t byfield on Sat, 27 Dec 2003 20:52:31 +0100 (CET) |
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
<nettime> the semantic web for beginners |
or: M*ke M*ney F*st as a part-time ontological envelope-stuffer! even boiled down to plain-vanilla ascii, this is *the* best discussion i've seen about the chimerical 'semantic web.' i'm sending this version to nettime as a sort of nostalgic doff of the hat to 'collaborative text filtering' (which doesn't seem to be doing very well these days, at least not in this neighborhood), but don't miss the page itself: < http://poorbuthappy.com/ease/semantic/ > as is typical of self-styled 'social software' blogosophers, the sub- stance of the discussion feels a bit like the patter of rain upon the Disciplinary Oceans (which is fine, if only they'd admit it). starting with their categorical mistake of insisting that this is 'ontology' and NOT epistemology, which very rarely gets mentioned in this context. (a quick empirical dip: googling 'semantic web' with 'ontology' yields 58000+ hits, but with 'epistemology'? 800+.). it is and it isn't; but the line separating the two realms of inquiry was never so clear -- and certainly shouldn't be taken for granted in this contect, given the substance of the debate. this tendency to glom on to ontology is partly attributable to the curious, uh, semantic swerve that 'ontology' took when computer-types got their hands on it. the easiest way to make this point is by ex- ample, say, that of yahoo's jerry yang in a '96 issue of WiReD: In 1996, [Yahoo's first "ontologist" 'Ninj' Srinivasan] was adding categories and making changes to the ontology almost every day. Now major adjustments are becoming much more in- frequent. She pointed to this as support for Yang's assertion -> that "at some point, our scheme will become relatively stable. -> We will have captured the breadth of human knowledge." -- http://www.wired.com/wired/archive/4.05/indexweb_pr.html 'hello, world!' but only partly attributable: because of course the practitioners who kept their hands firmly on the semantic steering wheel through the decade during which that swerve was 'locked in' were very much about making all that is solid melt into bits and, in the process, making millions -- in other words, making all that is solid melt into their pockets (sound familiar?). so it's not like all this blabla about 'on- tology' was only and unquestionably received wisdom; or even to the extent that it was, wasn't there some talk about a 'revolution' that'd chase out all the old computational ways and bring in the new? goodbye epistemology, hello ontology -- just without the hassle. inquiries beyond mere representation were the tprgic lot of all those goofy left- ists sniping at WiReD. but Real Men did windows -- *only* windows: * From The Free On-line Dictionary of Computing (09 FEB 02) [foldoc]: * * ontology * * 1. <philosophy> A systematic account of Existence. ^^^^^^^^^ even in this impoverished definition, they make oblique reference to all that Actual Stuff. but let us move on: * 2. <artificial intelligence> (From philosophy) An explicit * formal specification of how to represent the objects, concepts * and other entities that are assumed to exist in some area of * interest and the relationships that hold among them. * * For {AI} systems, what "exists" is that which can be * represented. When the {knowledge} about a {domain} is * represented in a {declarative language}, the set of objects * that can be represented is called the {universe of discourse}. * We can describe the ontology of a program by defining a set of * representational terms. Definitions associate the names of * entities in the {universe of discourse} (e.g. classes, * relations, functions or other objects) with human-readable * text describing what the names mean, and formal {axioms} that * constrain the interpretation and well-formed use of these * terms. Formally, an ontology is the statement of a {logical * theory}. * * A set of {agents} that share the same ontology will be able to * communicate about a domain of discourse without necessarily * operating on a globally shared theory. We say that an agent * commits to an ontology if its observable actions are * consistent with the definitions in the ontology. The idea of * ontological commitment is based on the {Knowledge-Level} * perspective. * * 3. <information science> The hierarchical structuring of * knowledge about things by subcategorising them according to * their essential (or at least relevant and/or cognitive) * qualities. See {subject index}. This is an extension of the * previous senses of "ontology" (above) which has become common * in discussions about the difficulty of maintaining {subject * indices}. * * (1997-04-09) so if you like philosophical slapstick, here you go: the very fine text of an excellent 'graphic novel' depicting a bunch of bloggers debating the hows and whys of 'how' -- a processual how -- to make groundless knowledge fungible. or, alternatively, if you're interested in getting a handle on this 'semantic web' stuff, here's a primer that explains *why* you won't be able to: platonism is alive and well, and its objects are just as elusive. at least, that seems to be the de facto consensus in the absence of any clear and distinct consensus. :) cheers, t - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Themes and metaphors in the semantic web discussion. November 15, 2003, [1]Peter Van Dijck. November 22 note: I had some versioning problems, and the article may now have a few slight differences with the November 15 version that are not indicated. In this article, I am trying to identify some of the themes and metaphors used in discussions about the semantic web. [2]Comments here. Last week Clay Shirkey posted an essay called [3]The Semantic Web, Syllogism, and Worldview, which generated a lot of responses on various websites. The essay asked the question: "What is the semantic web good for?". It echoes a feeling shared by many people, and expressed by JayT in the comments [4]here: If the Semantic Web was anywhere even close to half-as-good as the claims that are being made for it, it would ALREADY have gained massive widespread acceptance. Meaning VAST implementation. I printed out many of the discussions. While reading all this (using my trusty yellow highlighter), some recurring metaphors and themes emerged: * [5]What Is It * [6]Top Down Bottom Up * [7]Ontology Of Everything * [8]The Simple Life * [9]It Is Growing * [10]RDF Versus XML * [11]Real World Value * [12]Here Today * [13]Useful While We Are Get There * [14]Clay Misunderstands Syllogisms * [15]Doer Versus Talker My apologies in advance for being simplistic in this write-up - despite having developed [16]a language that has some semantics in it, I am far from an expert on the semantic web. Before we start, a few points about this writeup: * Any tips for improving the semantics of the HTML of this article are appreciated. * If I misquoted someone, or got a picture wrong (not everyone had a FOAF file), please let me know. I don't have an editor. * If you know where I can find a picture of [DEL: Joe Gregorio :DEL] , [17]JayT, [DEL: Dare Obasanjo :DEL] or [DEL: Tom Hoffman :DEL] , let me know. I'll be quoting lots of people - here's the cast: [pictures of everyone quoted in this article] Now let's get started. [18]What Is It? A starting theme of this discussion (and many others) is What Is It. (The different themes are bolded and capitalized.) The semantic web is an idea promoted by [19]Tim Berners Lee, a vision. But many people are confused about what exactly that idea is. A good thing about discussions like this is that many people try to define what the semantic web really is. Definitions go from the technical to the poetic: Paul Ford [20]states it technically: The Semantic Web is a framework that rigidly defines a means for creating statements of the form "Subject, Predicate, Object" or "triples," in a machine-readable format, where each of Subject, Predicate, Object is a URI. Shelley Powers [21]is more poetic: My idea of semantic web is if I can look for a poem that uses a metaphor of bird as freedom, and get back poems that have bird as metaphor for freedom. [22]Top Down Bottom Up. Once people start discussing what the semantic web is, two seemingly opposite visions of the semantic web emerge. We'll call them the "top-down" and the "bottom-up" semantic web. The theme is called Top Down Bottom Up. This theme is politically and emotionally loaded, and causes much misunderstanding. Someone should do a serious linguistic analysis of all this, but meanwhile: The top-down semantic web is discussed in terms of: * capital S Semantic Web. * the grand vision * all encompassing * global ontology * artificial * "the new world order of authoritarian classification systems" The bottom-up semantic web is discussed in terms of: * bottom up * already here * small victories * real * useful today Shirky, in his essay, says the proponents of the semantic web have a top-down view of the semantic web. This top down approach consists of two parts. According to Shirky, proponents of the semantic web want to make people use RDF (a building block of the semantic web). This is mostly true: the semantic web people want you to use RDF to share your semantic data, but they realize lots of semantic data is and will be shared in other forms - see [23]RDF Versus XML. By the way, the way the RDF