- From: <noah_mendelsohn@us.ibm.com>
- Date: Wed, 13 Sep 2006 19:13:18 -0400
- To: "T.V Raman" <raman@google.com>
- Cc: www-tag@w3.org
Raman, Mostly this looks fine and ready to go. I did notice a few more purely editorial issues, though the first two are significant and need to be fixed. That one is that the status section has a statement "An informal guide to previous discussion of this topic is available and may be useful to reviewers of this draft." The link is in fact to the draft itself. I strongly suspect that this entire sentence traces to your perhaps having adopted the metadataInUri draft finding as a starting point. It had such a link to informal discussion, so my guess is that this entire sentence is to be deleted. Anyway, the link is bogus. The other significant comment is that you also have carried over the usual reference to RFC 2119, but in fact the formal uppercase SHOULD, MUST, MUST NOT, etc. are not used in the finding. I suggest that the sentence in the status, and the corresponding bibliography entry be dropped. (Warning: I do comment later on one place in which the finding uses the odd capitalization Should Not; my recommendation is to lowercase that one, but if you uppercase it, then obviously you should keep the RFC 2119 reference after all.) Other less significant comments follow. I think most of these are worth fixing, but I leave that entirely to you. There are a fair number of them. If you want to fix only the two above and ship, I'm fine with that. I don't think the suggestions below are the sort of thing that would trigger another round of formal reviews. Most are typos or barely above that in significance. I'm afraid these are in the order I discovered them, which is not necessarily document order. If you find that you do want to fix these but that the editing is burdensome, I'll be glad to share the load. We can pick a time when the CVS copy is clean and I'll be glad to help patch up some of these, particularly where the issues are just punctuation and spacing. --- From 2.1.1 list item 5: "In contrast, links meant for machine consumption, e.g., Atom/RSS feeds might use the HTML link element." I think that this is hard to parse, partly due to the position of the commas and partly due to the wording. I think what you meant was that Atom/RSS feeds are examples of resources to which an HTML document might link for purposes of machine consumption. My eye tends to hang up on the comma after the e.g., and tries to parse as "Atom/RSS feeds might use the HTML link element", before backtracking and realizing that's not intended. Proposed rewording: "In contrast, links meant for machine consumption, e.g. links to Atom/RSS feeds, might use the HTML link element." --- From 2.1.1: "If no content negotiation is in place, Serve a canonical representation of the content at http://example.com/ubiquity/resource" The word "Serve" should be lowercase. --- The punctuation that terminates list items is inconsistent throughout the finding. For example in the abstract, there's a list of: Representations appropriate for different delivery contexts, Representations in different languages, I'm not sure why these end with commas. The style guides I've seen mainly say to use periods for list items if and only if they are each complete sentences, and otherwise to use no punctuation at all. In any case, the commas look suspect. --- In chapter 3: "As can be seen from the use-cases and suggested solutions enumerated in the previous section, pointers to Web Resources (URIs) can either: * Be canonical URIs, i.e., have no context hard-wired. * Can encapsulate partial context, e.g., encapsulate language, * Encapsulate multiple context bits, e.g., language and device profile, * Capture all context, i.e., the creator of the URI guarantees that all state is completely captured by the URI." I think the word "Can" should be deleted from the second item in the list. --- In the abstract: "This document explores the issues that arise in this context, and attempts to define best practices that help toward: * Preserve the One Web while enabling content publishing to a multiplicity of delivery contexts. * Enable automatic discovery of the available representations. * Enable the creation of RESTful URIs that remain representation agnostic while delivering the correct end-user experience." I think it would be better grammar to delete the word "toward", so the list can be interpreted as "practices that help Preserve the One Web", "practices that Enable automatic discovery", etc. --- In the very first sentence of the introduction: "There has always been a need to serve user-agent specific contents for a given URI --- thus highlighting the distinction..." you have approximated an emdash with three hyphens, which looks sort of bad. For your convenience, this is a real emdash — you can copy it if you like, or else use the entity —. A useful guide to dashes and their use in HTML is at [1]. -- I hesitate to raise this, because it's more a question of patching around bugs than doing things right, but both Firefox and IE tend to render the text in <code> elements too small relative to the text around them. I've seen this discussed on the Web, but I'm not sure why the problem occurs. Anyway, to get around this, I reluctantly made the drafts of metadata in URI point to a stylesheet (which happens to be metadataInURI.xsl in the same doc directory as your finding) which has in it: code {font-family: monospace; font-size: 130%;} pre {font-family: monospace; font-size: 120%;} in the generated HTML <style>. I find this makes the text sizes line up better. Others may disagree. --- Chapter 3: "URIs are cheap, we suggest creating as many distinctive URIs as is meaningful." I believe that the correct punctuation to join the independent clauses is a semicolon: "URIs are cheap; we suggest creating as many distinctive URIs as is meaningful." --- Similar issue also in chapter 3: "The hyperlink structure of the Web is crucial for content discovery; When creating..." One option is to leave the semicolon, I think, but then the word "When" following the semicolon should be lowercase. Otherwise, I think you could make the semicolon a period, and leave the "When" in caps. --- Also in that same list, the 3rd item has: "in the absence of user context; Or equivalently" There too, I think the "Or" should be lowercase. --- Chapter 3, same list: "Contrast these findings with the metadata in URIs and state finding which each enumerate use cases where user context Should Not be encapsulated by URIs" That "Should Not" has only initial caps. I think it should either be all caps or no caps. My vote, as signaled above, would be to make them lowercase, as nowhere else in the finding do we use RFC 2119 directives. If you want to make them uppercase, then obviously you should not follow my advice above on getting rid of the RFC 2119 reference in the status and biblio. --- Chapter 4: "URIs are cheap, create them as needed, publish them to the Web, and ensure that they are appropriately linked in to the rest of the Web. Thus, each representation of interest should get it's own URI and there should be one additional URI representing the generic resource." More independent clause punctuation. I think that should either be "URIs are cheap. Create..." with a period, or else "URIs are cheap; create..." with a semicolon. The comma seems incorrect to me. --- Chapter 4: "Thus, given one of the alternatives for a resource,ensure " Missing space before the word ensure. --- That's it. Sorry I didn't catch more of these before. Again, I'm very happy with the substance of the finding. Thanks for the hard work on this. Noah [1] http://alistapart.com/stories/emen/ -------------------------------------- Noah Mendelsohn IBM Corporation One Rogers Street Cambridge, MA 02142 1-617-693-4036 --------------------------------------
Received on Wednesday, 13 September 2006 23:13:35 UTC