Following on from my previous post about Wikispecies (which generated some discussion on TAXACOM) I've played some more with Wikispecies.
AS a first step I've added a Wikispecies RSS feed to my list of RSS feeds. This feed takes the original Wikispecies RSS feed for new pages (generated by the page Special:NewPages) and tries to extract some details before reformatting it as an ATOM feed. Specifically, I extract GUIDs such as IPNI and Index Fungorum identifiers, bibliographic references (which I will later parse to try and extract identifiers such as DOIs), and latitude and longitude if the Wikispecies page has type locality information. Having the later means that the RSS feed can be displayed as a map (Google Maps can take a RSS feed with geotagged items and display it on a map for you).
The map below is live, so it will show any geotagged items in the current Wikispecies feed.
View Larger Map
Showing posts with label GeoRSS. Show all posts
Showing posts with label GeoRSS. Show all posts
e-Biosphere Challenge: visualising biodiversity digitisation in real time
e-Biosphere '09 kicks off next week, and features the challenge:
Originally I planned to enter the wiki project I've been working on for a while, but time was running out and the deadline was too ambitious. Hence, I switched to thinking about RSS feeds. The idea was to first create a set of RSS feeds for sources that lack them, which I've been doing over at http://bioguid.info/rss, then integrate these feeds in a useful way. For example, the feeds would include images from Flickr (such as EOL's pool), geotagged sequences from GenBank, the latest papers from Zootaxa, and new names from uBio (I'd hoped to include ION as well, but they've been spectacularly hacked).
After playing with triple stores and SPARQL (incompatible vocabularies and multiple identifiers rather buggers this approach), and visualisations based on Google Maps (building on my swine flu timemap), it dawned on me what I really needed was an eye-catching way of displaying geotagged, timestamped information, just like David Troy's wonderful twittervision and flickrvision.com. In particular, David took the Poly9 Globe and added Twitter and Flickr feeds (see twittervision 3D and flickrvision 3D. So, I took hacked David's code and created this, which you can view at http://bioguid.info/ebio09/www/3d/:
It's a lot easier to simply look at it rather than describe what it does, but here's a quick sketch of what's under the hood.
Firstly, I take RSS feeds, either the raw geoFeed from Flickr, or from http://bioguid.info/rss. The bioGUID feeds include the latest papers in Zootaxa (most new animal species are described in this journal), a modified version of uBio's new names feed, and a feed of the latest, geotagged sequences in GenBank (I'd hoped to use only DNA barcodes, but it turns out rather few barcode sequences are geotagged, and few have the "BARCODE" keyword). The Flickr feeds are simple to handle because they include locality information (including latitude, longitude, and Yahoo Where-on-Earth Identifiers (WOEIDs)). Similarly, the GenBank feed I created has latitude and longitudes (although extracting this isn't always as straightforward as it should be). Other feeds require more processing. The uBio feed already has taxonomic names, but no geotagging, so I use services from Yahoo! GeoPlanet™ to find localities from article titles. For the Zootaxa feed that I created I use uBio's SOAP service to extract taxonomic names, and Yahoo! GeoPlanet™ to extract localities.
I've tried to create a useful display popup. For Zootaxa papers you get a thumbnail of the paper, and where possible an icon of the taxonomic group the paper talks about (the presence of this icon depends on the success of uBio's taxonomic name finding service, the Catalogue of Life having the same name, and my having a suitable icon). The example above shows a paper about copepods. Other papers have a icon for the journal (again, a function of my being able to determine the journal ISSN and having a suitable icon). Flickr images simply display a thumbnail of the image.
What does it all mean? Well, I could say all sorts of things about integration and mash-ups but, dammit, it's pretty. I think it's a fun way to see just what is happening in digital biodiversity. I've deliberately limited the demo to items that came online in the month of May, and I'll be adding items during the conference (June 1-3rd in London). For example, if any more papers appear in Zootaxa, or in the uBio feeds I'll add those. If anybody uploads geotagged photos to EOL's Flickr group, I'll grab those as well. It's still a bit crude, but it shows some of the potential of bringing things together, coupled with a nice visualisation. I welcome any feedback.
Prepare and present a real-time demonstration during the days of the Conference of the capabilities in your community of practice to discover, disseminate, integrate, and explore new biodiversity-related data by:
- Capturing data in private and public databases;
- Conducting quality assurance on the data by automated validation and/or peer review;
- Indexing, linking and/or automatically submitting the new data records to other relevant databases;
- Integrating the data with other databases and data streams;
- Making these data available to relevant audiences;
- Make the data and links to the data widely accessible; and
- Offering interfaces for users to query or explore the data.
