[OSM-talk] OpenStreetMap Pledge

Tom Carden tom at tom-carden.co.uk
Tue May 30 22:36:51 BST 2006


On 30/05/06, Christopher Schmidt <crschmidt at crschmidt.net> wrote:
> On Tue, May 30, 2006 at 07:59:21PM +0100, Tom Carden wrote:
>>
>> Before the 10th of June, so you can show it at Where 2.0?  ;)
>
> So *you* can show it at Where 2.0.

I shouldn't have used a ;) there - it was a genuine question... why the
two week time limit on the offer?  To be honest, if money is just to speed
things up, I doubt it will work given that Steve will be at Reboot8 this
week, and that he has just moved house, and that he has his own things to
prepare for Where 2.0.  For better or worse, he's still the main
bottleneck when it comes to implementing serious server-side changes.

I won't be at Where 2.0, but I take it you meant OpenStreetMap and not me
personally :)

> I've already got a half dozen things
> to present, more than I have the presentational time, energy, or
> materials to support.

And so does Steve, of course.

>> Would planet.osm be adequate if the frequency was daily and not
>> monthly?  Steve - how long does it take to generate planet.osm at the
>> moment?
>
> Yes, but mikel told me that it would be impossible to generate
> planet.osm nightly.

Ah, OK.  That's why I was asking Steve.

> I would much prefer planet.osm data.
>

Cool.  So if we could efficiently generate monthly planet.osm, and daily
diffs, then that format would be OK.

>> People have been asking for a full db dump, that Steve hasn't done
>> yet, for whatever reasons, for ages now.  I think we await OSMF, and
>> lawyers, and proper privacy policy, before distributing the data in
>> full.
>
> I'm not asking for that.

OK, but it's what Mikel suggested in this thread, that's why I mentioned it.

>> How long does it take you to generate tiles from planet.osm now?  Or
>> do you generate them on the fly?  (And if so, why can't it be done on
>> OSM's servers on the fly?).
>
> They're generated on the fly. Why OSM can't generate accurately
> projected image tiles on the OSM server on the fly in a reasonable
> amount of time that I've never gotten a reasonable answer to.

It wasn't a rhetorical question.  I meant why can't *your* implementation
run on OSM's servers?

> I don't care what anyone does with it: I've been told that the methods I
> want to use for rendering maps don't work for OpenStreetMap.

What methods are those, and where were you told that?  We had a long
discussion a while back but that was all to do with mixing/comparing
GoogleMaps and OSM - nothing to do with rendering.  Several people are
working on their own rendering code, some of it on openstreetmap.org -
what makes your way different?

>> Does OpenLayers have a place to link to data providers, like
>> Google/Yahoo/MS do in their APIs?  That would be essential IMO.
>
> You can certainly create one.

I'm sure I can.  But you were offering to host and create one.  I was
suggesting that if/when you do, you make sure to link to openstreetmap.org
so that people can be made aware that what they are viewing can be edited
by them.

>> >under the same license as OSM.
>>
>> Naturally, since you wouldn't have a choice ;)
>
> My point was that I'm not looking for special treatment from OSM:

> I'm
> just looking for a useful data export.

Sounds like you're asking to jump the queue of requests in exchange for
money, with a time limit.  I'm not a fan of that way of doing things, but
that's my personal opinion - I don't speak for OpenStreetMap; currently
only Steve can do that.

> Persons have expressed concerns
> in the past about Wikipedia accepting large donations, and for OSM, $500
> is relatively large:

It's the first publicly disclosed donation I'm aware of.  Haven't a clue
how large it is in relation to other donations, since Steve's paypal
account isn't public.

> there is no concern that I'm attempting to buy out
> OSM,

I never implied there was.  There might be now you've mentioned it :)

> I just want something that I can do something useful with.

The usefulness of OSM is undeniable on a local level as others have
demonstrated independently (Weybridge, Sutton Coldfield, etc.) with far
fewer extra demands on Steve's time.  On a global/worldwide level as you
are requesting, I slightly question how useful it really is right now, but
I /seriously/ question what the rush is.

Best,

Tom.






More information about the talk mailing list