Commons:Wiki Loves Monuments 2011/Tools

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

For Wiki Loves Monuments 2011 tools are being developed for various tasks.

General guidelines[edit]

All tools should follow these general guidelines

  • The tool should be open source.
  • The source should be published.
  • Keep internationalization in mind.
  • Preferably the tools should be developed at the toollabs

Current tools[edit]

Tools related to the current project.

Monuments lookup[edit]

Since monuments tend to be numbered in groups based on location, this quick lookup method is handy for finding out if specific or nearby monuments are already uploaded to Commons with a proper monument number (this lookup has been particularly useful for discovering not-yet-categorized pictures on commons for use in articles about monuments). Sample monument lookup:

http://commons.wikimedia.org/w/index.php?title=Category:Rijksmonumenten_with_known_IDs&from=+00019237

Monuments-map/Dutch monuments in OpenStreetMap[edit]

Based on the rijksmonumenten tool a proof of concept was made to display all the monuments currently in the database on a map[dead link].

On this map the icons for monuments that have a photo are small, the icons where a photo is missing are large. As background map is OpenStreetMap on client side runs OpenLayers, so it runs in each browser and can also include as iframe in other websites.

It's hacky and needs an i18n. It should be replaced with the search and export tool.

WMCH[edit]

http://www.wikilovesmonuments.ch

Database browsing[edit]

URL: http://wlm.wikimedia.pt/tools/plist/

  • Allows coordinate insertions and corrections to be made;
  • Statistics: the aim is to monitor evolution of coordinate, image, and article coverage before, during, and after the contest. It's built on a schema extensions (and data collected throught ETL to gather a lot of information), so it might not work immediately for other chapters. Mail me directly if you'd like to see it on yours.
  • Boxed Google maps rendering ([1])

Csv to template[edit]

With csv2template you can easily make structured lists. Multichill (talk) 14:57, 16 May 2011 (UTC)[reply]

Display nearby monuments[edit]

Augmented reality app for mobile phones[edit]

Bots[edit]

  • semi-automatically add coordinates from monument list to Wikipedia articles
  • Make a gallery of unused photos so people can add them to monument lists
  • Bot to move images from commonsCategoryBase to subcategories based on the monument template and Commonscat links at the Wikipedia

Austrian tools[edit]

If anyone could use one of these features just contact me :) --AleXXw 23:40, 12 December 2011 (UTC)[reply]

Duplicate IDs in wikilists[edit]

Spain[edit]

Different lists at http://toolserver.org/~platonides/wlm/[dead link] (by status, user, province, id...). Used its own categorization bot.

Wanted tools[edit]

Search and export tool[edit]

The monuments database is expanding rapidly, but we don't have a good and convenient way to get information from this database. The tool should consist of two parts: Searching and output format.

Api / Code: https://fisheye.toolserver.org/browse/erfgoed/api

Searching: It should be easy to make a selection of the monuments
  • Search by location
    • bounding box
    • distance (for example monuments in 10 km radius, find nearest monument)
  • Search for monuments without an image
  • (add more requirements here)
  • Search by text fields
    • select by country, municipality
    • search by whole name, match a string in name
  • Text search & location search combined
Output formats: The tool should be able to output the results in different formats
  • Static or dynamic KML for Google Earth and Google Maps
  • Gpx for you GPS
  • TomTom POI format
  • Layar format
  • HTML lists to print
  • Structured formats like JSON or XML so it can be used like the MediaWiki api (list of formats)

The tool should be available in multiple languages. The input side of the monuments database was written in python, this tool should probably be written in php. For the structured lists, some MediaWiki api code could be reused.

Overview of how the flow should be


Ideas on parameters[edit]

  • Common parameters (applicable to all orders)
    • props = [country|id|name|address|municipality|lat|lon|image|source|changed]: the properties which should be returned. By default all of them.
    • output = [kml|gpx|poi|html|layar|json|xml]
  • Specific orders action =
    • search: searches monuments containing the specified text OR "*" to show all monuments within the following paramaters:
      • srcountry = countrycode_(lang) OR "*"
      • srmunicipality = municipality
      • sraddress = address
      • srcoortinates = [yes|no|both]
      • srimage = [yes|no|both]
      • srinrange = lat, lon, range
    • statistics: gives statistical data, of a given country
      • stcountry = countrycode_(lang) OR "*"
      • stprops = [withimage|withcoordinates|noimage|nocoordinates|add more]
      • stformat = [percent|number|both]

Please note: Just ideas. Feel free to improve, delete or whatever you want.