Originally I planned to enter the wiki project I've been working on for a while, but time was running out and the deadline was too ambitious. Hence, I switched to thinking about RSS feeds. The idea was to first create a set of RSS feeds for sources that lack them, which I've been doing over at http://bioguid.info/rss, then integrate these feeds in a useful way. For example, the feeds would include images from Flickr (such as EOL's pool), geotagged sequences from GenBank, the latest papers from Zootaxa, and new names from uBio (I'd hoped to include ION as well, but they've been spectacularly hacked).
After playing with triple stores and SPARQL (incompatible vocabularies and multiple identifiers rather buggers this approach), and visualisations based on Google Maps (building on my swine flu timemap), it dawned on me what I really needed was an eye-catching way of displaying geotagged, timestamped information, just like David Troy's wonderful twittervision and flickrvision.com. In particular, David took the Poly9 Globe and added Twitter and Flickr feeds (see twittervision 3D and flickrvision 3D. So, I took hacked David's code and created this, which you can view at http://bioguid.info/ebio09/www/3d/:
It's a lot easier to simply look at it rather than describe what it does, but here's a quick sketch of what's under the hood.
Firstly, I take RSS feeds, either the raw geoFeed from Flickr, or from http://bioguid.info/rss. The bioGUID feeds include the latest papers in Zootaxa (most new animal species are described in this journal), a modified version of uBio's new names feed, and a feed of the latest, geotagged sequences in GenBank (I'd hoped to use only DNA barcodes, but it turns out rather few barcode sequences are geotagged, and few have the "BARCODE" keyword). The Flickr feeds are simple to handle because they include locality information (including latitude, longitude, and Yahoo Where-on-Earth Identifiers (WOEIDs)). Similarly, the GenBank feed I created has latitude and longitudes (although extracting this isn't always as straightforward as it should be). Other feeds require more processing. The uBio feed already has taxonomic names, but no geotagging, so I use services from Yahoo! GeoPlanet™ to find localities from article titles. For the Zootaxa feed that I created I use uBio's SOAP service to extract taxonomic names, and Yahoo! GeoPlanet™ to extract localities.
I've tried to create a useful display popup. For Zootaxa papers you get a thumbnail of the paper, and where possible an icon of the taxonomic group the paper talks about (the presence of this icon depends on the success of uBio's taxonomic name finding service, the Catalogue of Life having the same name, and my having a suitable icon). The example above shows a paper about copepods. Other papers have a icon for the journal (again, a function of my being able to determine the journal ISSN and having a suitable icon). Flickr images simply display a thumbnail of the image.
What does it all mean? Well, I could say all sorts of things about integration and mash-ups but, dammit, it's pretty. I think it's a fun way to see just what is happening in digital biodiversity. I've deliberately limited the demo to items that came online in the month of May, and I'll be adding items during the conference (June 1-3rd in London). For example, if any more papers appear in Zootaxa, or in the uBio feeds I'll add those. If anybody uploads geotagged photos to EOL's Flickr group, I'll grab those as well. It's still a bit crude, but it shows some of the potential of bringing things together, coupled with a nice visualisation. I welcome any feedback.
H1N1 Swine Flu TimeMap
Tweets from @ attilacsordas and @stew alerted me to the Google Map of the H1N1 Swine Flu outbreak by niman.
Ryan Schenk commented: "It'd be a million times more useful if that map was hooked into a timeline so you could see the spread.", which inspired me to knock together a timemap of swine flu. The timemap takes the RSS feed from niman's map and generates a timemap using Nick Rabinowitz's Timemap library.
Gotcha
Although in principle this should have been a trivial exercise (cutting and pasting into existing examples), it wasn't quite so straightforward. The Google Maps RSS feed is a GeoRSS feed, but initially I couldn't get Timemap to accept it. The contents of the <georss:point> tag in the Google Maps feed looks like this:
Turns out there's a minor bug in the file timemap.js, which I fixed by adding
Postscript
Nick Rabinowitz has fixed this bug.
Ryan Schenk commented: "It'd be a million times more useful if that map was hooked into a timeline so you could see the spread.", which inspired me to knock together a timemap of swine flu. The timemap takes the RSS feed from niman's map and generates a timemap using Nick Rabinowitz's Timemap library.
Gotcha
Although in principle this should have been a trivial exercise (cutting and pasting into existing examples), it wasn't quite so straightforward. The Google Maps RSS feed is a GeoRSS feed, but initially I couldn't get Timemap to accept it. The contents of the <georss:point> tag in the Google Maps feed looks like this:
<georss:point>
33.041477 -116.894531
</georss:point>
Turns out there's a minor bug in the file timemap.js, which I fixed by adding
coords= TimeMap.trim(coords); before line 1369. The contents of the <georss:point> taginclude leading white space, and because timemap.js splits the latitude and longitude using whitespace, Google's feed breaks the code.Postscript
Nick Rabinowitz has fixed this bug.
Subscribe to:
Posts
(
Atom
)