EOL API Discussion Group

With growing interest in using the Encyclopedia of Life API (application programming interface), comes the need for an openly available place to ask and answer questions. This is the place.

The Newsfeed for this EOL Community gathers updates associated with the items belonging to its Managed Collections, including activities of its members and comments from other EOL users.

Add a new comment

Newsfeed

  • Profile picture of Kristen Lans who took this action.
    Kristen Lans added "Mephitis mephitis (Schreber, 1776)" to the collection "The discussion collection".

    about 2 years ago

  • Profile picture of Roger Espinosa who took this action.

    Roger Espinosa commented on "EOL API Discussion Group":

    @Nathan Wilson: Hm; exact=1 looks like the key. I think what I'm looking for is more metadata about the classification on these results pages, to save having to ping each page to be sure it's information we can pull in. (For the moment, waving aside whether that's an issue we should be concerned with.)

    about 2 years ago

  • Profile picture of Tanya Dewey who took this action.

    Tanya Dewey commented on "EOL API Discussion Group":

    Here's the use case: use the EOL API to gain access to video shared with EOL and display on a video tab on ADW animal species accounts. I believe the concern is that the API (like EOL content in general) gives a user sometimes **very many** taxonomic name results for a single taxon. I think Roger is concerned about having to handle a huge taxonomic name mess as an api query result. Instead, if we could use particular hierarchies that we know match ours, then we would get only relevant results. I will clarify with him as well, but that is why we were hoping to find a way to parse content associated with taxon names from a single, or set of, hierarchies.

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "EOL API Discussion Group":

    @Nathan Wilson: Good point about VoLE. I read "This API gives access to a single hierarchy and its internal relationships" in the hierarchy entries documentation and thought it might have been talking about the EOL taxon concept hierarchy, not the alternatives.

    about 2 years ago

  • Profile picture of Nathan Wilson who took this action.

    Nathan Wilson commented on "EOL API Discussion Group":

    @Cyndy Parr: You can certainly walk the tree of any provider listed in provider_hierarchies.

    about 2 years ago

  • Profile picture of Nathan Wilson who took this action.

    Nathan Wilson commented on "EOL API Discussion Group":

    @Tanya Dewey: I think you are looking for the 'exact' option in search. For example, http://eol.org/api/search/1.0/Lama%20glama?exact=1 returns exactly one result which has an EOL page id of 309018. From that you can get the hierarchy entry id for Catalog of life with: http://eol.org/api/pages/1.0/309018 and finding the dwc:taxonID next to the dwc:nameAccordingTo that says 'Species 2000 & ITIS Catalogue of Life: Annual Checklist 2010'. This gives you 34296333, which you can then place in it's hierarchy started with: http://eol.org/api/hierarchy_entries/1.0/34296333 I'm not sure what is meant by 'getting knocked off'. VoLE seems to be working reasonably.

    about 2 years ago

  • Profile picture of Nathan Wilson who took this action.

    Nathan Wilson commented on "EOL API Discussion Group":

    @Cyndy Parr: As far as I know there is no way through the API to filter the content based on a specific hierarchy. It is not clear to me why that raises a concern about parsing multiple hierarchies. It does mean that you may get content returned for a high level taxon that is not associated with any of the descendants of that taxon in that hierarchy. This might lead to some confusion (birds showing up on a page for reptiles for example or images for taxa that aren't included in that specific hierarchy), but in most cases that's not going to be 'wrong'. I need to understand the actual use case better. It is important to realize that all of our content is associated with a classification that comes from that content provider, but it may be partial or even completely flat. So the only content clearly associated with a hierarchy like Catalogue of Life is content provided by them which will be very little. Most of the other connections between content and names are inferred roughly speaking through the 'pages' or 'taxon concepts'. Under the hood we actually do aggregate the data up the approved classification trees to populate our Hierarchy Entry pages and as far as I know that aggregation is not currently available through the API. However, it is not clear that would actually address the need in this case.

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "EOL API Discussion Group":

    I can't speak to the problems with service interruptions. I think I'm starting to see the problem -- looks like you can walk the EOL tree using http://eol.org/api/docs/hierarchy_entries. And you can get the root of the provider hierarchies, as well as specific responses for a page for a provider, but how do you walk the tree of a specific provider? I'm not sure. You might have to get the tree directly from them then query us for each name or provider-specific identifier.

    Regarding authorities in parentheses, there are specific rules about which are in parens: http://en.wikipedia.org/wiki/Binomial_nomenclature#Authority . Some providers include authorities while others don't. Does that help?

    about 2 years ago • edited: about 2 years ago

  • Profile picture of Tanya Dewey who took this action.

