- From: Harry Halpin <hhalpin@ibiblio.org>
- Date: Tue, 26 Sep 2006 15:56:37 +0100
- To: public-grddl-wg <public-grddl-wg@w3.org>
Overall, good - I think it's ready to ship as a working draft with the following minor syntactic changes. I'd also *really really* like to have working sample "example" files with data and links to engines like SPARQL and GRDDL implementations which people can run "out of the box", which we don't for much of this tutorial. IMHO that's the main distinction between the tutorial and the use-case document, so let's do it. Anyone up for helping? However, I won't block the release of this document as Working Draft even though the examples need more work. General note: There's lots of sentences like this: "Since XFN values are found on 'a' elements, " that should be formatted like this: "Since XFN values are found on <code>a</code> elements," for transparency. I'll note them as I encounter them with the "*-> <code>".. note. 1) Introduction I really want to start the primer off as "easy-to-read" and am so a bit unsure about the rather technical first sentence,primarily because I don't believe it's exactly correct, as RDF is not really KR or meta-data for many of purposes in our use cases, but just a generalized data format that uses URIs for identity keys for easy aggregation. Instead of starting a primer with a potentially contentioous statement (as well as referencing an "enduring conversation thread" which most readers of the Primer may not care about, let's just simplify. I've also think re-arranging the paragraph breaks a bit makes it bit easier to understand - remember the audience may not be people who read this may not know what "knowledge representation" means nor XSLT mavens. "XML and RDF technologies address very separate and often orthogonal problem spaces: message and structured document formats, meta data, and knowledge representation. Publishers of distributed web content meant for both human and machine consumption have much to gain from standards that enable both technologies. The enduring conversation thread on embedding RDF in XHTML is a demonstration of this itch which (if properly scratched) has the potential to greatly increase the richness of web content specifically, and data in general....(first 3 paragraphs") -> "Publishers of web content meant for both human and machine consumption have much to gain from standards that enable both XML and RDF, and by taking advantage of both publishers can greatly increase the richness of web content specifically and data in general. In particular, there is a need to easily convert between XML vocabularies for specialized purposes, like XHTML for displaying documents to users, and RDF for merging and querying data. GRDDL provides an inexpensive set of mechanisms for bootstrapping RDF content from uniform XML dialects in such a way as to shift the burden of formulating RDF to transformation algorithms written specifically for these dialects. XML Transformation languages such as XSLT are quite versatile in their ability to process, manipulate, and generate XML and the use of XSLT to generate XHTML from single-purpose XML vocabularies is historically celebrated as a powerful idiom for separating structured content from presentation. GRDDL shifts this idiom to a different end: separating structured content from its authoritative meaning (or semantics). The way in which GRDDL empowers authors of web content can be considered somewhat analogous to allowing a non-native speaker learn the spoken form of a new language first, before attempting to master its written form - rather than trying to learn both simultaneously." Letting a transformation algorithm convert XML to RDF lets a user not be forced to recode the same data in both XML and RDF, instead using XML transformation languages such as XSLT that are quite versatile in their ability to generate RDF from XML. " 2) Scheduling Example Who is Jane? :) "In this use case Jane is trying to schedule a meeting with three other friends." -> "In this use case a hypothetical user Jane is trying to schedule a meeting with three of her friends." I know it's pedantic, but let's include the DOCTYPE in our HTML example, since otherwise the user can't cut and paste the example and get it to validate in the W3C Validation Service: Add to first example after sentence: "This microformat uses HTML class attributes to associate particular meanings with elements in the markup:" INSERT-> "<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">" Also, I bet it has to be done on your "after GRDDL" example file as well [1]. And the markup validtor gives this problem with that document [1] as well "on line 19 it contained one or more bytes that I cannot interpret as utf-8" "The profile URI for GRDDL is http://www.w3.org/2003/g/data-view""-> it would good explaining to add both a period to the end of that sentence and then a sentence explaining what profile URIs are and why they're important/nice. However, I seem to have writer's block right now and can't phrase it right...maybe say "They provide a uniform place for links to transformations and documentation to be stored for web-based vocabularies." I'd also add the XHTML DTD to your Embedded RDF example, at least in the linked to example file [2] I think it would be nice if we linked to the W3C GRDDL RDF-in-HTML web service so people know that it's really running code. [3] "Then she needs to add a link elemenk" *-> <code>link</code> "The link element contains" *-> <code>link</code> It might be useful to show them the RDF produced (maybe in Turtle syntax, with a link to Turtle syntax), or at least have a link to it. Finally, the "scheduling calendar" is sort of anti-climatic, because we just demonstrated to people that we can take microformats and embedded RDF and transform it to RDF/XML. I believe the audience for this document may very well not be people who are ardent RDF users or proponents, but general web masters and "desperate perl hackers" that might want to know why producing RDF is useful - and they may read the tutorial up to this point, say "So What?" and then give up. The best way would be by example. I think it's okay to go to Working Draft in final form, but my (without chair hat on) personal preference would really be to show them that the calendar files can be merged and a meeting scheduled via a SPARQL query. 