Commons:Wikimaps/Wikimaps plan for 2015

These are initial thoughts about what the Wikimaps project would deal with in the coming year. Participate in shaping the scope. Here are our thoughts.

Wikimaps possible themes and projects for 2015 These are initial thoughts about all of the things what the Wikimaps project could deal with in the coming year. Participate in shaping the scope. Here are our thoughts. Please comment.

Tools

Collaboration Wikimaps & OpenHistoricalMap

The Wikimaps and the OpenHistoricalMap projects have been working closely together. Could we launch common projects?

Comments or participants:

Commons – Images – Scanned maps -axis

Warper

In 2014 our goal is to recreate the control point interface, and fit it in a new overall site structure and layout. In 2015 we would deal with items that we have not yet had the time to focus on: Making map sets, mosaicing, editing and adding metadata. Batch operations for adding already georeferenced images from repositories.

Comments or participants:

• Comment: British Library and NYPL maps already exist on Commons, which have geo-referencing in their home libraries. (See Category:Maps with links to external georeferencing, NYPL maps now identified and ready to add just as soon as I get the bot permissions). We could start porting those to WMlabs and the warper within a matter of days. Warping of maps already on Commons in large numbers could also start as soon as desired. What (if any) remaining things should do we need to think about, before we make it so? Jheald (talk) 09:53, 18 August 2014 (UTC)
  • Experience from the British Library suggests that announcing "batches" of maps ready to georeference, and then tracking progress on the batch in real-time, is a good way to build interest -- better than simply an open-ended process. Could we be ready to go with all of the old maps currently on Commons as a first "batch", ideally quite soon? Jheald (talk) 10:32, 18 August 2014 (UTC)
  • WMUK is planning a campaign starting with an all-day launch event at the British Library on October 31st (RSVP), to try to identify and Flickr-tag the remaining maps in the Mechanical Curator collection, using the index to the collection that we have been building here on Commons. There could be up to 10,000 of them. All advice or help or participation with this event would be very welcome, including any introductions to anyone who could build support in the OSM / historical SM community to get involved (to do this is going to need a crowd). After the launch event, the list will be live on Commons for Wikimedians anywhere in the world to work through. As soon as all the maps have been identified and tagged, the BL will then launch a new round of geo-referencing based on them, which we will then be able to batch copy to here. Jheald (talk) 11:03, 18 August 2014 (UTC)
• Comment: Humphrey of University of Portsmouth has talked about getting his full UK Ordnance Survey mosaic from the early 1900s onto our tile-server as a layer we could warp against (as well as some other goodies). User:Kolossos and he were talking about it, and it sounded like a real runnner. Any update on this? Jheald (talk) 09:57, 18 August 2014 (UTC)
• Comment: For compatibility with OSM-based projects, we may need to be a bit more careful to record exactly which map layers we're referencing against. In particular, referencing against Google maps may be a problem for some re-users. Jheald (talk) 09:59, 18 August 2014 (UTC)

Map search

The Wikimaps project pilots a maps search. The search should at least cover the maps at Wikimedia Commons, but it would be preferable to be able to search and rectify any openly available map, and save it in Commons. Realizing the search requires synchronization between the Commons Wikibase, Wikidata and Wikimedia search. Must involve key technologists in the projects. Evaluate and propose technologies.

Comments or participants:

• Comment: It is possible to index the georeferenced maps from WikiMaps in the http://OldMapsOnline.org project for free - see http://project.oldmapsonline.org/contribute. Klokantech

Aerial imagery

Working with aerial images requires many questions to be tackled.

• Storage for imagery, both hobbyist and GLAM
• Rectification requires freely available elevation models
• The control point interfaces may be shared, using different processing

Because of the specialized nature of the project, it would require dedicated funding + project frame.

Comments or participants:

Wikimedia Commons / Wikipedia access points

Wikimedia Commons is going through changes that will affect the whole work flow. The media viewer will be the access point to all metadata and tools for an image, but none are implemented yet. Image metadata will be stored as structured data, but only gradually over a period of maybe several years. How do we work now? We need to see if we contribute directly to that development, or if we design our work flow to best fit our needs, and offer the results as guidelines for the development.

Comments or participants:

• Comment: Not entirely. The media viewer will not be the access point to all metadata and tools. Current revision plans now place the Media viewer essentially as an alternative way to view elements of a Wikipedia page, with primarily just the information on that Wikipedia page being re-exposed. For more detailed information the user will be prompted to go to a separate File page view, essentially the present file page.