Versus XML discussion is seen as an extension of the Top Down Bottom Up discussion is misguided. People seem to think XML is bottom-up, and RDF is top-down. But as Danny Ayers rightly [24]points out, XML is a top down technology: Shirky's last words are "... the big advantage of this bottom-up design and adoption is that it is actually working now." How much of that bottom-up design and adoption features XML, a spec that for most people is handed down from above? The second part of the top down approach, according to Shirky, is that semantic web people want to create top-down ontologies. Shirky really misses the ball here - most proponents of the semantic web don't believe in a global ontology. See [25]Ontology Of Everything. So in short, Top Down Versus Bottom Up is not really a valid discussion in terms of RDF Versus XML. In terms of Ontology Of Everything, it is just a plain misunderstanding of what the proponents of the semantic web propose. One reason why people get confused in this Top Down Bottom Up argument is that it is a theme with emotional and political undertones. Shelley Powers [26]captures the feelings of many people who believe in bottom-up. Notice the political and emotional undertones in this discussion. Tim [referring to Tim [DEL: Berners Lee :DEL] Bray], man, you got to get down, son. Scrabble in the hard pack with the rest of us plain folk. Yank off that tie, and put on some Bermudas and hang with the hometown gang for a bit. You been with the Big Bad Business Asses too much -- you forgot your roots. [27]Ontology Of Everything. As part of the Top Down Bottom Up theme, many people (starting with Shirky) seem to think the semantic web people are trying to promote an ontology of everything. Clay Shirky [28]writes: [About XML] With such a foundation, making formal agreements about the nature of whatever was being described -- an ontology -- seemed a logical next step. So Shirky seems to be saying that semantic web relies on top-down ontologies. Not so. The proponents of the semantic web are well aware that building a top-down ontology or taxonomy that works for everybody and tries to describe everything won't work. Shelley Powers [29]writes: There never was a suggestion that all metadata work cease and desist as we sit down on some mountaintop somewhere and fully derive the model before allowing the world to proceed. Dan Brickley [30]writes: One point I was particularly puzzled by [...] was your apparent impression that we're working towards a single, global, monolithic ontology. Danny Ayers [31]says: Another tiresome premise is that Semantic Web somehow involves a global ontology. Nothing could be farther from the truth. Semantic Web technologies (RDF, OWL) allow you to define ontologies which can be entirely local. If there are parts you can share or map, great, if not, no big deal. Alex Wright [32]writes: RDF's primary reason for being is to let local communities of interest create their own ontologies - not to enforce some nefarious New World Order of authoritarian classification systems. Dan Brickley [33]explains a bit further why the semantic web doesn't propose a top-down ontology: Unlike vanilla XML, RDF vocabularies can be freely mixed together in data without prior agreement. So you often see ad-hoc combinations of Dublin Core, RSS1, MusicBrainz, RDF-calendar, FOAF, Wordnet, thesaurus, Geo-info etc etc frequently deployed together, despite the fact that the creators of those various vocabularies barely knew each other. This strikes me as the height of loosly-coupled pragmatism rather than a wide-eyed effort to build a monolithic universal category system. I think that suffices to debunk the idea that the semantic web is about an ontology of everything. [34]The Simple Life. A theme related to the Top Down Bottom Up theme is The Simple Life. In The Simple Life, valuable things happen small scale, and grow organically. Little improvements are valuable. I think The Simple Life is an underlying theme that doesn't often get stated explicitly, but influences many of the discussions of the semantic web. Echoes of the Simple Life theme are found throughout these discussions. Clay Shirky [35]writes: In an echo of Richard Gabriel's [36]Worse is Better argumment, the Semantic Web imagines that completeness and correctness of data exposed on the web are the cardinal virtues, and that any amount of implementation complexity is acceptable in pursuit of those virtues. The argument seems to be that simplicity works, whereas the semantic web tries to impose too much complexity. Shelley Powers [37]writes about simplicity as well: I'll let you in on a little secret: my semantic web is not The Semantic Web. They won't give nobel prizes for it, and it