- From: Jon Phipps <jphipps@madcreek.com>
- Date: Tue, 27 May 2008 13:20:37 -0400
- To: Ed Summers <ehs@pobox.com>
- Cc: SWD Working SWD <public-swd-wg@w3.org>, Diego Berrueta <diego.berrueta@fundacionctic.org>
Thanks Ed. Great review! More inline below... --Jon On May 26, 2008, at 9:50 PM, Ed Summers wrote: > > Hi Jon & Diego: > > I finally got round to reviewing the recent changes [1]. Since most of > the changes this time included rewriting URLs I focused mainly on > them. > > In general I was wondering if the examples and the steps should to be > consistent in the use of URLs as they were in the previous working > draft of March 14th, 2006 [2]. At the moment the examples for each of > the recipes seem to use a location of: > > http://www.w3.org/2006/07/SWD/recipes/examples-20080421/ > > Whereas the steps for each recipe use: > > http://yourhost.com/examples/ > > I think I would prefer the steps in all the recipes to be consistent > in the host and paths from the examples. This would include the sample > HTTP requests as well. At the very least I would recommend using > example.com or what have you, instead of yourhost.com. I _think_ that the only place where we use 'yourhost.com' is in the test messages and response for each recipe. My help-desk experience tells me that we can't use the resolvable W3-space URI because people will just copy the test URI, get the documented response (which will tell them nothing of value) and call it a day. We could certainly use example.org, although it would seem to me to be less clear that they should substitute the correct host location of their own files. On the other hand, if I'm correctly assuming a lack of understanding to that degree, nothing could be sufficiently clear. Unless Diego or Ralph has an objection, I'm fine with changing 'yourhost.com' in the released doc to 'example.org' wherever it occurs. > > In addition there were some problems with the files being served up > at w3.org: > [snip] > I think if it's too much trouble to get the server side files correct, > perhaps the easy thing to do would be to simply use example.com > throughout. I think having non-resolvable examples would be preferable > to resolvable but broken examples. We emphatically agree!! (says Jon, presumptuously speaking for Diego as well) -- correct examples are definitely what we're trying to provide. Thank you so much for the fine-grained checking. I've fixed all instances of 'isegserve'-based URIs in the documents, replaced the example1.rdf and example2.rdf files, and corrected all instances of incorrect capitalization of 'class'. > > LastIy I got to thinking that RDFa + GRDDL would provide a really nice > way of publishing human and machine readable versions of a RDF > vocabulary in one document without having to play around with any > content negotation. But I realize this is the 11th hour, and there > probably isn't much time to do this. Perhaps it could be another note, > if someone were to find time to do this? I guess RDFa hasn't made it > through the REC process yet, so maybe this is premature. An excellent point, but I think we already decided that providing recipes for RDFa and GRDDL was out of scope (for this version of the document at least) and inserted the following to cover that particular portion of our posterior: "Finally, it should be noted that the Recipes described in this Cookbook are not the only way to publish a vocabulary or ontology for use by Semantic Web applications. RDFa and its cousin GRDDL may in the near future provide an effective method for publishing documents for use by both people and machines. But a useful discussion of RDFa and GRDDL is well beyond the scope of this document." > > //Ed > > [1] http://www.w3.org/2006/07/SWD/recipes/draft-20080421.html > [2] http://www.w3.org/TR/2006/WD-swbp-vocab-pub-20060314/ >
Received on Tuesday, 27 May 2008 17:21:25 UTC