From ryan at technorati.com Thu Feb 1 10:08:51 2007 From: ryan at technorati.com (Ryan King) Date: Thu Feb 1 10:08:56 2007 Subject: [uf-dev] nested @class="value" properties In-Reply-To: <45C15152.6030703@pallas.us> References: <45C15152.6030703@pallas.us> Message-ID: On Jan 31, 2007, at 6:32 PM, Derrick Lyndon Pallas wrote: > I asked about this in IRC today and Tantek asked me to send an > email to the list. The question is: is @class="value" recursive. > For instance, > >
> type > > foo > quux > bar > >
> > In this example, is the value of property "foo quux bar" or just > "foo bar"? Personally, I would expect it to be "foo bar" because > that's what I'd expect from the following fragment: See, I would expect it to bee "foo quux bar", because all three are text nodes of a node that has @class="value". I don't know if value excerpting should be recursive or not, because I don't know how authors would expect it to work. Do you have any examples from the wild of recursive value excerpting? -ryan -- Ryan King ryan@technorati.com From ryan at technorati.com Thu Feb 1 10:10:05 2007 From: ryan at technorati.com (Ryan King) Date: Thu Feb 1 10:10:09 2007 Subject: [uf-dev] and @xml:base In-Reply-To: <45C1562C.9080505@pallas.us> References: <45C1562C.9080505@pallas.us> Message-ID: <1B77882E-8EDF-4F1E-BB4C-812721CF2D69@technorati.com> On Jan 31, 2007, at 6:53 PM, Derrick Lyndon Pallas wrote: > What is the href of the anchor in the following document, which has > a base URL of "http://www.example.com/" > > > "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> > > > Confusing > > > > text > > > > Is it or www.example.com/bar/link.xml>? In other words, how does > interact with @xml:base? Is this question outside the scope of this > mailing list? ~D Honestly, I think the right answer is to ignore xml:base. Browsers don't implement it interoperably because it isn't specified in XHTML. I think we should remove the xml:base tests from the suite. (sorry, I was the one who put them there) -ryan -- Ryan King ryan@technorati.com From brian.suda at gmail.com Thu Feb 1 10:27:39 2007 From: brian.suda at gmail.com (Brian Suda) Date: Thu Feb 1 10:27:43 2007 Subject: [uf-dev] nested @class="value" properties In-Reply-To: References: <45C15152.6030703@pallas.us> Message-ID: <21e770780702011027m7da82519qcbcc36fc333ea06c@mail.gmail.com> On 2/1/07, Ryan King wrote: > On Jan 31, 2007, at 6:32 PM, Derrick Lyndon Pallas wrote: > > > I asked about this in IRC today and Tantek asked me to send an > > email to the list. The question is: is @class="value" recursive. > > For instance, > > > >
> > type > > > > foo > > quux > > bar > > > >
> > > > In this example, is the value of property "foo quux bar" or just > > "foo bar"? Personally, I would expect it to be "foo bar" because > > that's what I'd expect from the following fragment: > > See, I would expect it to bee "foo quux bar", because all three are > text nodes of a node that has @class="value". I would actually expect 'foo quux barfoobar' it would find the first value="foo quux bar" and then append the next two values "foo quux barfoobar" to the string. -brian -- brian suda http://suda.co.uk From derrick at pallas.us Thu Feb 1 19:10:47 2007 From: derrick at pallas.us (Derrick Lyndon Pallas) Date: Thu Feb 1 19:11:04 2007 Subject: [uf-dev] nested @class="value" properties In-Reply-To: <21e770780702011027m7da82519qcbcc36fc333ea06c@mail.gmail.com> References: <45C15152.6030703@pallas.us> <21e770780702011027m7da82519qcbcc36fc333ea06c@mail.gmail.com> Message-ID: <45C2ABB7.1070105@pallas.us> Brian Suda wrote: > On 2/1/07, Ryan King wrote: >> See, I would expect it to bee "foo quux bar", because all three are >> text nodes of a node that has @class="value". > > I would actually expect 'foo quux barfoobar' it would find the first > value="foo quux bar" and then append the next two values "foo quux > barfoobar" to the string. > > -brian According to , "Sometimes only part of an element which is the equivalent for a property should be used for the value of the property. ... For this purpose, the special class name "value" is introduced to excerpt out the subset of the element that is the value of the property." Is "value" itself a property or just a marker? Because if it is a property in its own right, then it should apply to itself, i.e. it is recursive. If it is a marker, then I think Brian may be right and the wording should be changed to indicate this. (In hCard, it appears to be a property.) Taking