- From: carmen <_@whats-your.name>
- Date: Tue, 3 Dec 2013 01:32:05 +0000
- To: semantic-web@w3.org
> If anyone thinks a more positive timeline is possible, do please sketch one... all i can offer is a <FACT>, your speculative-fiction was too realistic to top without a time-investment longer than it took to make the change: 00:00s Alt-2 (switched wm to vi) 00:03 found K.rb 00:05 find line where URI is defined, change to: 00:08 Type = W3 + "ns/rdf#type" :w! 00:10 Alt-4 (back to URXVT) 00:12 rm -rf cache/ 00:15 fg httpd, ctrl-c, up-arrow return (restarted) input triplrs are using the constant shorthand, so theyre now emitting the new value, cache with the old data is gone, view-dispatcher's looking for the new Type URI. and so on. most libraries are setup like this no? there's a "CURIE" map at the top of Nathan's js3.js to map the "prefix"/"dirent"/"qname"/"CURIE"-alikes for the fractal-repetition of their slightly-different contexts especially given the problem-space of merging data from as many sources & expression-schemes as exist on the WWW, i'd expect most RDF libraries probably have a robust input/front-end types&identifiers-normalization layer, hinted with machine-learning AIs running on data culled from cralwers & the BTC? this just seems crucial eg there's DC's 1.1/elements vs purl/types or whatever they changd around 2006-8.. "Type" type-tag <- ex-MetaWeb-Freebase | Schema.org | Cyc | no URI just a <h1> in a wiki and :Symbols in .py code | some non-URI tag in HTTP 1.1 or IANA/MIME tables, etc have you opened up all the popular frameworks and tried changing it to see if it's difficult? i'd have forgotten about this already except my CONFIRM unsubscribe requests always seem to disappear to a digital abyss
Received on Tuesday, 3 December 2013 01:32:36 UTC