- From: Matthew Raymond <mattraymond@earthlink.net>
- Date: Wed, 23 Feb 2005 11:53:43 -0500
Ian Hickson wrote: > On Tue, 22 Feb 2005, Matthew Raymond wrote: >>>Please discuss hCalendar and hCard issues with Tantek; >> >>Why? > > Because Tantek is the editor of those specs. I fail to see how I could have a reasonable debate over whether hCalendar/hCard should be included in a WHATWG spec without actually holding such a discussion on the mailing list. Remember, I'm proposing a complete alternative to hCalendar and hCard, not a modification of it. A debate over competing proposals is outside the scope of any one proposal. >>I have no intention of developing a Calendar/Card system for HTML5 >>outside of a public forum, and the Wikis are outside of my control. > > Neither do I. I don't intend to develop a Calendar/Card system for HTML5 > at all, in fact, since someone else is already working on a system for > this. When it comes to editing that part of the WA1 spec, if there exist > proposals that can just be imported wholesale, I would much rather do that > than come up with yet another way of doing things. Why use it at all? If you want others to do the development of calender/card markup for you, why not just drop it from WA1 entirely? The only thing WA1 introduces on top of hCalendar/hCard are the <calendar> and <card> container elements. Otherwise, it's pretty much straight from Tantek's spec, and there are no obvious dependencies in other parts of the WA1 spec. (Not that I don't think including calendar/card markup in WA1 is a good idea...) Alternatively, why not just break WA1 into sections or modules (i.e. "Web Applications 1.0 - iCalendar Mappings"), sorta like CSS3, and develop each module individually? (I'll leave it up to you as to how to submit WA1 to standards organizations. You may want to submit all sections together to avoid breaking dependencies between them, or perhaps one at a time so that any one has a better chance of getting through.) >>>I'm not planning on touching the relevant sections of WA1 until >>>Tantek's drafts are much more stable. >> >>Why are we holding up the Calendar/Card portion of WA1 based entirely on >>the work of a single individual who is not a member of WHATWG? > > We're not. WA1 is likely to become a huge spec. I expect to be quite busy > doing lots of WA1 work for months. I hope that hCalendar and hCard will be > stable once most of the rest of the WA1 spec is ready; if it is, then we > will be able to re-use that work and save ourselves a lot of time and > effort. Reuse is fine so long as everyone agrees that a particular solution is the best or most practical. What concerns me is that I don't have a nice little wiki to post my ideas to or organize my thoughts with, whereas Tantek has his company's developer site, so I'm at a disadvantage. Since I and other WHATWG contributors don't have our own wiki, even if myself and other contributers find a better solution, you aren't going to put it into the WA1 spec until Tantek says he's done with his project. A good example are those two links you posted: http://developers.technorati.com/wiki/hCalendarFAQ http://developers.technorati.com/wiki/hCalendarIssues In both, Tantek is directly replying to a message I posted in this list, and he can add and change material over time to better support his position. Not that I think he shouldn't be able to do this. In fact, I've been asking for a WHATWG wiki for quite some time. Right now, sadly, the "XUL Alliance" does a better job of providing such resources. Anyone know a good free Wiki site, by the way? Oh, one other thing. If you're going to like to a spec, the page should be static, not a wiki. Otherwise, you could inadvertently allow anyone and their mother to change part of a WHATWG spec at will. >>Granted, he's vastly more qualified than I am, but with regards to the >>WHATWG charter, he has no more authority than I do. Are we going to >>start carving up WHATWG specs and handing out pieces to specific >>individuals outside of the WHATWG for them to control?!? If someone who >>worked on Browser-X comes up with a menu system for WA1, do we just put >>everything on hold until he finishes that draft?!? > > There's no "carving up" here; it's just a matter of not reinventing the > wheel. You'll notice that the <canvas> part of WA1 is almost identical to > the Apple implementation in early Tiger betas. This is not coincidental. > Similarly, we're reusing much of the work that the HTML and DOM working > groups have done over the past few years in terms of HTML and DOM errata. > And we're using suggestions from www-html, Slashdot, and bug systems; not > just in this mailing list. Much of the stuff you just mentioned was already implemented in a browser of some sort or approved by a standards organization. Tantek is not a standards organization, and as far as I know, hCalendar and hCard have not been implemented in a browser. In fact, after reading the spec more closely, it looks more like it's designed for automated extraction of information by webcrawlers rather than for actual calender controls and the like. > I'm not saying that your suggestions on <calendar> aren't