[Humanist] 29.29 billions of pages' worth

Humanist Discussion Group willard.mccarty at mccarty.org.uk
Fri May 15 07:32:23 CEST 2015

                  Humanist Discussion Group, Vol. 29, No. 29.
            Department of Digital Humanities, King's College London
                Submit to: humanist at lists.digitalhumanities.org

        Date: Thu, 14 May 2015 10:55:32 -0400
        From: Hugh Cayless <philomousos at gmail.com>
        Subject: Re:  29.24 billions of pages' worth
        In-Reply-To: <20150513051508.A367A668A at digitalhumanities.org>

Technical transitions are not religious wars, it is true, but technology is, to an unreasonable extent, fashion driven. Those of us who’ve been around the block a few times are very familiar with this sort of cycle:

1. New technology appears that compares favorably with an established technology, achieving greater speed or ease of use, often by dropping big chunks of the established technology.

2. New tech acquires many evangelists, becomes very popular. Old tech’s users are derided and have to defend their decision not to be fashionable. Inexperienced users choose the new tech based only of its hype and contribute to that hype.

3. Users begin to realize parts of the established tech that the new one dropped were actually useful. They begin to add them to the new tech.

4. The realization dawns that the new tech is now just as clunky/slow/undesirable as the old. Maybe it’s totally unusable now, or maybe it’s reached a sustainable level of maturity. Some people drift back to the established tech if it’s still viable, leaving a core of dedicated users, some move on to another new tech. Maybe the two continue to coexist, like vi and emacs. Old tech users (if any remain) lead a chorus of "I told you so."

5. Rinse, repeat.

JSON is at about step #3 right now. Time will tell if JSON is to XML as XML was to SGML or if it’s like the NoSQL movement is to the RDBMS (for reference see e.g. https://dennisforbes.ca/index.php/2010/03/24/the-impact-of-ssds-on-database-performance-and-the-performance-paradox-of-data-explodification/). 

There are no silver bullets. There is no single "right" technology. The choice of what technology to use to accomplish a task should be made based on its affordances, it’s maturity level, the help you can get from its user community, and its overall suitability for your data and your requirements. HathiTrust and its developers seem to me to have made a perfectly sensible non-ideological technology decision.

You, Desmond, are a sort of XML kakangelist :-). I can understand finding a technology flawed and even irritating (I feel much the same way about RDF), but you seem to me to raise your dislike of XML to the level of ideology, and I don’t think that’s a reasonable basis for deciding whether or not to use a technology.

All the best,

 *  Hugh A. Cayless, Ph.D
 *  hugh.cayless at duke.edu
 *  Duke Collaboratory for Classics Computing (DC3)
 *  http://blogs.library.duke.edu/dcthree/

More information about the Humanist mailing list