Kathy Hill

Cicada researcher

The EOL Profile Newsfeed contains comments left for its owner by other members, EOL Community invitations, and gathers updates associated with the items in the owner's EOL watch list.

Add a new comment

  • Profile picture of Jennifer Hammock who took this action.

    Jennifer Hammock commented on "Rubenstein Fellows":

    @Kathy Hill: More good questions! Regarding Flickr, that's correct, those photos only persist on EOL as long as they remain on Flickr, so if you are Flickr account space-limited that is not the ideal option. For scientific images, Morhbank can be convenient; they have bulk upload tools and they share with EOL. (http://www.morphbank.net/About/HowToContribute/). The new Scratchpads are also likely to be a good option for this. Regarding text entries from multiple sources: a direct text entry article and a Scratchpad article, for the same chapter of the same taxon, will both be retained. Many chapters on some taxon pages display multiple entries from different sources. The only time content is overwritten is when an existing, shared object changes at the partner site of origin. In that case the revised object overwrites the original on EOL at the next partner harvest.

    over 2 years ago

  • Profile picture of Kathy Hill who took this action.

    Kathy Hill commented on "Rubenstein Fellows":

    @Cyndy Parr: Thanks!

    over 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Rubenstein Fellows":

    @Kathy Hill: Right, while we hope that partners stay online forever, if they don't we would simply stop harvesting and keep what we have if the partner is willing. Also, we do keep old versions of data we have harvested (though we probably can't do that forever) so can serve as a backup repository even for sites that do stay online.

    over 2 years ago

  • Profile picture of Kathy Hill who took this action.

    Kathy Hill commented on "Rubenstein Fellows":

    @Cyndy Parr: Thanks Cyndy, that is very useful to know, and I assume it applies to all data objects. So we have to hope that all content partners stay online "forever" (or I guess you could just not harvest from them again if they are gone)?

    over 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Rubenstein Fellows":

    @Kathy Hill: Kathy, yes, if you remove an image from your Scratchpad or from Flickr (or from any other EOL content partner) the image will be removed from EOL the next time we harvest.

    over 2 years ago

  • Profile picture of Kathy Hill who took this action.

    Kathy Hill commented on "Rubenstein Fellows":

    @Bob Corrigan: I'm just trying to figure out what would be the best platform for me to use to share my Cicadidae data. I have my own website www.insectsingers.com but it seems that it would probably be easier to use Scratchpads or Lifedesks which are already set up to work with EOL. Since I am totally new to this, I am trying to figure out which is better. I still have some trouble understanding how EOL works. After EOL harvests the information from somewhere like Lifedesks or Scratchpads or Flickr, does it then store the information itself? If an image, say, is removed from the Scratchpad or Flickr site, and EOL "reharvests" from this place (or however it works), does the image remain with EOL?

    over 2 years ago

  • Profile picture of Bob Corrigan who took this action.

    Bob Corrigan commented on "Rubenstein Fellows":

    @John P. Sullivan: Ahh... that explains it.

    over 2 years ago

  • Profile picture of Fernando Jerep who took this action.

    Fernando Jerep commented on "Rubenstein Fellows":

    @John P. Sullivan: Very nice initiative, great project !

    over 2 years ago

  • Profile picture of John P. Sullivan who took this action.

    John P. Sullivan commented on "Rubenstein Fellows":

    @Bob Corrigan: "AFRICHTHY" only looks intimidating to pronounce: AFRIC-THEE from Africa + ichthy, the Greek root for fish. :-)

    over 2 years ago

  • Profile picture of Bob Corrigan who took this action.

    Bob Corrigan commented on "Rubenstein Fellows":

    @John P. Sullivan: That's great news - we (EOL) are working closely with the Scratchpad team and look forward to your feedback. Well done creating AFRICHTHY (that's a very hard acronym to pronounce)!

    over 2 years ago

  • Profile picture of Bart C. who took this action.

    Bart C. commented on "EOL Discussion Group":

    @Bob Corrigan: Good to know something similar is already in your queue. I will be on the lookout for it in the future. Thanks for the info about the barriers.

    over 2 years ago

  • Profile picture of Bob Corrigan who took this action.

    Bob Corrigan commented on "EOL Discussion Group":

    @Roderic Page: Agreed on your last point. The API as it exists today is a first step - but not the final one.

    over 2 years ago

  • Profile picture of Bob Corrigan who took this action.

    Bob Corrigan commented on "EOL Discussion Group":

    @Bart C.: This is a feature we have in our queue, and the barrier to implementing it is primarily one of there being higher priorities for us right now. Getting it right would involve the creation of a new "link" object (as distinct from a text object), a parser to process a defined set of RSS feeds at a regular frequency, and some UI/UX work to place it in the right place on a taxon page.

    The most complex part is the automation - we could add #1 and #3 and rely on users to provide us with links, which could in turn be curated, ranked, collected, commented by other users.

    If you'd like to chat, just drop me a note here or on twitter (@eolpm).

    over 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "EOL Discussion Group":

    @Cyndy Parr: Two separate issues to separate: 1. What does Google Refine send to service to make match? 2. What does Google Refine expect back? Taking last one first, Google Refine expects an id and a measure of confidence for that id, that's it. If service supports preview then user can "see" the match, which is where we could build a preview widget based on EOL API. Regarding 1, perhaps I misunderstood but I got sense @Jennifer Hammock was asking whether Google Refine could use additional information when trying to make the match. For example, if I know the name is for a plant I could tell the service this so it would ignore animal homonyms. Google Refine can pass on additional columns to the service (or the service can support "types") but this only helps if service itself can make use of that information. Because EOL API is just searching on a text string I can't tell it to filter by, say taxonomic group or geography). I may have responded to a question that wasn't actually being asked ;) I may also have let my frustration with the EOL API slip out. As with other projects such as Mendeley, if you design the API without actually using it then it is invariably crap. Until you can reproduce the EOL web pages using just API calls the API will be unsatisfactory.

    over 2 years ago • edited: over 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "EOL Discussion Group":

    @Roderic Page: Rod, not sure what you mean, the Pages API method returns nearly all the text and image data we have for a page identifier organized by subject. As you say, some of that information (like distribution) might be helpful for users as they manually merge.

    over 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "EOL Discussion Group":

    @Bart C.: Bart, it is okay if you leave your original comment. If you want to you can leave a comment thanking the person for their prompt fix to the machine tags. We love the Flickr community because they are so quick to respond!

    over 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "EOL Discussion Group":

    @Cyndy Parr: This isn't straightforward to do, Google Refine expects to get an id back. One can send the reconciliation service more information from other fields in the database, but it's not clear to me that EOL's API supports anything other than name searching. Another approach would be to create a preview (such as exists for Freebase topics) where the Refine user could see an image of the taxon and some text and use that to decide if the match was correct. I'd toyed with doing this but got bored. May revisit it.

    over 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "EOL Discussion Group":

    @Roderic Page: Jen may have been suggesting to pull in and display other fields of information to assist the reconciliation in Google Refine. For example the EOL API can offer higher taxa, common names, distribution, habitat, etc.

    over 2 years ago

  • Profile picture of Bob Corrigan who took this action.

    Bob Corrigan commented on "EOL Discussion Group":

    @Roderic Page: Rod, good news, we are a few weeks away from deploying our notification system. Stay tuned.

    over 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "EOL Discussion Group":

    @Jennifer Hammock: Sorry for delay in replying (EOL desperately needs an email notification system). I limit the number of results from the EOL search to 3 to keep things manageable. Automatch should pull out best matches if there's one unambiguous best hit. Not totally sure what you mean by "pulling additional columns"...

    over 2 years ago