welcome; quite > the opposite. However, at the moment, your suggestions would be best given > to Tantek, since Tantek is actively working on this kind of work, and has > studied the requirements in detail. I myself will not be working on the > calendar and business card parts of the spec for a long time. Actually, you gave a specific time you'd be working on it... "I'm not planning on touching the relevant sections of WA1 until Tantek's drafts are much more stable." So if the above is the case, I wouldn't be able to get anything regarding calendars or cards considered for inclusion in the specification until Tantek decides to complete his hCalendar and hCard specs, both of which date back to mid October of last year. > You are of course more than welcome to make independent proposals for > including calendar data and "business card" data in HTML(5). Why would anyone do that if the guy editing the target spec is going to ignore them for a period of time determined by a third party? It's one thing to be too busy to work on it at a given moment, but it's another thing to say that you definitely WON'T work on it until someone else is ready. > If your > proposals are technically superior to other proposals, then, once it comes > to including a calendar mechanism in the WA1 spec, the WHATWG members > (probably represented by me) will use your proposals. When will that be? When you're ready, or when Tantek is? (No offense to Tantek. This is an argument of procedure rather than character.) >>Might I also point out that the wikis you posted URLs to have NO >>COPYRIGHT NOTICES, NO OBVIOUS LINKS TO A COPYRIGHT POLICY and NO >>COPYRIGHT METADATA, and that the domain is that of THE COMPANY >>TECHNORATI. Who's to say his company won't just turn around and claim >>copyrights and patents to the whole mess? > > All good points to raise with Tantek. Not if you're linking to them through the WA1 spec. Once something becomes part of a WHATWG spec, these concerns become _your_ problem. (Although, if anything Tantek posts on the wiki becomes part of the WHATWG spec, there's the same issue, since he's already posted it... Has nothing to do with my method, though, since I use a different approach.) >>If you and Tantek want hCalendar and hCard as part of a WHATWG >>specification, you should post material here and allow people to comment >>on it publicly. A specific solution should not get special treatment >>regardless of who came up with it. I wouldn't mind seeing Tantek as a >>member of WHATWG personally, but freezing a portion of the spec so that >>a single individual can work on it completely outside of this working >>group is just wrong. > > I hope this e-mail has clarified any misunderstandings; specifically, that > comments are indeed welcome here, that Tantek's proposals aren't getting > special treatment, and that the relevant parts of the spec aren't frozen. If you're waiting on Tantek to finish his spec BEFORE working on the Calendar/Card section of WA1, that's special treatment. I doubt I could hold up portions of a WHATWG specification in a similar way. >>P.S.: I will be directing all my comments and ideas with regards to >>Calendar/Card markup to this mailing list only. I invite Tantek to do >>the same. > > Tantek's proposals are also, as I understand it, aimed at areas that are > out of scope for WHATWG, such as adding calendar metadata to existing > HTML4 documents that can be processed by Web crawlers such as > Technorati's. Thus I wouldn't be surprised if he did not feel it > appropriate to do the work in this forum. Actually, I was wondering if that was the case earlier, and looking at the spec now, I'm even more inclined to believe that... Hmm. I'm wondering if a calendar control might not be in order. Something like this: | <input type="calendar" name="calendar1" data="myCalendar.ics"> ..And for a version with fallback (just to annoy you ;) )... | <dataentry type="calendar" name="calendar1" data="myCalendar.ics"> | <div class="vcalendar"> | <div class="vevent" id="events" repeat="template"> | <p> | <label>URL: | <input type="text" class="url" | name="calendar1.event[events].url" | value="http://www.web2con.com/"> | </label><br> | <label>Start Date: | <input type="text" class="dtstart" | name="calendar1.event[events].dtstart" value="2005-10-05"> | </label><br> | <label>End Date: | <input type="text" class="dtend" | name="calendar1.event[events].dtend" value="2005-10-07"> | </label><br> | <label>Summary: | <input type="text" class="summary" | name="calendar1.event[events].summary" | value="Web 2.0 Conference"> | </label> | </p> | <p><button type="remove">Remove Event</button></p> | </div> | <p><button type="add" template="event">Add Event</button></p> | <p><button type="submit">Submit</button></p> | </div> | </dataentry> The above degrades into WF2, and then into HTML 4.01, while using hCalendar. Not sure we need something as specific as a calendar control, though. There may not be a use case for it. Then again, if webmail programs become more like Outlook and combine scheduling with email, there may be a significant use case for this.
Received on Wednesday, 23 February 2005 08:53:43 UTC