won't be a deafening flash or a blinding roar. It will just make my life a bit easier than what what it is now. Some folks who like the Semantic Web won't necessarily like or agree with my simple, little small 's', semantic, small 'w' web. [38]It Is Growing. The Simple Life, with it's connotations of farming, gardening, has a related theme called It Is Growing. This theme says that the semantic web is growing already, and all we need to do is take care of it, like a garden. In Shirky's essay, his [39]conclusion is exactly this: Much of the proposed value of the Semantic Web is coming, but it is not coming because of the Semantic Web. The amount of meta-data we generate is increasing dramatically, and it is being exposed for consumption by machines as well as, or instead of, people. But it is being designed a bit at a time, out of self-interest and without regard for global ontology. It is also being adopted piecemeal, and it will bring with it with all the incompatibilities and complexities that implies. Shiry seems to think this is opposite to what the semantic web people are proposing. He makes the same mistake as in the Ontology Of Everything theme. Most proponents of the Semantic Web seem to agree with him on this. Shelley Powers [40]writes: I also agree, conditionally, with Clay when he concludes [this]. But Clay's reasoning is flawed if he believes that this isn't the vision shared by those of us who work towards the Semantic Web. Steve Cayzer [41]writes: [about Its Growing] That sounds to me a lot like the vision being pushed by members of W3C that I talk to. [42]RDF Versus XML. A sub-theme of Top Down Bottom Up is RDF Versus XML. In semantic web discussions, people often end up talking about RDF and XML, and which one is better. This, again, is an emotionally and politically loaded theme. Most people agree that RDF is kind of complex. Many people think it's not useful to use something so complex when they can do the same thing in simple XML without worrying about RDF. RDF is seen by some as an overly complex technology, trying to solve a problem XML and HTTP already solve. Paul Ford [43]writes: [The essay claims that] the Semantic Web is a technological pipe dream: an over-specified solution in search of a problem. And Joe Gregorio [44]says: This is exactly the point I made in [45]The Well-Formed Web, that the value that the proponents of the Semantic Web were offering could be achieved just as well with just XML and HTTP, and we are doing it today with no use of RDF, no need to wait for ubiquitous RDF deployment, no need to wait for RDF parsing and querying tools. Dare Obasanjo [46]sees it like this: The difference between the RDF proponents and the XML proponents is fairly simple. In the XML-centric world parties can utilize whatever internal formats and data sources they want but exchange XML documents that conform to an agreed upon format, in cases where the agreed upon format conflicts with internal formats then technologies like XSLT come to the rescue. The RDF position is that it is too difficult to agree on interchange formats so instead of going down this route we should use A.I.-like technologies to map between formats. [...] Thus, if you are an XML practitioner RDF doesn't change much except new transformation techniques and technologies to learn. Shelley Powers, disagrees. Commenting on Dare's comment above, she [47]writes: Dare is saying that we don't need RDF because we can use transforms between different data models; that way everyone can use their own XML vocabulary. This sounds good in principle, but from previous experience I've had with this type of effort in the past, this is not as trivial as it sounds. By not using an agreed on model, not only do you now have to sit down and work out an agreement as to differences in data, you also have to work out the differences in the data model, too. In other words -- you either pay upfront, once; or you keep paying in the end, again and again. [...] However, don't let me stop you from using XML and your own home grown data model and rules and regs. But we won't let this stop us from using RDF and RDF/XML. The point I'm trying to make is this: the semantic web is here. It snuck in quietly while the rest of us were debating. It is viral, slowly putting out little tendrils of applicability throughout the web. The only problem we're really having is that we're not recognizing it now because no huge rocket burst into the air going "Semantic Web is here! Semantic Web is here!" [48]Real World Value. Another recurring theme is Real World Value. The key word here is real. Real World Value says the semantic web is useful for real problems, opposing the idea that the semantic web is a solution looking