    Tanya Dewey commented on "EOL API Discussion Group":

    These are Roger's original comments re: his thoughts about getting data from a single hierarchy. There are also comments here on problems he's been having with the API in general. "You find a provider with "provider_hierarchies", then all its roots with "hierarchies", and then you can walk it with "hierarchy_entries". If you have a list of scientific names, you have to use "search", extract each page id and invoke "pages" to get the data. There's nothing in the search results to tell you which provider/classification/whatever that page is tied to. Just a lot of fuzzy ambiguity, e.g. What do you actually check? That the starts with "Lama glama" and maybe has parenthesis but not a sub-species name? Can you guarantee that all the tags have their authorship in parenthesis? (The raw CoL data doesn't.) If you've got references to good examples of EOL api usage, that'd be great to see. I couldn't find *anything* recently that wasn't "fuzzy" somehow (searching Google and github, mostly). For the IUCN data, I found the IUCN collection and was trying to harvest that but even trying to put in 10-60 second delays (and 5 minute delays every 10 requests) I kept getting knocked off for service interruptions."

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "EOL API Discussion Group":

    Tanya and Roger had questions about using the API:

    The issue of using the eol API for grabbing video came up, but we have some concerns. The primary concern is that we don't want to end up having to parse multiple hierarchies. We thought perhaps there was a way to identify a particular hierarchy, like Catalogue of Life, and ask for data objects of a particular type (video) associated with taxon names in that hierarchy. Do you have a recommended approach for that?

    If you are trying to walk a tree (which I'm guessing you want to do for animals), grabbing video, looks to me like you first find the number of the hierarchy with http://eol.org/api/provider_hierarchies/1.0 . COL is 529. Then you use that with with http://eol.org/api/docs/hierarchy_entries to find and walk the parent child relationships and get the EOL identifiers . Then finally you use the pages method to ask for videos for each EOL identifier. Would that work?

    But perhaps you also could just do plain search and pages call by scientific name, check for your own objects on the page to verify the match between your identifier and ours, and pull the videos that are associated with pages that have your content on them.

    about 2 years ago • edited: about 2 years ago

  • Profile picture of Cyndy Parr who took this action.
    Cyndy Parr changed the description of the community "EOL API Discussion Group".

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.
    Cyndy Parr changed the description of the community "EOL API Discussion Group".

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.
    Cyndy Parr created the community "EOL API Discussion Group".

    about 2 years ago

  • Profile picture of Robert Hole, Jr who took this action.

    Robert Hole, Jr commented on "EOL Learning and Education Community":

    @eoleducation: Thank very much. I had seen that list. I'm putting together several regional lists in addition to the CA general lists and (as I see you're manager of that collection) you're welcome to add any/all of them if you see fit.

    about 2 years ago

  • Profile picture of eoleducation who took this action.

    eoleducation commented on "EOL Learning and Education Community":

    @Robert Hole, Jr: @Robert, thanks for the note! We are excited about your California Biodiversity community. I think it will be a great place for people to come together to share on collections on EOL! There is a collection of region specific collections here: http://eol.org/collections/16402. Thanks for the info and we will be sure to join the California Biodiversity community!

    about 2 years ago

  • Profile picture of Robert Hole, Jr who took this action.

    Robert Hole, Jr commented on "EOL Learning and Education Community":

    I've started a California Biodiversity community and you're welcome to join and participate. I also offer it as inspiration to use for schools/classes to put together similar communities for some location they are working and living in. For instance "animals of Centerville Elementary School" or "Flowers we see on Our Miss Brooks' class field trips". And for those in California, it could be used as a jumping off point for making such collections/communities, limiting the number of taxa your students have to search through.. I would love to to hear about other regional collections around the world. I'm putting together a collection of those collections I find around here, regardless of taxa included.

    about 2 years ago

  • Profile picture of Valter Jacinto who took this action.

    Valter Jacinto commented on "Rapid Response team":

    bug on EOL: In Tachinidae there are moths! Resource: EOL Group on Flickr http://eol.org/pages/9019/worklist

    about 2 years ago • edited: about 2 years ago

  • Profile picture of Valter Jacinto who took this action.

    Valter Jacinto commented on "Rapid Response team":

    This comment was deleted.

    about 2 years ago • deleted: about 2 years ago

  • Profile picture of Valter Jacinto who took this action.

    Valter Jacinto commented on "Rapid Response team":

    Sedum brevifolium - This page is not included in any of the classifications that are browsable in EOL. Is this a new species ??? http://eol.org/pages/5545066/overview

    about 2 years ago

  • Profile picture of Valter Jacinto who took this action.

    Valter Jacinto commented on "Rapid Response team":

    2 different species in the same EOL page: Teucrium polium | Teucrium capitatum | http://eol.org/pages/5365621/overview

    about 2 years ago