed".
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
database everytime you import diffs, but only the changed ones.
One more issue I see is what to do with features that cross these
boundaries? Which style will/should they get? Should the import split
them into 2 features, each on one side of the boundary?
--
Lennard
_
so work for diffs and changed polygons), or
by doing this itself.
What were the other issues you had thought up?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
John Smith wrote:
> While my patch works, I don't know if this is the best solution to the
> problem or not:
>
> http://trac.openstreetmap.org/attachment/ticket/1666/osm.xml-patch
I commented on the ticket.
--
Lennard
___
talk
tion, it could've just as
easily been done by you or him by just deleting those objects.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Shaun McDonald wrote:
> It's really old news, that tile.openstreetmap.org is using the
> minutely diffs.
But Steve was out at a conference when we switched over to the new tile
server. Which is fast enough that it doesn't drop any render requests
any mo
nvention of =designated, it seems we all just agree to disagree.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
e. I
suppose editors could automatically add/remove the tag, but that just
gets me back to my first point: why?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Valent Turkovic wrote:
> Presets speedup many things, so it would be nice to have also presets for
> source= tag.
Then create your own presets for your desired source=* tags.
--
Lennard
___
talk mailing list
talk@openstreetmap.or
can type, so I don't see
why you'd say it doesn't support source=*. You're not always, every
time, relying only on built in presets, are you?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
Kenneth Gonsalves wrote:
> select node from planet_osm_point where
> golf='green' as golfmarkers
>
> but the flag is not showing. What could be wrong?
select way from planet_osm_point
--
Lennard
___
t see that.
>
> Any ideas why?
Yes. It was rendered as a tunnel, but then overwritten by the non-tunnel
variant.
I've committed a fix, which will probably show up on the map in a few days.
--
Lennard
___
talk mailing list
talk@openstreetm
John Smith wrote:
> Apparently there is a 1000 member per relation limit.
Not true.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
ifferent for every single street).
ZIP code systems are not always of the "one number for a big block of
houses/streets" variety.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
od_tile to synch it with the new mapnik
- Clear your cache, restart apache
- Try again
To all that interests this: don't rely on tools included by your linux
distribution to be up to date.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
ht
Peter Körner wrote:
>> It reaches 100% when /Tromelin Island /is set to not-ok.
I got Bahasa Indonesia at 229/230 with 2 countries (Tromelin, Turkey) as
not ok.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetm
hs rather than years.
Very good progress has been made recently on offset rendering for the
LineSymbolizer. I'm indeed hoping we can use this something this year.
http://trac.mapnik.org/ticket/180
--
Lennard
___
talk mailing
Ian Dees wrote:
> How would I go about doing that? I suppose I could use curl or something
> similar, but is there a better way?
If you can run perl scripts, have a look at:
http://trac.openstreetmap.org/browser/applications/utils/revert
--
L
Why did you tag all those nodes with landuse=cemetery?
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
do that as well.
--
Lennard
___
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
> Lennard wrote:
> The general solution is "maxspeed is the highest of the maxspeeds of all
> classes of vehicle on that road".
> See also the signs we have in continental europe when you enter a country:
> there is usually a large sign specifying the maximum speeds on di
1 - 100 of 133 matches
Mail list logo