On 8/12/2015 8:43 AM, Bernhard Eversberg wrote:
> … misleading metaphors like “relevance” and “ranking” should > not be tolerated in catalogue interfaces.
> Any critically thinking patron might well be amused – or enraged – and > call out, “Relevance? How can a machine know anything about my > intentions when I just enter two or three words? That’s preposterous!” > (Well, they don’t, but that’s because library catalogs are of > no critical relevance to them.)
In regular search engines, “relevance” is supposed to be determined not only by citation analysis (how many pages link to a specific page) but also by algorithms, much of those based on computers “knowing us” by using in-depth browsing history, scanning our emails, by knowing where we are and have been (with mobile computing), and so on. Ostensibly, they do this in order to make the advertising we see more pertinent to what we want.
As a result, when we see an ad in a printed newspaper, it is undirected and aimed at everyone, but an online/email ad can zoom into what we want and we see advertising directed at us. This is the theory, but the reality is a disaster. More and more people are using adblockers on the web. I use AdBlock Plus too. (http://www.theguardian.com/technology/2014/sep/09/blocking-adverts-napster-for-the-publishing-industry)
Why? Well, for one thing, I think this demonstrates that the “relevance” everybody praises is just rubbish. I admit: I want to see ads, that is: if they really are “relevant” to me–but 99.999% of the ads I see on the web are not relevant at all and (based on the Guardian article above) it looks as if I am not the only one to think so.
If “relevance” doesn’t work with ads (which are incredibly important on the web) then why are we supposed to believe it works better with search engines? After all, it’s using the same technologies. I mentioned in a post sometime back, that I suspect (but cannot prove) that the Googles not so much give us listings of sites in order of “relevance,” but the Googles have managed to move the completely irrelevant sites–those that used to torment everyone by coming up first in pre-Google search engines–much farther down the list, often to the second page of results where they will never be seen.
This would also account for people, including myself, who look at the top three to ten sites before giving up and trying something else. They do it because the listings they see are not relevant.
Therefore, I suspect that the most relevant sites to your search may be buried much farther down in the list, but no one would ever find them. I also don’t know how this could be researched. But in any case, I think that the idea of “relevance” is pretty fuzzy even in state-of-the-art search engines.
What relevance means in library catalog is obviously something different from search engines. But no matter: the public demands relevance ranking, no matter what it means or how it is implemented. They just want to believe it works and to open that “Pandora’s Box” would be just too frightening!
James Weinheimer email@example.com
First Thus http://blog.jweinheimer.net
First Thus Facebook Page https://www.facebook.com/FirstThus
Personal Facebook Page https://www.facebook.com/james.weinheimer.35 Google+ https://plus.google.com/u/0/+JamesWeinheimer
Cooperative Cataloging Rules http://sites.google.com/site/opencatalogingrules/
Cataloging Matters Podcasts http://blog.jweinheimer.net/cataloging-matters-podcasts The Library Herald http://libnews.jweinheimer.net/