[OpenAjaxMobile] Analysis of W3C's latest geolocation APIs and proposals for how to generalize

[Rotan Hanrahan]

Thanks Jon. We're thinking along the same lines.

As you probably know, one of the key ingredients of good specification management is correct and consistent labeling, which is why the W3C has a well-defined process and a strict publications mechanism. Still, your prediction is probably not far off, for all the reasons you describe. My concern is only to ensure that the first visible contact between OAA and the proposed geo-location group gets off to a good start.

Regarding the possible roles of the three groups you have mentioned, the outline seems reasonable. It wouldn't be true, however, to say that W3C stays out of security specs, given that the consortium has been responsible for SOAP security extensions, the Web Security Context, Cross-site Access Controls, various security requirements for XML encryption, XML key management, Web Services and so on... though I concede that security wouldn't be their speciality. Indeed, it is probably fair to say that W3C groups that are not dedicted to security matters would welcome such work being shouldered by external experts.

Finally, I want to pick up on one particular word from your original message: simple. I completely agree that the primary geo-location API should be simple. There is a danger, however, that the appearance of a W3C group dealing with geo-location API issues might be seen as "the only game in town" and will come under pressure from many directions to have everyone's favourite features included in the specification. That would be dangerous; it would be a recipe for continuous argument and perpetual procrastination. Perhaps one of the best things we could do for the nebulous working group is to give it some opinion on what we would consider to be the "bare minimum" for a working geo-location API. If they can create such an API in a reasonable time, that would be a success. We need the geo-location "axioms". The orthogonal atomic features. No more, no less.

I'm no GL expert, so perhaps others could jump in here to separate the must-haves from the nice-to-haves.

---Rotan





From: Jon Ferraiolo
Sent: Wed 03/09/2008 18:19
To: OpenAjax Alliance discussion list on Mobile Ajax
Subject: Re: [OpenAjaxMobile] Analysis of W3C's latest geolocation APIs and proposals for how to generalize


Hi Rotan,
Yes, of course, the Geolocation WG does not yet have an approved charter, and even if the charter is approved as is, there are lots of potential changes that might happen along the way. Therefore, I agree that from a formal process perspective, the current geolocation spec technically is just a proposal. However, given the people who are involved in developing the spec (e.g., Google, the mapping leader, and Microsoft, also big in mapping), the degree of public discussion so far, emerging partnership between W3C and the mobile operator community (in the form of OMTP), and the prevailing political winds (favorable winds towards HTML5-related efforts), my prediction is that this particular spec has a really good chance of remaining largely intact. Of course, others might read the tea leaves differently. Also, as one of the people at OpenAjax Alliance, I like the APIs that they have developed, so at least one person at OpenAjax will be supportive.

I definitely encourage appropriate liaison activities between W3C and OpenAjax Alliance (and OMTP), whether formal or informal. Right now there are 3 industry groups that are active in the device APIs area, W3C, OMTP and OpenAjax Alliance. We should do our best to all march in the same direction, and as far as I can tell, we all have sincere intents towards working together in an effective manner. Here is my current thinking about how to divide the work up:

* W3C develops the API specs that defines some (or maybe all) of the actual JavaScript device APIs
* OMTP develops security specs that the W3C can reference. W3C stays out of security specs. (Traditionally, a difficult area for the W3C)
* OMTP defines JavaScript APIs for technologies that W3C hasn't launched yet but that the industry needs. They specs might be transitioned to W3C at a future time.
* OpenAjax reviews all of the above work and provides feedback, particularly on how well the above work aligns with the needs of Ajax developers
* OpenAjax might define some JavaScript APIs for technologies that the industry needs but OMTP and W3C haven't started yet
* OpenAjax might launch on open source JavaScript shim layer activity that allows JavaScript access to legacy device APIs, but such an effort probably requires the available of mobile browser plugins such as Aplix's WebVM for such an effort to be viable.

Jon


[...]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://openajax.org/pipermail/mobile/attachments/20080903/93a3f456/attachment.html