tag:blogger.com,1999:blog-7419001987588063203.post3352577642450497664..comments2024-02-15T00:29:00.066-08:00Comments on Vitruvian design for scholarship in the humanities: Digital scholarship must be technology-agnosticNeel Smithhttp://www.blogger.com/profile/10590621399352493304noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-7419001987588063203.post-85156312284513049402013-05-27T04:58:07.854-07:002013-05-27T04:58:07.854-07:00I can't http://besteditingservices.org/ you th...I can't <a href="http://besteditingservices.org/" rel="nofollow">http://besteditingservices.org/</a> you the number of hours I have spent dealing with the sigma insanity in Unicode. It is *utterly* mad. Greek had the misfortune of being an early entrant into the Unicode space, and presumably they learned from their mistakes when it came to dealing with Arabic.<br />Assmahttps://www.blogger.com/profile/14036886108127843427noreply@blogger.comtag:blogger.com,1999:blog-7419001987588063203.post-19432690479890013992012-03-02T07:55:32.477-08:002012-03-02T07:55:32.477-08:00I find the CTS URN system very interesting, but th...I find the CTS URN system very interesting, but there is one thing that puzzles me in the ontology of the text that lies behind it. The system follows the FRBR approach, where the "expressions", "manifestations" and "items" of a text emanate from the "notional text". This is the librarian's (and the scribe's) perspective. The philologist's perspective is quite opposite. For us the "text" originates from its sources, its witnesses (MSS, papyri, etc.), that is from the "items". The witnesses ("items") are the only reality of the text. In other words, the philologist's ontology of the text is (should be) document-based. This is, by the way, what makes the Homer Multitext project so interesting to me.<br /><br />From this perspective, a controversial point lays in the passage from, say, urn:cts:greekLit:tlg0012.tlg001:1.1-1.20 to urn:cts:greekLit:tlg0012.tlg001.manuscriptX:1.1-1.20 and to urn:cts:greekLit:tlg0012.tlg001.manuscriptY:1.1-1.20. This assumes that the first 20 verses of the Iliad in our "notational text" called "Iliad" (that is in fact the positive result of a scholarly consensus and is not abstract in itself) are unproblematically mappable upon the first 20 verses in manuscript X and in manuscript Y. How does the CTS URN system manage a MS Y having a "spurious" 11th verse after the 10th "canonical" verse? This might work as a counterexample showing the difference between a notional text-based ontology and a document-based ontology of the text.Paolohttps://www.blogger.com/profile/14410410835949255850noreply@blogger.comtag:blogger.com,1999:blog-7419001987588063203.post-87968505842692831672012-02-12T21:18:32.197-08:002012-02-12T21:18:32.197-08:00[In reply to Sebastian] The good news about using ...[In reply to Sebastian] The good news about using URN citations is that they are primarily implemented as URLs through the http:// structure. But what makes having a URN better than having a simple URL is that the URN is the consistent part of the URL - that is, whenever the text service is hosted somewhere else (which happens all the time in say, the Homer Multitext Project) - it is still actionable and useful. One neat outcome recently discovered is that one can type a URN into Google and Google has been able to pull the appropriate text through whatever text service it finds. <br /><br />As for the longevity of links/URLs - I read an article (http://worldcat.org/arcviewer/5/LEGAL/2011/06/15/H1308163631444/viewer/file2.php) that talks about the issues of preservation when relying on links/URLs to point to a particular source. I think that this is a very scary concept when the goal is to preserve content and preserve access to content. <br /><br />So - my comment is, URLs provide transient access to information. But, one can have their cake and eat it too since the URLs to access these texts are built on this stable, long-term solution of a URN referring to a text.Anonymoushttps://www.blogger.com/profile/10099038282015624717noreply@blogger.comtag:blogger.com,1999:blog-7419001987588063203.post-83148429273557279982012-02-09T05:55:29.674-08:002012-02-09T05:55:29.674-08:00Agreed that "http:" identifiers derived ...Agreed that "http:" identifiers derived from URL addresses could be used independently of their original intended function, and are here for a long time, but are they expressive enough? Taken out of their planned use an address, they aren't parseable with any specified semantics.<br /><br />Aren't we at risk of looking for nails we can bang our hammer on?Neel Smithhttps://www.blogger.com/profile/10590621399352493304noreply@blogger.comtag:blogger.com,1999:blog-7419001987588063203.post-83030582961365530442012-02-08T09:17:15.241-08:002012-02-08T09:17:15.241-08:00Hi Neel,
I think there are advantages to using id...Hi Neel,<br /><br />I think there are advantages to using identifiers that are both unique going forward and actionable now.<br /><br />It is true that an address like http://nomisma.org/id/igch1544 has a specific meaning in today's technological environment. As you say, you can paste it into a browser and get to a web page.<br /><br /> I don't think that means that access to the definition of the concept "IGCH 1544" will go away if the DNS system changes. "http://nomisma.org/id/igch1544" can remain unique, even if access to the definition needs to be mediated through a new location (a new URL).<br /><br /> Taking your longer perspective, I think sequences of characters beginning "http://..." can be put out on the internet in such a way that they implement very long-term identification. That some 1000s of years from now, a future scholar may be able find sufficient traces of certain "http://" identifiers to be able to both figure what they identified and to find that content. And I do think that ability to recover is made more likely by the fact that the content behind "http://" identifiers can be copied without intervention of the "owner" of the identifier.<br /><br />To summarize: only time will tell but I think we can point to "http://" identifiers as one route to long-term viability and technological independence.Sebastian Heathhttps://www.blogger.com/profile/00323339285494849021noreply@blogger.com