[Historic] the "spherical mercator" of time?

[Jeff Meyer]

Aaron - do you mean Carden and Migurski or another Tom &/or MIke?

Dan - your demo is cool, but is that as fast as it can go? ; )

Mikel - are you thinking of the tiles being dependent on just physical
characteristics of the earth's surface or also of animal-imposed activity
(e.g. national borders like Dan depicts, or paths of explorers, battles,
changes in roads, street names, animal habitats, etc.).

For some time ranges (e.g. pre-industrial era), it would seem that decade
or century-wise (or never at all) updates to shorelines, etc. would be
fine. After that, it would seem that man's influence on the shape of
shorelines and other terraforming (e.g. the Netherlands, Venice, Seattle
after 1860, SF, Boston, Dubai, Krakatoa, canals, you name it...) would make
a higher resolution approach more interesting. That said, the relative % of
tiles requiring updates over these periods would seem pretty small.

- Jeff


On Thu, Aug 23, 2012 at 8:35 AM, Dan Vanderkam <danvdk at gmail.com> wrote:

> Depending on the situation, we may be saved by the fact that most features
> on a historical map won't change that frequently. For instance, in my
> countries of the world demo (* see below), I used two layers: a tiled
> geographical layer (which never changes) and a country boundaries layer
> (which rarely changes on short timescales). Compositing features
> client-side becomes a more compelling approach with historical maps, since
> the features can change or remain the same independently of one another.
>
>   - Dan
>
> * http://danvk.org/whm/map.html. Disclaimers: this works best in Chrome.
> You can pan with the mouse and use the scroll wheel to zoom. It uses
> unlicensed commercial data, so please don't share the link.
> On Aug 23, 2012 5:54 PM, "Aaron Straup Cope" <aaronofmontreal at gmail.com>
> wrote:
>
>> It's definitely a thing that's been talked about in the past at Stamen.
>>
>> If Mike and/or Tom aren't on the list already we should drag them in.
>>
>> On 8/23/12 10:46 AM, Mikel Maron wrote:
>>
>>> One thing I'm pondering is how to efficiently and easily share time
>>> aware geodata. Prior to Google gifting the world spherical mercator
>>> tiles, web maps were rendered on request, based bounding box,
>>> dimensions, projection. WMS was slow. Naturally I'd expect we'd use
>>> cached tiles for serving time data, but I don't see an obvious way of
>>> this not exploding to all arbitrary times in history. Perhaps enough to
>>> begin by expanding the tile template to /{time}/{zoom}/{x}/{y}, and
>>> maybe limiting time to year or decade or century. Is there something
>>> more clever? What ways do historians bring methodical order to the times
>>> and eras in history that are of most common interest and utility?
>>> * Mikel Maron * +14152835207 @mikel s:mikelmaron
>>>
>>>
>>> ______________________________**_________________
>>> Historic mailing list
>>> Historic at openstreetmap.org
>>> http://lists.openstreetmap.**org/listinfo/historic<http://lists.openstreetmap.org/listinfo/historic>
>>>
>>
>>
>> ______________________________**_________________
>> Historic mailing list
>> Historic at openstreetmap.org
>> http://lists.openstreetmap.**org/listinfo/historic<http://lists.openstreetmap.org/listinfo/historic>
>>
>
> _______________________________________________
> Historic mailing list
> Historic at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/historic
>
>


-- 
Jeff Meyer
Global World History Atlas
www.gwhat.org
jeff at gwhat.org
206-676-2347
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/historic/attachments/20120823/46f72d78/attachment.html>