RDF: Difference between revisions
m ([tantek] added "Criticism: '''Difficult to use and poor DX''': From the AT Protocol FAQ: [https://atproto.com/guides/faq#why-create-lexicon-instead-of-using-json-ld-or-rdf Why create Lexicon instead of using JSON-LD or RDF?] <blockquote><p>RDF is intended for extremely general cases in which the systems share very little infrastructure. Itβs conceptually elegant but difficult to use, often adding a lot of syntax which devs don't" to "See Also") |
Tantek.com (talk | contribs) (finish blockquote) |
||
Line 51: | Line 51: | ||
* Criticism: https://news.ycombinator.com/item?id=31042118 <blockquote>billions of dollars in 2008-2010 timeframe on implementing Semantic Web technologies</blockquote> | * Criticism: https://news.ycombinator.com/item?id=31042118 <blockquote>billions of dollars in 2008-2010 timeframe on implementing Semantic Web technologies</blockquote> | ||
* Criticism: https://news.ycombinator.com/item?id=31046470 <blockquote>"These are so inaccurate that they are frequently misleading, and cannot be trusted." The Colonel later pulled me into his office and stated (rather comically): "You mean to tell me I've been paying people to draw cartoons for 3 years? We're not goddamned Disney here."</blockquote> | * Criticism: https://news.ycombinator.com/item?id=31046470 <blockquote>"These are so inaccurate that they are frequently misleading, and cannot be trusted." The Colonel later pulled me into his office and stated (rather comically): "You mean to tell me I've been paying people to draw cartoons for 3 years? We're not goddamned Disney here."</blockquote> | ||
* Criticism: '''Difficult to use and poor DX''': From the [[AT Protocol]] FAQ: [https://atproto.com/guides/faq#why-create-lexicon-instead-of-using-json-ld-or-rdf Why create Lexicon instead of using JSON-LD or RDF?] <blockquote><p>RDF is intended for extremely general cases in which the systems share very little infrastructure. Itβs conceptually elegant but difficult to use, often adding a lot of syntax which devs don't | * Criticism: '''Difficult to use and poor DX''': From the [[AT Protocol]] FAQ: [https://atproto.com/guides/faq#why-create-lexicon-instead-of-using-json-ld-or-rdf Why create Lexicon instead of using JSON-LD or RDF?] <blockquote><p>RDF is intended for extremely general cases in which the systems share very little infrastructure. Itβs conceptually elegant but difficult to use, often adding a lot of syntax which devs don't understand. β¦ </p<p>We looked very closely at using RDF but just didn't love the DX or the tooling it offered.</p></blockquote> |
Revision as of 00:26, 15 December 2022
This article is a stub. You can help the IndieWeb wiki by expanding it.
RDF is an abbreviation for Resource Description Format, an early-ish part of XML-related technology, that was pitched as the Semantic Web, used in Trackback and RSS 1.0 (but dropped by Pingback and RSS 2.0 respectively), later rebranded as Linked Data, experimentally published in the past, but not used by any indieweb use-cases in practice.
Microformats began in part as a reaction against the verbosity of RDF and extra technical layers. RDF developers eventually took some tips from Microformats in turn, and created RDFa and JSON-LD.
Example Uses
FOAF
FOAF (Friend of a Friend) is (was?) one of the more frequently used RDF vocabularies/formats, in many ways a reinvention of the vCard vocabulary (used in hCard microformats and h-card microformats2).
From this wiki (from IndieWeb community experience), FOAF was also:
- undiscussed at the original IndieWebCamp 2011
- briefly mentioned at HWC 2014-10-15 in Lisbon
- mentioned in the Indie Contacts session at IndieWebCamp 2014
- an example of the sidefile-antipattern
- indexed by the now defunct Google Social Graph API
- an example of past standards efforts
- noted as "dead" in the IndieWeb Summit 2019 State of the Suite (of Protocols) session
The general concept of a "friend of a friend" affording some trust (at least to leave comments) is used in the Vouch protocol.
Criticisms
complexity without benefit
The additional conceptual and implementation complexity costs of learning/coding RDF do not provide equivalent (or any in practice?) marginal benefit (certainly not to individual developers who have limited time/maintenance resources for their own websites):
- From: http://microformats.org/wiki/triples
The larger problem is that triples are more than just complicated, they are unnecessarily complicated.
syllogisms not useful
- syllogisms critique http://www.shirky.com/writings/herecomeseverybody/semantic_syllogism.html
provokes emotional ranting
For some reason, RDF and Semantic Web in general provoke strong negative emotional reactions from at least some developers:
- 2018-08-07: https://pleroma.site/notice/2962160
- "these people should be fucking ASHAMED of this work, it is a DISASTER. it is FILLED WITH RDF SEMANTIC WEB HORSESHIT. FUCK THE SEMANTIC WEB." @kaniini August 7, 2018
See Also
- JSON
- microformats
- RSS
- Aaron Swartz's unfinished book on the programmable web has a lot to say about RDF https://upload.wikimedia.org/wikipedia/commons/3/3f/Aaron_Swartz_s_A_Programmable_Web_An_Unfinished_Work.pdf
- HTTPRange-14
- https://ruben.verborgh.org/blog/2018/12/28/designing-a-linked-data-developer-experience/
- https://twitter.com/jaredhanson/status/1046929964614864897
- "Thereβs a diehard set of purists that will keep spinning the Semantic Web under the guise of different names, undeterred by actual adoption or market need." @jaredhanson October 2, 2018
- Criticism: 2019-02-10 Peter Molnar: A journey to the underworld that is RDF
- Criticism: threads https://twitter.com/YungTseTech/status/1125641970720755712
- "To be honest, I haven't yet seen a convincing case for the Semantic Web. Of course this could be because I'm some kind of unimaginative subliterate ... but ... (and I'm only putting this out here!) ... could it be that it doesn't solve a lot of problems people actually have?" @YungTseTech May 7, 2019
- https://unwalled.garden/docs/why-not-rdf
- ^^^
β¦ philosophy about RDF is YAGNI (You Ainβt Gonna Need It). We see RDFβs complexity as a turn-off to developers and something we should try to avoid if we can.
- Criticism, thread (includes some JSONLD criticism too, debunking of "simplicity" claims) https://twitter.com/pfrazee/status/1136719211735277574
- "Eh yeah but you're still encoding all that info somewhere, and it doesn't change the fact that RDF is actually a graph model with URL attributes that we're fighting to behave like a schema identifier" @pfrazee June 6, 2019
- Criticism: https://news.ycombinator.com/item?id=31042118
billions of dollars in 2008-2010 timeframe on implementing Semantic Web technologies
- Criticism: https://news.ycombinator.com/item?id=31046470
"These are so inaccurate that they are frequently misleading, and cannot be trusted." The Colonel later pulled me into his office and stated (rather comically): "You mean to tell me I've been paying people to draw cartoons for 3 years? We're not goddamned Disney here."
- Criticism: Difficult to use and poor DX: From the AT Protocol FAQ: Why create Lexicon instead of using JSON-LD or RDF?
RDF is intended for extremely general cases in which the systems share very little infrastructure. Itβs conceptually elegant but difficult to use, often adding a lot of syntax which devs don't understand. β¦ </p
We looked very closely at using RDF but just didn't love the DX or the tooling it offered.