for a problem. Paul Ford, for example, [49]writes: I believe that there is much of value in the Semantic Web framework which can be applied to real-world problems. [50]Here Today. Most people seem to agree that the semantic web is Here Today. The Semantic Web is already being built, and slowly becoming useful. This theme is often combined with Real World Value. Marc Canter [51]writes: The technologies that surround the Semantic web [...] are here today, while the dream of the semantic web is still years away. Alex Wright [52]puts it this way: [...] the many smaller victories already taking place, like the real, tangible impact of RDF, RSS, and Web services. Though certainly it's a long way from the Alexandrian fantasies of its more breathless boosters, the foundations of the Semantic Web are already taking shape. He [53]continues: Shirky nonetheless bases his argument on a central fallacy: the Semantic Web as monolith, as a single "thing" to be opposed or supported. The Semantic Web is not an all-or-nothing proposition; it is a rubric describing a set of distinct (though related) technologies - RDF, FOAF, OWL, RSS, XML - all of which are designed to improve machine-to-machine communication [...]. And those technologies, like it or not, are already here. Danny Ayers [54]writes: This isn't an either-or situation. There are systems based on Semantic Web technologies working now. The web is working now. Alex Wright [55]quotes this [56]PDF: "It may be that - at least in the short term - that there are many semantic webs rather than The Semantic Web; they may - even in the long term - take us where we need to go." Shelley [57]combines Ontology Of Everything and Here Today and expands It Is Growing to It Is Being Uncovered (one of many themes not discussed in this article). To the Semantic Web people there is no issue about building a global ontology -- it already exists on the web today. Bit by bit of it is uncovered every time we implement yet another component of the model using a common, shared semantic model and language. [58]Useful While We Are Getting There. The Here Today theme is often extended to include Useful While We Are Getting There. This theme says that even though we haven't reached a point yet where a lot of information is available in RDF, the semantic web is already useful. Steve Cayzer [59]writes: Many RDF apps get by perfectly well without any fancy inference rule machinery, exploiting the RDF data model as a handy mechanism for mixing independently created data vocabularies. Danny Ayers [60]agrees: Semantic Web technologies can solve immediate problems in effective ways. You don't need the Semantic Web to find that RDF is a very useful tool for representing web-like, partially structured data. The Semantic Web is a vision, it may even be a pipe dream (hey Clay, you missed a cliche!), but the technologies in that pipe dream are solving real-world problems today. Tom Hoffman [61]states it more strongly: What bothers me about Shirky's essay is that the building blocks of the Semantic Web are useful even if we never achieve nirvana. [62]Clay Misunderstands Syllogisms. In the essay, Shirkey says that the semantic web won't work because it uses syllogisms: statements about something that can be used to deduce certain things (like: Peter is a Belgian. Peter wrote this article. Therefore, this article was written by a Belgian). Clay Shirky [63]writes: The Semantic Web is a machine for creating syllogisms. [...] Syllogisms are Not Very Useful. Other people refute Shirkey's statement that syllogisms don't work. Paul Ford [64]writes: But logical reasoning does work well in the real world - it's just not identified as such, because it often appears in mundane places, like library card catalogs and book indices [...]. Relational databases are used as an example to refute Shirky's point. They also depend on statements about the world, and they have proven themselves to be useful. Dan Brickley [65]says: The Semantic Web project, viewed as an effort to make it easier to publish, mix, share and consume data on the Web, depends on logic in pretty much the same way SQL or UML depend on logic. Danny Ayers [66]agrees: Exactly the same logical arguments can be made just as well against relational databases. And look how useless they are... [67]Doer Versus Talker. The final theme that jumped out for me is Doer Versus Talker. It basicaly says: "put up or shut up". It's not much of an argument, really, but I thought I'd mention it anyway. Paul Ford [68]writes: Quite a bit of work has been done on trust metrics, semantic disambiguation, ontology exchange, triple storage, and query semantics. Some of it is doubtlessly going down the wrong path, but some is equally