(Requirements in Dutch)

Possible classes, code snippets to use[edit]

Helpful texts[edit]

Android app[edit]

WLM Android app

See also a augmented reality app for iPhone & Android, from NL 2010: http://rijksmonumenten.info/

Matching images from Commons to monuments[edit]

A tool that displays in Google maps or in Google Earth

  • 1) monuments
  • 2) all geolocated images in Commons in 1 km radius of those monuments which are not yet tagged with a monumentID template in Commons
  • + this tool lets bind those images with monuments -- drag and drop image onto monument location icon would be nice

... as a result image on Commons get's monumentID template added to it

Flickr import bot[edit]

Below is a list of all the known Flickr repositories and the information needed to customize the bot

Language Group Monument ID Templates Categories Status
ro http://www.flickr.com/groups/wlm2011ro/ We're using a tag the contains the LMI code preceded or not by lmi= (some users put it, some don't). Regular expression:
((lmi=)?)([A-Z]{1,2})-(I|II|III|IV)-([a-z])-([A-Z])-([0-9]{5}(\.[0-9]{2})?)
{{Wiki Loves Monuments 2011|ro}}{{Monument istoric|CT-I-m-A-02567.}} Category:Historical monuments in Romania Strainu is uploading
nl http://www.flickr.com/groups/wlm2011nl/ We're using a tag the contains the RM code preceded or not by rm= (some users put it, some don't). Regular expression:?? {{Wiki Loves Monuments 2011|nl}}{{rijksmonument|12345}} Category:Rijksmonumenten Multichill is uploading
es http://www.flickr.com/groups/wikilmes/ We're using a tag the contains the BIC code preceded by BIC= {{Wiki Loves Monuments 2011|es}}{{BIC|RI-51-1234567}}{{BIC|RI-51-1234567-12345}} Category:Cultural heritage monuments in Spain Uploaded by Wiki-Bot. Code at svn.
ca http://www.flickr.com/groups/wikilovesmonumentscat We're using a tag the contains the BIC code preceded by BIC=, even if it's not BIC {{BIC}}{{Béns Andorra}}{{Mérimée}} Category:Cultural heritage monuments in Catalan Countries Joancreus is uploading

Simplified Commonist[edit]

See also Commons:Tools/Commonist

Mass upload for Wiki Loves Monuments has so far been a tough thing to do, especially for the new and inexperienced participants of the contest.

Wikimedia Foundation developers have done a great job on preparing & fixing UploadWizard and enabling UploadCampaigns that we used in September. Unfortunately, UploadWizard has a small but important flaw, which is, of course, its limitation to only 10 files that can be uploaded simultaneously, and the lack of an at least semi-automated upload method (preparing descriptions is a true pain).

I would like to ask someone to create a simplified version of Commonist or another easy-to-use, preferably Java-based (or at least Windows-friendly) mass upload tool for use by the participants of the national contests. Here are some proposed requirements and ideas for such a programme (hint: I am not a programmer, so feel free to suggest your own ideas!):

  • Would need to be as similar to the UploadCampaign as possible:
    • No need to choose the wiki (Wikimedia Commons set by default);
    • No need to choose the licence/permission (Creative Commons Attribution-ShareAlike 3.0 by default);
    • No need to insert the source (should be set as Own work by default);
    • No need to choose categories (will be dealt with by templates);
    • No need to insert coordinates (could be taken from EXIF data);

Instead:

  • The programme should allow to insert the ID of the monument;
  • It should have the WLM logo somewhere;
  • It should be as much localisable as possible;
  • More?

Possible features:

  • It should have some hard-coded (but localisable) information on the rules of contest or the licence;
  • An "info" tab (with a basic FAQ or some links) would be a plus.
  • Sexiness (i.e. easy-to-use and shiny) would be a plus.
  • More?

IRC channel[edit]

If you don't have an irc client you can use http://webchat.freenode.net


Wiki Loves Monuments 2011

Concept - Philosophy - Timeline - Progress - Monuments database - Tools - Action points - Upload wizard - Sponsors - Press - Highlighted pictures
Lists per country - May Meeting - European Heritage Days - Commons infrastructure - Structured lists
Andorra - Austria - Belgium & Luxembourg - Denmark - Estonia - France - Germany - Hungary - Netherlands - Norway - Poland - Portugal - Romania - Russia - Spain - Sweden - Switzerland