[OHM] : TimeSlider - calling attention to early prototype

[Tim Waters]

Hello Karl,

welcome!

On 19 May 2014 14:57, Karl Grossner <karlg at stanford.edu> wrote:

> Hello all,
>
> I've just now become aware of the OHM effort (hi Susanna) and very
> interested to learn what the development goals are. Is there a link to some
> material laying them out?
>

We have a few wiki pages:
http://wiki.openstreetmap.org/wiki/Open_Historical_Map  and this mailing
list.


> For example, is the plan to provide the means for loading copyright-free
> scans of historic maps and an editor for digitizing their contents?
>

There are tools available to handle this - for example mapwarper.net of
which I am the developer of. And there's also the sister project from the
Wikimedia Foundation - Wikimaps http://wikimaps.wikimedia.fi/ of which one
part plans to use the Wikimedia Commons with their own warper system to
store, annotate and load up historic maps.

 > and an editor for digitizing their contents?

The editor is this very project: OpenHistoricalMap.org :-)

I can envisage it being made easier to, given a WMS or Tile endpoint from
an OOC historical map to more easily digitize in OHM. We use the
OpenStreetMap tools, but our work flow is probably going to be a bit
different in some places.


> My interests in this include both the spatial and temporal, and the
> joining of those two. A couple of things that might interest: I've done
> some work with colleague Elijah Meeks on representing historical time (
> http://dh.stanford.edu/topotime), and there is a discussion under way
> right now in GeoJSON world about adding a "when" object to the forthcoming
> GeoJSON-LD standard (several threads, listed here:
> https://github.com/geojson/geojson-ld/issues).
>
> I'm very interested in the prospects for developing over time a global
> historical atlas that includes vector roads and rivers along with cities
> and boundaries. There are a few schemes aiming at such a thing, therefore
> many people talking about similar issues but in different conversations.
> One thing about digitizing is its so time-consuming, getting the "right"
> encoding scheme down beforehand becomes really important. Not sure how, but
> merging the discussions somehow makes sense.
>

OHM can take in roads, rivers and anything else that used to exist!  It
would be great to see the course of a river changing over time, I know
there's a fair bit of interest in this within the community.

OHM is evolving in a very OpenStreetMap way - adhoc, community, itch
scratchy way. OpenStreetMap tended to avoid worrying about getting the
right schema - in favour of a "folksonomy" (With GIS databases, you set the
schema first, and then fill in the data, with OSM you fill in the data and
make up the schema as you go) Basically it's better to get stuff in the map
first and then worry about how to deal with it - I think it's an approach
that really works. What that probably means is that the really interesting
brain stimulating stuff with vague dates, and fuzzy locations and
geographical ambiguity won't be solved before the easy stuff is added and
working first.

You are correct in that it is time consuming. However in my mind:
 1) They said the same thing 10 years ago regarding OpenStreetMap and look
at it now :-)
You are never working alone. Crowdsourcing best practices, Varying sizes of
work, durations and community are absolutely key here. Keeping things easy
and simple are key in my opinion - otherwise it will be boring. But it's
not impossible.
 2) There's the potential to use automated vectorisation for some datasets
(See recent discussion on this list)
 3) There's a huge untapped knowledge base of amateur historians, local
experts, geographers, databases, archives, libraries and museums all around
the world.  It's exciting times!

Cheers and welcome again!

Tim

>
> best
> Karl
>
> --------------
> Karl Grossner
> Digital Humanities Research Developer
> Stanford University Libraries
> Stanford,CA US
> www.kgeographer.org
>
>
> ------------------------------
>
>
>
>> ----------------------------------------------------------------------
>>
>> Message: 1
>> Date: Wed, 14 May 2014 09:19:51 -0300
>> From: Rob H Warren <warren at muninn-project.org>
>> To: historic at openstreetmap.org
>> Subject: [OHM]  TimeSlider - calling attention to early prototype
>> Message-ID: <E1B5BD07-1BB0-4947-9C8C-004E29012AF1 at muninn-project.org>
>> Content-Type: text/plain; charset=us-ascii
>>
>>
>> Figuring all of this is something that will take place in a few
>> iterations. Things to keep in mind: The Gregorian calendar starts around
>> 1582 and anything before is only really valid as a year.
>>
>> The interface between the slider and rendering was originally meant to be
>> an add-on to the tile renderer with something like startDate, endDate being
>> part of the tile URL.
>>
>> Tim, nice work with the extra layers. What do you think of using TIME-OWL
>> style named time periods to deal with things like "Roman Era" for
>> "historic" tags? This avoids splitting hairs when rendering with dates and
>> help people tag things properly. We can borrow some previous time data from
>> [1].
>>
>> rhw
>>
>> [1] hypermedia.research.southwales.ac.uk/kos/star/time-periods/
>>
>>
>>
> _______________________________________________
> Historic mailing list
> Historic at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/historic
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/historic/attachments/20140519/2efc34d3/attachment.html>