likely to prove worthwhile. [...] By sneering at a few researchers, Shirky maligns the patient, methodical work of hundreds of others. Marc Canter [69]writes: But this is where we divide up the world between doers and talkers. Danny Ayers, Dan Brickley and LOTS of others are doing what it takes to BUILD the semantic web. Poo pooing it and saying it can't be done doesn't help anything. Comments on this article can be posted [70]here. References 1. http://petervandijck.net/ 2. http://poorbuthappy.com/ease/archives/002335.html 3. http://www.shirky.com/writings/semantic_syllogism.html 4. http://poorbuthappy.com/ease/semantic/#realworldvalue 5. http://poorbuthappy.com/ease/semantic/#whatisit 6. http://poorbuthappy.com/ease/semantic/#topdownbottomup 7. http://poorbuthappy.com/ease/semantic/#ontologyofeverything 8. http://poorbuthappy.com/ease/semantic/#thesimplelife 9. http://poorbuthappy.com/ease/semantic/#itisgrowing 10. http://poorbuthappy.com/ease/semantic/#rdfversusxml 11. http://poorbuthappy.com/ease/semantic/#realworldvalue 12. http://poorbuthappy.com/ease/semantic/#heretoday 13. http://poorbuthappy.com/ease/semantic/#usefulwhilewearegettingthere 14. http://poorbuthappy.com/ease/semantic/#claymisunderstandssyllogisms 15. http://poorbuthappy.com/ease/semantic/#doerversustalker 16. http://xfml.org/ 17. http://www.hyperorg.com/blogger/mtarchive/002172.html 18. http://poorbuthappy.com/ease/semantic/#whatisit 19. http://www.w3.org/People/Berners-Lee/ 20. http://www.ftrain.com/ContraShirky.html 21. http://weblog.burningbird.net/fires/002040.htm 22. http://poorbuthappy.com/ease/semantic/#topdownbottomup 23. http://poorbuthappy.com/ease/semantic/#rdfversusxml 24. http://dannyayers.com/archives/002017.html 25. http://poorbuthappy.com/ease/semantic/#ontologyofeverything 26. http://weblog.burningbird.net/fires/002040.htm 27. http://poorbuthappy.com/ease/semantic/#ontologyofeverything 28. http://www.shirky.com/writings/semantic_syllogism.html 29. http://weblog.burningbird.net/fires/002007.htm 30. http://lists.w3.org/Archives/Public/www-archive/2003Nov/0010.html 31. http://dannyayers.com/archives/002017.html 32. http://www.agwright.com/blog/archives/000787.html 33. http://lists.w3.org/Archives/Public/www-archive/2003Nov/0010.html 34. http://poorbuthappy.com/ease/semantic/#thesimplelife 35. http://www.shirky.com/writings/semantic_syllogism.html 36. http://www.jwz.org/doc/worse-is-better.html 37. http://weblog.burningbird.net/fires/002040.htm 38. http://poorbuthappy.com/ease/semantic/#itisgrowing 39. http://www.shirky.com/writings/semantic_syllogism.html 40. http://weblog.burningbird.net/fires/002007.htm 41. http://jena.hpl.hp.com:3030/blojsom-hp/blog/technologies/semweb/?permalink=A324DB4E341E5CD2F166DF5B773F1CC1.textile&smm=y 42. http://poorbuthappy.com/ease/semantic/#rdfversusxml 43. http://www.ftrain.com/ContraShirky.html 44. http://bitworking.org/news/Skirky_on_the_Semantic_Web 45. http://bitworking.org/news/WellFormedWeb 46. http://www.25hoursaday.com/weblog/CommentView.aspx?guid=1eaabc62-b9f7-4ef2-b980-da29b6015dd5 47. http://weblog.burningbird.net/fires/002040.htm 48. http://poorbuthappy.com/ease/semantic/#realworldvalue 49. http://www.ftrain.com/ContraShirky.html 50. http://poorbuthappy.com/ease/semantic/#heretoday 51. http://blogs.it/0100198/2003/11/08.html#a1964: 52. http://www.agwright.com/blog/archives/000787.html 53. http://www.agwright.com/blog/archives/000787.html 54. http://dannyayers.com/archives/002017.html 55. http://www.agwright.com/blog/archives/000787.html 56. http://www.ht03.org/papers/pdfs/7.pdf 57. http://weblog.burningbird.net/fires/002007.htm 58. http://poorbuthappy.com/ease/semantic/#usefulwhilewearegettingthere 59. http://jena.hpl.hp.com:3030/blojsom-hp/blog/technologies/semweb/?permalink=A324DB4E341E5CD2F166DF5B773F1CC1.textile&smm=y 60. http://dannyayers.com/archives/002017.html 61. http://stone.tuttlesvc.org:880/2003_11_09.html#000371 62. http://poorbuthappy.com/ease/semantic/#claymisunderstandssyllogisms 63. http://www.shirky.com/writings/semantic_syllogism.html 64. http://www.ftrain.com/ContraShirky.html 65. http://lists.w3.org/Archives/Public/www-archive/2003Nov/0010.html 66. http://dannyayers.com/archives/002017.html 67. http://poorbuthappy.com/ease/semantic/#doerversustalker 68. http://www.ftrain.com/ContraShirky.html 69. http://blogs.it/0100198/2003/11/08.html#a1964: 70. http://poorbuthappy.com/ease/archives/002335.html # distributed via <nettime>: no commercial use without permission # <nettime> is a moderated mailing list for net criticism, # collaborative text filtering and cultural politics of the nets # more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body # archive: http://www.nettime.org contact: nettime@bbs.thing.net