3) Buying a Guitar Example "Stephan wishes to buy a guitar, so decides to check reviews" -> "Stephan wishes to buy a guitar, so he decides to check reviews" "Microformats define simple formats which can easily convert between HTML and RDF through the use of GRDDL" -> I'd link to http://www.microformats.org or add a sentence saying "For more information about microformats please go to http://www.microformats.org" w/i a link because members of our audience may not know what microformats are. "To extract a vCard-RDF from HTML you can use ( @@TODO hCard2vcardrdf.xsl ???)" -> I'll e-mail Norm Walsh about this and cc the listserv. For purely aesthetic reasons: "<?xml version="1.0" encoding="utf-8"?> <rdf:RDF xmlns:rdf ="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:vCard="http://www.w3.org/2001/vcard-rdf/3.0#" > " -> "<?xml version="1.0" encoding="utf-8"?> <rdf:RDF xmlns:rdf ="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:vCard="http://www.w3.org/2001/vcard-rdf/3.0#">" "Another microformat that allows for more information to be gleaned from the document is XFN" -> link to some URI about XFN. "Using values in the rel attribute" *-> <code>rel</code> "Since XFN values are found on 'a' elements," *-><code>a</code>, delete 's. Again, I would like to link in this XFN example to *full* HTML examples, not just have HTML code snippets in the code. "This allows for use to modify the circle of trust from our direct friends to first-order friends of our friends." What is "first-order" friends? Friends of Friends? Use different wording than "direct friends" and "first order friends" to explain whatever you're getting at. "seed URL" -> "seed URI" Also explain what you mean by "seed URI" - I think I know but our audience might not. "That FoaF file will then give us an additional list of URLs " -> "That FoaF file will then give us an additional list of URIs ". Add a sentence explaining that "FOAF (Friend of a Friend) is a popular RDF vocabulary for describing social networks of friends and personal data, used by services such as Livejournal." Then link to FOAF and Livejournal.[4,5]. Using "URL" is not quite as cool as using "URI" :) "power of the rel="me"" -> "power of the <code>rel="me"</code>" "The Del.icio.us links" -> The <a href="http://del.icio.us">Del.icio.us</a> links <a>because of the rel="me"</a> While I won't _block_ the transition of this to Working Draft based on the fact that it doesn't have links to running code, I might add this example does not. I'd like to *run* the example and use this rel="me" technique to consolidate identity. Does it actually work? Is it implemented? If so, where? I vaguely remember reading something Tantek wrote but can't find the link. "The Del.icio.us links could be encoded into RDF" -> This is rather hypothetical and so more suited for the use-case document. Can we do this in reality? And it would be nice to have a "guitar-rocker45" user as well on del.icio.us to help the code. "owner "Stephan Smith" who has an RDF-vCard at "example.org" and has data in other places on other services such as claimid.com and guitar-rocker.com. All of these can be merged to form a bigger picture of "Stephan Smith" at "example.org"" -> again, all rather hypothetical. "Guitar site" -> please make some sample files for this, and possibly a more informative ficitonal name, such as "GuitarExample.org". Again, won't block, but making sample files for this needs to be done. "rel="me" value" *-> <code>rel="me"</code> "Using GRDDL is it possible to glean information about the guitar in the form of product specifications supplied by the manufacture and reviews from site members." -> "Using GRDDL we can automatically glean information about the guitar in the form of product specifications supplied by the manufacture and reviews from site members." It's not "possible" - we can :) "SPARQL engine " -> please add a sentence explaining SPARQL, like "<a href="http://www.w3.org/TR/rdf-sparql-query/">SPARQL (The SPARQL Protocoal and RDF Query Language)</a> is a query language for RDF similar in spirit to the use of the query language SQL for relational databases." "If Stephan was looking for a Guitar" -> "If Stephan was looking for a guitar" "Using a seeded list of XFN URLs given by Stephan that are converted to FoaF, we can match the URLs to any URLs from the FoaF generated from the guitar reviews. Now we have a list of members that Stephan trusts relative to the guitar site. We can then match the URLs of the reviewers to the URLs in the XFN list." -> "Using a seeded list of XFN URIs given by Stephan that are converted to FoaF, we can match the URIs from his XFN list to any URIs from the FoaF generated from the guitar reviews. Now we have a list of reviewers that Stephan trusts on the guitar review site." Ala the SPARQL queries, I'd like to actually run them on some data but I can't because we don't have any sample files attached that we can actually use GRDDL on to generate RDF. It would also be nice, if we really want people to learn from this tutorial, to assume they don't have SPARQL installed, and give them a link to a SQARQL implementation such as the one with Jena. "This SPARQL result is in XML or JSON and can easily be consumed by another application." -> Again, I won't block, but we really should not just say these sort of things but provide a concrete example. Perhaps an example stylesheet that generates HTML would be good, or a link to something like the Tabulator[6]. I realize I did lots of complaining about there not being links to concrete example files. However, over all the content is really good and I'm quite pleased. I'm absolutely packed and busy till the end of the week, but after the weekend I'm willing to sit down and devote a whole day to work closely with someone (Brian Suda I assume, but anyone else whose interested) in manufacturing these example files.) [1]http://research.talis.com/2006/grddl-wg/robin-hcal-grddl.html [2]http://research.talis.com/2006/grddl-wg/david-erdf.html [3]http://www.w3.org/2003/11/rdf-in-xhtml-demo [4]http://www.foaf-project.org/ [5]http://community.livejournal.com/ljfoaf/ [6]http://dig.csail.mit.edu/2005/ajar/release/tabulator/0.7/tab.html -- -harry Harry Halpin, University of Edinburgh http://www.ibiblio.org/hhalpin 6B522426
Received on Tuesday, 26 September 2006 19:57:07 UTC