This idea is active.
Search Engines »

Search Engines to add "Learning" as search context (just you can with images, videos, etc)

Search Engines could add "Learning" to their standard home page options of Web, Images, Videos, Maps, News, Shopping and E-mail.


Submitted by EricShepherd 3 years ago

Vote Activity Show

Comments (6)

  1. Moderator

    Nice one -- it would be great if it were as easy to find learning resources as it was to figure out what time a movie playing..

    3 years ago
  2. Yes. The microformat idea one enabler as we need to figure out how to tell the search engine that something is learning content.

    Any ideas on other ways to tell that something is learning?

    3 years ago
  3. "Learning Content" meaning what? Formal learning, or content that can be used for learning?

    Upon reflection, content might have a "degree of learning-ness" that measures how likely it is to be used (or useful) as a learning resource conditioned on a context. If that is true, the static approach to identifying a resource as a "learning" based on its format or provenance is limited. Even if we restrict ourselves to formal learning resources I would argue that we cannot in general rely on tags of any sort that the author or creator has to put there.

    For example, if every video had to be labelled with a "video tag" by its creator, most videos would probably never be labelled as such and never be found in a search for videos. Fortunately, the MIME type gives it away, so we CAN search for videos, but there isn't a "learning MIME type". So my two cents is that we should be looking at using pattern recognition and machine intelligence instead, and I believe this is where search is going. Methods that rely on the intrinsic properties of resources are scalable, whereas relying on people to label resources is not.

    3 years ago
  4. i disagree with the idea and i agree with Robby Robson.

    3 years ago
  5. If we assume with Robby that anything could be labelled educational (and you can't control or scale the use of the 'tag') one possible approach to creating the tab above would be to follow Creative Commons approach in DiscoverEd / or Google Scholar and think about educational in terms of curated collections. Provide a search of a set of known collections - either by domain, known repository, or perhaps declared 'faux-pository' [trying to think how to include slideshare collections for example].

    there is an overhead but partly the LR could address some of that and there are other existing declared collections - eg OCWC members; [it might help to know how Google maintains the list of sites Google scholar indexes].

    3 years ago
  6. Searching is a behavior driven by one's mental architecture (I'm calling it MA for now). MA is not universal but influenced by culture, including teh culture into which a person is born, that within which the person functions day-to-day, and the culture of the profession. And, all those engaged in teaching do not have the same MA, especially in adult learning. There is an interaction between the culture of teaching and the subject area which is being taught. For example, those teaching an art course possess a different MA than those engaged in teaching aeronautical engineering. In order to arrive at a search engine that meets the needs of diverse MAs, the engine should be able to learn and to adapt to input so that future searches are more immeidately productive in giving usable results. It may be helpful to engage and share metacognitive strategies each of us employ in searching for learning content as an initial step to identifying requirements for effective search engines for learning content.

    A second part of this relates to how the content is structured to provide the "triggers" that indicate teh content should be included in search results. What are current means for searching the content in addition to data included in metatags? How can content metadata and other information be structured so that it facilitates a successful search?

    In total, the TOE I see for search engines is one that responds to behavior driven by diffent MAs. Structure of the learning object would also incorporate facets associated with the MA of the intended user population.

    3 years ago