jump to navigation

Alternative stimulus response June 7, 2013

Posted by Mia in Frontiers.
comments closed

Something’s been bugging me.  There is a huge swell of interest in alternative metrics as an indicator of scholarly and social impact. These alternatives are distinguished from traditional bibliometrics which focus primarily on ISI’s journal impact factor (jif).

The scholarly journal framework has dematerialized into a loose, unbounded container — although “journal” is still a useful model and I don’t think it has outlived its purpose quite yet. It still meaningfully represents a collection of thematically related, stand-alone information units – otherwise known as “articles” — and regardless of whether the collection is completely virtual.

The direct dissemination of information will continue to inspire the creation of new and emerging models which will help us understand how scholarly information is propagated, absorbed, and digested into the broader community. Perhaps it is more of an organic process than we realize.

Some are calling for the inclusion of altmetrics into tenure packages, but I would be wary of endorsing such general recommendations, particularly since it is academia which misappropriated the use of the JIF for determining research ‘value’ in the first place.

There are tremendous differences in the way in which various disciplines conduct, generate and communicate research/results.

The click of a link may be an interesting pattern generator, but so much further analysis has to be conducted on those patterns in order to gain understanding of what they might meaningfully represent.

Internet/infotainment phenomena have very short life spans.  Reputation management is on the rise. Using Capitalizing on SEO tactics to drive more traffic in a particular direction may boost altmetric “scores”, but like a downpour on parched earth, there’s a lot of runoff with no absorption.   The lack of the term ‘altmetrics’ in this blogpost title is intentional.

Reflection and digestion take time.  Unless, of course, you’re just clicking links and aren’t actually reading, reflecting or digesting.

ISI’s Journal Impact Factor’s characteristics and methodology are well-documented. The journal impact factor has been well-studied and well-critiqued over the years.  It is so well-understood by the scientific research community that its rejection as a measurement (see here, and the recent DORA) has been greeted with widespread enthusiasm.

Now altmetrics need to be scrutinized with a similar kind of rigor along with a healthy dose of skepticism.  Or they run the very real risk of becoming altmetricks instead.


Forget Big Brother March 9, 2013

Posted by Mia in Uncategorized.
comments closed

We’re getting inundated with all this hype about Google Glasses.  How many people will be lining up to pay Google for the privilege of being volunteer recruits as walking-humanoid-google-streetview-cameras?  I’m no conspiracy theorist, but these GG things are hardly benign.  They’re the latest assault and invasion of personal space yet.

Plus, they so remind me of that Star Trek episode — you know, the one where everyone is wearing a game headset and getting addicted to sucking those creepy little funnels into the holes.  Until no one on the Enterprise is doing anything else but feed their addiction to this game.

The images and video alone that would be captured by these glasses and stored on google servers, along with geo coordinates, timestamps, and google image (face) recognition?

Not a Good Idea.

Discoverability December 18, 2012

Posted by Mia in Frontiers, Uncategorized.
comments closed

We’re embarking on a discovery layer project and looking at the few contenders out there in the marketplace.

I am not yet sold on building a pre-harvested index composed of all different types of indexing terminology and depth of granularity thrown in together. How reasonable it is to have something like, oh, I don’t know, say ninety or so different content types, so that you can filter your results?

The goalpost just moved.  Its easier to start your search and get some meaningful results, and that is very good.   

But to refine a search might mean you are now faced with an abundance of choice as to which content types to tick off as you are scrolling through the list of terms that you can check off.

We’re back to listing terms in an alphabetical sequence and asking users to (once again) browse through the terms to find ones that strike their fancy!

It’s no wonder relevancy ranking algorithms are closely guarded secrets.  Special sauce, indeed.

Oh, look, it’s snowing! Thank you, WordPress. It’s that time of year.

Typography counts November 26, 2012

Posted by Mia in Frontiers, Uncategorized.
comments closed

A recent study by Larson and Picard on the Aesthetics of Reading has been circulating on the internets (thanks to Fadayev’s UsabilityPost). One thing the authors tested for is whether good typography can be measured to have a positive effect on mood. The interesting thing is that reading comprehension and speed were not measured, but rather the reader’s subjective impression of how long it took to read the piece (the duration). The authors of the study that reading is an engaging task, but reading when high quality typography is present is “more engaging” — as measured by users consistently (and by a wide margin) underestimating the amount of time it took them to complete the reading.

…high quality typography appears to induce a positive mood, similar to earlier mood inducers such as a small gift or watching a humorous video. This is an exciting finding because there are important differences between good and poor typography that appear to have little effect on common performance measures such as reading speed and comprehension.

It’s great to see the importance of typography getting attention. Recently the UK GDS team (responsible for launching the new website gov.uk) was deciding on legibility, so they looked to a font widely adopted for highway signage:

Accessibility is key at GDS and ideally we’d like a typeface that’s good enough for us not to need an ‘easier to read’ font option for the dyslexic and those with other visual or cognitive issues. We tried lots of different ones and the best was Transport. This shouldn’t come as a surprise – it’s the typeface designed by Jock Kinneir and Margaret Calvert in the late fifties/early sixties for Britain’s national motorway and road sign system. They spent years testing for legibility in all sorts of extreme conditions; in the rain, at night, at speed.

A font that withstood years of testing under extreme driving conditions should score well on lowering the stress quotient for sure.


RDA is starting to impinge on things October 13, 2012

Posted by Mia in CLUES/WebPAC, Resource Description and Access.
comments closed

A new MARC field snuck up on me — the 264.  The problem with this field is that it was only invented in 2011 to handle new RDA rules for interpreting imprint information that can’t be accomodated by the existing 260.  Quoting from MARC bibliographic (emphasis mine):

Information in field 264 is similar to information in field 260 (Publication, Distribution, etc. (Imprint)). Field 264 is useful for cases where the content standard or institutional policies make a distinction between functions


Huh? A field [for handling imprint], which is similar to an already existing field [which handles imprint], only different?  This ‘definition’ really eludes me, I must admit, but I’ll leave it to others to interpret.

Records started arriving with a 264 indicator 1, not a 260, so naturally no imprint information was displaying to the public.  Rules for handling the import and display of the 260 (like on webpub.def) didn’t include the 264 because the field hadn’t existed previously.  None of our MARC load tables knew about this field.

I’m surprised that this field came in under the radar.  I am wondering how much relevance rules that may look at the — guess what? — 260|c field need to be modified.  I can’t see how that can be avoided.

And why this change now, when the MARC format is on its way out?  The whole thing particularly mystifies me, since many of the date complexities are only applicable to physical entities — entities which are a diminishing proportion of our collection.