a look at xhtml2vcard.xsl, it just takes the normalization of the text() nodes. However, the value of the element #ref is clearly "foo bar" in my original example and that's what I expect the include-pattern to include. ~D From bewest at gmail.com Thu Feb 1 19:11:28 2007 From: bewest at gmail.com (Benjamin West) Date: Thu Feb 1 19:11:32 2007 Subject: [uf-dev] nested @class="value" properties In-Reply-To: <45C15152.6030703@pallas.us> References: <45C15152.6030703@pallas.us> Message-ID: <8ad71be30702011911g5132ecfdt6f74f708c200f079@mail.gmail.com> This is really an include-pattern issue, isn't it? Should it be added to the wiki at ? The idea behind include is that this block: --------- >
> type > > foo > quux > bar > >
> Should be identical to... >
> foo > quux > bar >
> this block: -------------- >
> type > > > > ------------- >From : "the object include completely replaced by the subtree that it references" So that much is clear. However, what about the #ref element and it's children? Should a parser produce the same result for #ref as when it's included? If the results are different, this could place extra burden on authors. What does XSLT do if you ask for all the text nodes with class="value" for a property? That would be what works currently. Ben From bewest at gmail.com Thu Feb 1 19:18:16 2007 From: bewest at gmail.com (Benjamin West) Date: Thu Feb 1 19:18:20 2007 Subject: [uf-dev] nested @class="value" properties In-Reply-To: <8ad71be30702011911g5132ecfdt6f74f708c200f079@mail.gmail.com> References: <45C15152.6030703@pallas.us> <8ad71be30702011911g5132ecfdt6f74f708c200f079@mail.gmail.com> Message-ID: <8ad71be30702011918j61215acco47620b0b899d5d67@mail.gmail.com> Derrick has a good point with his recursive interpretation. From a usability perspective, it is the most consistent model. The value of the fragment is "foo bar". With @value treated recursively, when the other examples are evaluated, the result is still "foo bar", making this option the most consistent authoring experience. On 2/1/07, Benjamin West wrote: > This is really an include-pattern issue, isn't it? Should it be added > to the wiki at ? > > The idea behind include is that this block: > --------- > >
> > type > > > > foo > > quux > > bar > > > >
> > > > Should be identical to... > > >
> > foo > > quux > > bar > >
> > > > this block: > -------------- > >
> > type > > > > > > > > > ------------- > From : > "the object include completely replaced by the subtree that it references" > > So that much is clear. However, what about the #ref element and it's > children? Should a parser produce the same result for #ref as when > it's included? If the results are different, this could place extra > burden on authors. What does XSLT do if you ask for all the text > nodes with class="value" for a property? That would be what works > currently. > > Ben > From andy at pigsonthewing.org.uk Fri Feb 2 00:32:57 2007 From: andy at pigsonthewing.org.uk (Andy Mabbett) Date: Fri Feb 2 08:16:32 2007 Subject: [uf-dev] nested @class="value" properties In-Reply-To: <8ad71be30702011918j61215acco47620b0b899d5d67@mail.gmail.com> References: <45C15152.6030703@pallas.us> <8ad71be30702011911g5132ecfdt6f74f708c200f079@mail.gmail.com> <8ad71be30702011918j61215acco47620b0b899d5d67@mail.gmail.com> Message-ID: In message <8ad71be30702011918j61215acco47620b0b899d5d67@mail.gmail.com>, Benjamin West writes >On 2/1/07, Benjamin West wrote: >> >
>> > type >> > >> > foo >> > quux >> > bar >> > >> >
>> > >> >> Should be identical to... >> >> >
>> > foo >> > quux >> > bar >> >
[top posting corrected] >Derrick has a good point with his recursive interpretation. From a >usability perspective, it is the most consistent model. The value of >the fragment is "foo bar". Where are you getting the space from? I make it "foobar" in each of the above. -- Andy Mabbett * Say "NO!" to compulsory ID Cards: * Free Our Data: * Are you using Microformats, yet: ? From rbach at rbach.priv.at Fri Feb 2 09:06:28 2007 From: rbach at rbach.priv.at (Robert Bachmann) Date: Fri Feb 2 09:04:37 2007 Subject: [uf-dev] hcalendar test suite patch In-Reply-To: <19b880020701191923x5f80cddbj81c8183f91402a49@mail.gmail.com> References: <19b880020701191923x5f80cddbj81c8183f91402a49@mail.gmail.com> Message-ID: <45C36F94.7030409@rbach.priv.at> Mike Samuel wrote: > Attached is a patch for the tests at > http://hg.microformats.org/tests?cmd=manifest;manifest=8323e6e893f8f60ded243c97a060f4befd4a2b20;path=/hcalendar/ Thanks, I've applied it. -- Robert Bachmann (OpenPGP KeyID: 0x4A5CCF10) From bewest at gmail.com Fri Feb 2 13:15:43 2007 From: bewest at gmail.com (Benjamin West) Date: Fri Feb 2 13:15:47 2007 Subject: [uf-dev] nested @class="value" properties In-Reply-To: References: <45C15152.6030703@pallas.us> <8ad71be30702011911g5132ecfdt6f74f708c200f079@mail.gmail.com> <8ad71be30702011918j61215acco47620b0b899d5d67@mail.gmail.com> Message-ID: <8ad71be30702021315n3725c07fi6997604a70ee52c6@mail.gmail.com> > >Derrick has a good point with his recursive interpretation. From a > >usability perspective, it is the most consistent model. The value of > >the fragment is "foo bar". > > Where are you getting the space from? I make it "foobar" in each of the > above. > > > -- > Andy Mabbett Correct: foobar. From faaborg at mozilla.com Mon Feb 5 15:58:20 2007 From: faaborg at mozilla.com (Alex Faaborg) Date: Mon Feb 5 15:58:40 2007 Subject: [uf-dev] The UI of microformat detection Message-ID: [note: reposting from microformats-discuss] I've posted a variety of conceptual mockups of microformat detection in Firefox 3 to my blog: http://blog.mozilla.com/faaborg/2007/02/04/microformats-part-4-the- user-interface-of-microformat-detection I would love to get some feedback on what designs you think work well, or don't, and if there are any other UI ideas I should be considering. Cheers, -Alex From faaborg at mozilla.com Mon Feb 5 16:06:58 2007 From: faaborg at mozilla.com (Alex Faaborg) Date: Mon Feb 5 16:07:04 2007 Subject: [uf-dev] Re: [uf-discuss] The UI of microformat detection In-Reply-To: <21e523c20702040436u10647966g884823b90e841c32@mail.gmail.com> References: <25301E2B-54BC-4CDE-B382-A344DFBD575F@mozilla.com> <21e523c20702040436u10647966g884823b90e841c32@mail.gmail.com> Message-ID: <860A01F6-72D5-4D2E-B35C-877C66EE1611@mozilla.com> > Would it possible/practical to have a configurable "post this > microformat" option. Yeah, that is the general model we are going for. In the mockups I primarily show client side apps, but they could just as well be Web apps (some of the google services I show in the mockups are), and the entire framework will be completely extensible. > I could post a hAtom entry to by > blogging service that would preformat a blog post referencing it I'm not sure if we are going to have native support for hAtom detection, since only a limited number of Firefox users are also active bloggers, but this would be a great extension for people who do blog. We are however thinking about how we can extend text areas to include inline microformat creators, and Tantek suggested that these might have access to all of the microformats you have run into recently, saying "it would be cool if browser already knew all of your recent proper nouns." So for instance if you are blogging about an event you are about to attend, you could easily add the hCalendar entry for the event to your post. -Alex On Feb 4, 2007, at 7:36 AM, David Janes wrote: > On 2/4/07, Alex Faaborg wrote: >> I've posted a variety of conceptual mockups of microformat detection >> in Firefox 3 to my blog: >> >> http://blog.mozilla.com/faaborg/2007/02/04/microformats-part-4-the- >> user-interface-of-microformat-detection >> >> I would love to get some feedback on what designs you think work >> well, or don't, and if there are any other UI ideas I should be >> considering. > > Alex, > > Would it possible/practical to have a configurable "post this > microformat" option. Then when I see a hCard, I could post it to a > webservice that would accept it; I could post a hAtom entry to by > blogging service that would preformat a blog post referencing it; I > could send a hCal to my calendar service... > > Regards, etc... > David > > -- > David Janes > Founder, BlogMatrix > http://www.blogmatrix.com > http://blogmatrix.blogmatrix.com > _______________________________________________ > microformats-discuss mailing list > microformats-discuss@microformats.org > http://microformats.org/mailman/listinfo/microformats-discuss From davidjanes at blogmatrix.com Tue Feb 6 04:28:20 2007 From: davidjanes at blogmatrix.com (David Janes) Date: Tue Feb 6 04:28:25 2007 Subject: [uf-dev] Re: [uf-discuss] The UI of microformat detection In-Reply-To: <860A01F6-72D5-4D2E-B35C-877C66EE1611@mozilla.com> References: <25301E2B-54BC-4CDE-B382-A344DFBD575F@mozilla.com> <21e523c20702040436u10647966g884823b90e841c32@mail.gmail.com> <860A01F6-72D5-4D2E-B35C-877C66EE1611@mozilla.com> Message-ID: <21e523c20702060428o35055014w3aeaff3e365617d0@mail.gmail.com> On 2/5/07, Alex Faaborg wrote: > > Would it possible/practical to have a configurable "post this > > microformat" option. > > Yeah, that is the general model we are going for. In the mockups I > primarily show client side apps, but they could just as well be Web > apps (some of the google services I show in the mockups are), and the > entire framework will be completely extensible. > > > I could post a hAtom entry to by > > blogging service that would preformat a blog post referencing it > > I'm not sure if we are going to have native support for hAtom > detection, since only a limited number of Firefox users are also > active bloggers, but this would be a great extension for people who > do blog. Hi Alex, Other possible uses for hAtom" - nicely printing a blog entry (or newpaper article) - capturing text to a personal store - sending to a social bookmarking site How about a configuration panel with checkboxes to decide what microformats one wishes to recognize? Or even better, enter your own "semantic HTML" information and then one could see all sorts of deployments, in situations such as intranets. The nice thing is that for "other" microformats ... and I assume there's going to be more and more of these ... the only things needed are (I think): - a name - a root class - an icon (optional to some degree) And then hopefully an ecosystem of developers and users will develop. I was attending a conference on Sunday called TransitCamp [1] talking about re-imagining how the website for our local transit would work. Needless to say I was pushing the whole concept of microformats/semantic HTML; one could imagine capturing routes, stations, store hours, etc. all from within the browser to do seamless cross-site trip planning. Regards, etc... [1] http://blogmatrix.blogmatrix.com/:entry:blogmatrix-2007-02-06-0000/ -- David Janes Founder, BlogMatrix http://www.blogmatrix.com http://blogmatrix.blogmatrix.com From steveivy at gmail.com Tue Feb 6 12:07:27 2007 From: steveivy at gmail.com (Steve Ivy) Date: Tue Feb 6 12:07:52 2007 Subject: [uf-dev] XFN Test Data Message-ID: Hi folks, While working on figuring out how to help add XFN support to mofo (http://mofo.rubyforge.org/), I found it useful to create a page with test data for XFN. http://deliciouslymeta.com/projects/xfn/test_data.html This page contains links with all the profile relationships (per http://gpmg.org/xfn/11), most if not all of the two-component combinations, and most if not all of the invalid two-component combinations. Tantek had some thoughts, here: http://rbach.priv.at/Microformats-IRC/2007-02-05#T163544 Ryan has expressed an interest in adding these to the test suite, which I'm fine with, one folks feel like it's covered the bases. --Steve -- Steve Ivy http://redmonk.net From brian.suda at gmail.com Wed Feb 7 07:45:41 2007 From: brian.suda at gmail.com (Brian Suda) Date: Wed Feb 7 07:45:57 2007 Subject: [uf-dev] HG XSLT Updates Message-ID: <21e770780702070745k260ffe4bi211def1810ab89d5@mail.gmail.com> Robert and I have been IMing back and forth with new ideas and improvements to the XSLT files. I took some time yesterday and brought alot of the code up to XSLT2 because of the ability to pass-back an node-set as a variable. So it allows to pass more than one-string back to the calling template. The idea behind all of this is to have a master mf-template.xsl to do all the heavy-lifting and then if you want to create a new template to convert hCard->CVS it is as easy as just calling functions to extract the data. No knowledge of the underlying mechanices of WHERE exactly to decode microformats from, but just that the function returns the string(s) you need. So i checked all the code in, but we are probably going to back-port alot of it to just EXSLT. XSLT2 can easily be added back in, the problem is that only saxon supports it at the moment, whereas XSLT1 with EXSLT is more readily supported. In the future we can migrate back to XSLT2 (we already have working code). Just a heads-up that in the weeks to come there will be some changes to the XSLT templates and the corresponding test cases as well. Robert has also been working on a very nice perl script to run the tests and display errors in a nice coloured format. If anyone has any questions/concerns just let us know. -brian -- brian suda http://suda.co.uk From connolly at w3.org Wed Feb 7 09:34:49 2007 From: connolly at w3.org (Dan Connolly) Date: Wed Feb 7 09:35:18 2007 Subject: [uf-dev] HG XSLT Updates In-Reply-To: <21e770780702070745k260ffe4bi211def1810ab89d5@mail.gmail.com> References: <21e770780702070745k260ffe4bi211def1810ab89d5@mail.gmail.com> Message-ID: <1170869689.7497.388.camel@dirk> On Wed, 2007-02-07 at 15:45 +0000, Brian Suda wrote: [...] Sounds like good stuff. Thanks for the update. > So i checked all the code in, Indeed, I see.. "updated hCard and hCalendar templates for XSLT2 ..." http://hg.microformats.org/x2v?cmd=changeset;node=f53fcd962608;style=gitweb and a couple after that; the latest seems to be: author Robert Bachmann Wed Feb 07 16:40:52 2007 +0100 (113 minutes ago) changeset 624ddd73e349 -- Dan Connolly, W3C http://www.w3.org/People/Connolly/ D3C2 887B 0F92 6005 C541 0875 0F91 96DE 6E52 C29E From jammie.knight at gmail.com Thu Feb 8 13:55:05 2007 From: jammie.knight at gmail.com (Jamie knight) Date: Thu Feb 8 13:55:13 2007 Subject: [uf-dev] hCard to Gmail Converter Message-ID: hiya, This is my first message to this list so I hope I don't make any mistakes! If I do please be gentle! A few months ago, I was clicking around with operator, trying to find a way to get a hCard into gmail, as gmail is the main e-mail service I use. After a lot of looking around and a bit of searching I found that there was a process which I could use to bring hCard formatted data into my gmail account. This process was quite simple; I have documented it more fully on my blog post entitled hCard to Gmail (http://jkg3.com/Journal/74/hcard-to-gmail) . In short, I used one service to convert the hCard to a vCard, then another service to convert this vCard to the CSV file the gmail contact importer required. >From here, I sought a way to automate this process. I downloaded copies of both scripts (they are both open source) and proceeded to hack them together. This was my first time with XLST, and my first time producing a CSV file within PHP, so after a lot of learning, I managed to "functionize" the CSV producing part of the code, and place it into the hCard to vCard code instead of it producing a file. I tried to upload this to my server, then I discovered that there was a different XLST service on my host to my local testing sever, after a few hours of fiddling around I decided that I did not know enough about XLST to be able to fix this problem, and instead required something that I could understand easier. So, my next move was to find a "simpler" way to extract my hCard data into a php array. Luckily for me, there was an exact tool for this staring me in the face. hKit hKit for those who are unaware, is a very useful bit of code by Drew McLellan which take a URL as input, and produces a PHP array of hCard objects as an output. find out more about hKit at http://allinthehead.com/hkit/ After downloading and playing with this script, I rewrote my converter, and produced a service which works. The process is still hCard ? vCard ? CSV I relise that this is not the best way to proceed so I am now learning more about producing gmail formatted CSV files. More information on the service (and a bookmarklet) is available at www.jkg3.com/stuff and i would welcome any input form the community regarding further development of this service. Thanks Jamie Knight From ryan at technorati.com Wed Feb 21 11:06:19 2007 From: ryan at technorati.com (Ryan King) Date: Wed Feb 21 11:06:26 2007 Subject: [uf-dev] Re: [uf-discuss] XOXO to JSON and back In-Reply-To: <6CE341F8-D3A2-4F26-AC80-CB8C854796CF@mac.com> References: <6CE341F8-D3A2-4F26-AC80-CB8C854796CF@mac.com> Message-ID: Moving to -dev.. On Feb 21, 2007, at 3:10 AM, Kevin Marks wrote: > A preliminary go at a bidirectional XOXO to JSON service: > > http://kevinmarks.com/cgi-bin/xoxotojson.py?url=http://kevinmarks.com > > and back again: > > http://kevinmarks.com/cgi-bin/jsontoxoxo.py?url=http%3A// > kevinmarks.com/cgi-bin/xoxotojson.py%3Furl%3Dhttp%3A//kevinmarks.com > > change the url parameters as it suits you, let me know what you think Very nice. A nit to pick, though: the Content-Type header is wrong for the first one. -ryan -- Ryan King ryan@technorati.com From ehs at pobox.com Fri Feb 23 09:59:45 2007 From: ehs at pobox.com (Edward Summers) Date: Fri Feb 23 10:00:21 2007 Subject: [uf-dev] operator source code Message-ID: <1B019F81-5B81-4927-82AC-06C2550249DD@pobox.com> Not sure if this has been mentioned on here yet but: http://www.kaply.com/weblog/2007/02/21/operator-source-code-is- finally-available/ //Ed