Comments on: Just in time or just in case? http://commonplace.net/2009/10/just-in-time-or-just-in-case/ Library2.0 and beyond Mon, 26 Jul 2010 15:40:36 +0000 hourly 1 http://wordpress.org/?v=3.0 By: Janneke http://commonplace.net/2009/10/just-in-time-or-just-in-case/comment-page-1/#comment-4067 Janneke Sat, 17 Oct 2009 15:54:41 +0000 http://commonplace.net/?p=825#comment-4067 As someone who teaches lukas' endusers at the UvA, I was very surprised how much the students seem to like metasearch. During my instruction last week, the metasearchtask generated the least amount of questions (Maybe because it was the first task and they were still fresh.. but nevertheless..) Btw, I would still like to maintain that my inner librarian doesnt like metasearch (and H&I for that matter), she cannot handle such a lack of control ;-). But as someone who likes a happy customer, I am very interested to see what you and your colleagues come up with next. As someone who teaches lukas’ endusers at the UvA, I was very surprised how much the students seem to like metasearch. During my instruction last week, the metasearchtask generated the least amount of questions (Maybe because it was the first task and they were still fresh.. but nevertheless..)

Btw, I would still like to maintain that my inner librarian doesnt like metasearch (and H&I for that matter), she cannot handle such a lack of control ;-) . But as someone who likes a happy customer, I am very interested to see what you and your colleagues come up with next.

]]> By: Lukas Koster http://commonplace.net/2009/10/just-in-time-or-just-in-case/comment-page-1/#comment-4038 Lukas Koster Fri, 16 Oct 2009 15:46:27 +0000 http://commonplace.net/?p=825#comment-4038 Hi Till, thanks for your thoughts. Yes, the end user perspective is a little bit out of the picture here, but implicitly present at the same time. Selecting databases and interpreting results are the most problematic issues for end users. Although statistics show that the vast majority of users (more than 85% I think) don't even try to find the most appropriate databases, they just enter a search term and hit "Search", that way just using the preselected default set of databases. Hi Till, thanks for your thoughts. Yes, the end user perspective is a little bit out of the picture here, but implicitly present at the same time.
Selecting databases and interpreting results are the most problematic issues for end users. Although statistics show that the vast majority of users (more than 85% I think) don’t even try to find the most appropriate databases, they just enter a search term and hit “Search”, that way just using the preselected default set of databases.

]]> By: tillk http://commonplace.net/2009/10/just-in-time-or-just-in-case/comment-page-1/#comment-4037 tillk Fri, 16 Oct 2009 15:36:43 +0000 http://commonplace.net/?p=825#comment-4037 Thanks for sharing that presentation. It's a good conclusion of problems with metasearch. I think, I'll nail it on my forehead whenever I meet customers using the metasearch products wo support :-) Though, what's somewhat missing is the user's perspective (or at least the presentation doesn't point out clearly that perspective). Metasearch is hard to understand and use for end users, I have only anecdotal evidence for that, no own evaluation results. The user experience often starts with hard decisions to be made (which databases to select?). Result list presentation is often unusal (at best), handling of these lists often being complicated (there often is a somewhat mysteriously sorted "single list" of (parts of) merged results, seperate lists for each searched database, somtimes added by automatic reloads of result numbers, lists or rankings and the like... Of course, we will never have all data for local indexing, even Serials Solutions won't get it all (although they have already more than half a billion records). And even with all data at your hands, you still may offer horrible user experience... So I totally agree with your conclusions, but nevertheless am looking forward for the day metasearch finally is obsolete... Thanks for sharing that presentation. It’s a good conclusion of problems with metasearch. I think, I’ll nail it on my forehead whenever I meet customers using the metasearch products wo support :-)
Though, what’s somewhat missing is the user’s perspective (or at least the presentation doesn’t point out clearly that perspective). Metasearch is hard to understand and use for end users, I have only anecdotal evidence for that, no own evaluation results. The user experience often starts with hard decisions to be made (which databases to select?). Result list presentation is often unusal (at best), handling of these lists often being complicated (there often is a somewhat mysteriously sorted “single list” of (parts of) merged results, seperate lists for each searched database, somtimes added by automatic reloads of result numbers, lists or rankings and the like…
Of course, we will never have all data for local indexing, even Serials Solutions won’t get it all (although they have already more than half a billion records). And even with all data at your hands, you still may offer horrible user experience…
So I totally agree with your conclusions, but nevertheless am looking forward for the day metasearch finally is obsolete…

]]>