Given the complexity and depth of some per-image data stored on some filepages (eg the templated data for images of aircraft), the file page should be assumed likely to remain much as it is for the foreseeable future. Information that relates to real-world entities that fulfill d:Wikidata:Notability will increasingly be drawn from Wikidata - eg information about underlying books, paintings, map-series that are the subject of the image; also information about original creators, where these are notable. Template development work will increasingly be done to present these on file pages. Other, per-image information will not be stored on Wikidata. Some will be stored on the Commons Wikibase. However, initially this may be restricted to a simplified machine-readable licensing summary, and a cloud of "topics" (ie things with Wikidata entries that somehow relate to the image), which will be used to enable combinatorial tag-like searching.

See also this essay, c: Commons:Wikidata/How GLAMs can help the Structured Data for Commons initiative, and a newly formed user-community on wikidata, d:Wikidata:WikiProject Structured Data for Commons for piloting such templates, and for discussing aspects of what should be stored where, and how. The Foundation's thoughts about what will be stored on Commons Wikibase are still very much in the air, and, as you say, this is something where they are expecting an informed community to come forward and help. Jheald (talk) 10:21, 18 August 2014 (UTC)

Geolocating images and documents

Expanding the Wikimaps tools to place photographs by pinning, control points and 3D placement by web interfaces and mobile augmented reality apps. The control point interface may be used on oblique aerial images, and possibly also some outdoor photography.

Comments or participants:

OpenHistoricalMap – Vector maps – data -axis

Vector rendering for historical maps

Historical map tiles need to be available for any time or period, rather than pre-rendered as bitmap tiles for selected points in time.

Comments or participants:

Wikimaps Gazetteer

Place names can be extracted from maps with the help of suitable interfaces. This is naturally part of the OpenHistoricalMap vectorization process. Wikidata has the flexibility to store name changes and variations, and the data exchange between Wikidata and the OpenHistoricalMap should be tackled. We must make sure that there exist suitable properties for expressing cultural variants as well as temporal changes for place names in Wikidata. Tools for extracting names and features should be made in OHM.

Comments or participants:

• Comment: Wikidata gazetteer question: There will come a scale of resolution at which gazeteer items will fail d:Wikidata:Notability. Is there any thought to start running an additional WikiBase, perhaps with the same properties Pnnn as WikiData, but with its own namespace, perhaps Gnnnn for high-resolution geographical features? Such a WikiBase could also store boundary information (maybe in a purpose-optimised type of list data type). Jheald (talk) 09:42, 18 August 2014 (UTC)
• Comment: There has also been some talk of seeing how much we could do with Humphrey at University of Portsmouth, to see how much of his existing highly developed UK historical gazetteer we might be able to port over. Any progress information on that would be useful. Jheald (talk) 09:44, 18 August 2014 (UTC)

Crowdsourcing apps

Evaluating historical geographic data by checking it on a historical map

• Geolocating images
• Marking images as maps in Commons
• Georeferencing on mobile, on location

Collaboration with organizations who have tackled crowdsourcing and maps apps would be desirable.

Comments or participants:

Nordic & GLAM focus

Wiki Loves Maps

In 2014 the Wikimaps Nordic project will rehearse with hackathons about maps with experiences from #Hack4DK, #Hack4FI and Apps4Finland. Wiki Loves Maps can be a further elaborated event model, possibly taking advantage of the experiences of the New York Library Map Hack.

We would continue with locations from 2014:

• Helsinki - Aleksanterinkatu. Many GLAMs are collaborating on gathering materials about the location.
• Umeå
• Tartu

Comments or participants:

Local history

The geo approach requires broader support for geographic knowledge building. We must tackle the notability limitation for local knowledge. Does it require a new wiki project for local history or can Wikipedia be expanded with unverified information?

• Local history wikis
• Local history storytelling sites Platsr, Kultur og Naturreise

Piloting locality

• The Archipelago Sea between Finland and Sweden. The area is bilingual: Swedish and Finnish-speaking, and the historical materials can be found in Finnish and Swedish archives. There are connections to local communities, universities and GLAMs.
• Worldwide: Event model for a cultural historical expedition? Working with any locality.

Comments or participants: