- From: Steve Lee <stevelee@w3.org>
- Date: Tue, 12 Mar 2019 15:21:05 +0000
- To: Janina Sajka <janina@rednote.net>, public-personalization-tf@w3.org
As a prototyping exercise I definitely agree with 1. After all it's one way of attaching a string (and so serialised data) to HTML. We could conceivable switch to others with minimal impact. I'm not so sure about 2 as it depends on how deep the reach is during protyping. Steve On 12/03/2019 12:20, Janina Sajka wrote: > Absolutely a consideration, Steve. But there's an important silver > lining in that cloud. By prototyping with data dash, we get two > important benefits: > > 1.) We get to demonstrate whether or not the concept works in real > world situations for real users. > > 2.) We get to document clashes. This is critical to pursuing any > custom minted attribute for personalization. If a custom attrib is > required, that decision can be backed by documentation and not just > reasonable conjecture. > > I actually think #2 is very important. But, so is #1. Meanwhile, we have > people on time limited proof of concept funding who can now move > forward. I just don't see how we can lose. > > Best, > > Janina > > Steve Lee writes: >> I still think the key issue is a potential global namespace clash. >> >> data-* attributes are a global namespace so if NOT treated as application >> scope specific (private) then clashes are likely to occur. These could break >> backward compatibility - a major "sin". >> >> Specifically, if we reserve any name then someone may have already used it - >> or want to. Any we will have broken their use of it. >> >> Here's a recent example of the serious problems clashes in a global >> namespace breaking backward compatibility [1][2]. ECMA wanted to add >> Array.prototype.contains() to the JavaScript standard but the very popular >> Moot Tools library had already used that so ESC had to define >> Array.protype.includes. >> >> Arguably Moo Tools should not have effectively reserved this name as they >> also could have broken someone else's code when Moo Tools was included. But >> the issues is the same in both cases - Global a namespace clash. >> >> We need to avoid this with data-* attributes. >> >> 1: http://2ality.com/2016/02/js-name-clashes.html >> 2: https://bugzilla.mozilla.org/show_bug.cgi?id=1075059 >> >> Steve >> >> On 11/03/2019 15:28, Charles LaPierre wrote: >>> That was the statements which was preventing us from using data- to >>> insert mathML into a page so that it could be used by a plugin etc. So >>> I was never quite sure why we couldn’t use data- for that purpose but >>> could for this purpose since both seem similar in my thinking. I don’t >>> remember who though putting the mathML into the data-mathml = “enter raw >>> mathML here” thought that was a violation of the use of data- But if >>> this isn’t the case we could have created a plugin which could interact >>> with the mathML. Or there could be data-latex as another option. Would >>> be really good to know why we can’t do that or if the spec on data- >>> needs to be updated. >>> >>> >>> Thanks >>> EOM >>> Charles LaPierre >>> Technical Lead, DIAGRAM and Born Accessible >>> Twitter: @CLaPierreA11Y >>> Skype: charles_lapierre >>> Phone: 650-600-3301 >>> >>>> On Mar 11, 2019, at 8:13 AM, lisa.seeman <lisa.seeman@zoho.com >>>> <mailto:lisa.seeman@zoho.com>> wrote: >>>> >>>> >>>> Folks, are we braking the html standard by using data- attributes? >>>> >>>> I quote >>>> >>>> " >>>> >>>> Custom data attributes >>>> <https://www.w3.org/TR/html50/dom.html#custom-data-attribute>are >>>> intended to store custom data private to the page or application, >>>> for which there are no more appropriate attributes or elements. >>>> >>>> These attributes are not intended for use by software that is >>>> independent of the site that uses the attributes. >>>> >>>> " >>>> >>>> >>>> >>>> >>>> From w3c schools (IE out in the wieled...) >>>> " >>>> *Note:*Custom attributes prefixed with "data-" will be completely >>>> ignored by the user agent." >>>> >>>> In pother words we are going against how dat- is intended to be used. >>>> >>>> >>>> All the best >>>> >>>> Lisa Seeman >>>> >>>> LinkedIn <http://il.linkedin.com/in/lisaseeman/>, Twitter >>>> <https://twitter.com/SeemanLisa> >>>> >>>> >>>> >>>> >>> >
Received on Tuesday, 12 March 2019 15:21:12 UTC