Should we split off the JSON encodings of geographic locations and contact information into separate documents, so they can be re-used by others?
Should we urge ECRIT (or some other working group) to re-specify LoST in JSON encoding? Or should we do the equivalent work here in PAWS? -Pete [email protected] wrote: > I can sense an agreement that we can go ahead and use json encoding > (only). > > I would then go ahead and instruct the editor to encode the data model > with json in the merged draft. > > - Gabor > > -----Original Message----- > From: ext Zhulei [mailto:[email protected]] > Sent: Thursday, September 13, 2012 9:17 PM > To: Bajko Gabor (Nokia-CIC/SiliconValley); [email protected] > Subject: Re: JSON vs XML > > Hi, > > I really went through e-mail discussions on this. My proposal is to > use json as data model of paws protocol issued by paws, in case that > wg addresses the trend of APIs of browser venders and keep some > discovery mechanisms open for any extra works. At the same time, we > have some problems to satisfy the needs to reuse ws schema encoded by > traditional devices, which will issue a separate wg document for xml > encoding standard for xml supporting industries. > > The reasons doing this are not technical issues, just we have broad > requirements to reuse TSWS generally for communication purposes, > including smart objects, ad hoc use, browser API, wifi broadband, > cellular etc. > > Best regards, > Zhu Lei > > -----邮件原件----- > 发件人: [email protected] [mailto:[email protected]] > 发送时间: 2012年9月14日 7:40 > 收件人: Zhulei; [email protected] > 主题: RE: JSON vs XML > > There was not much feedback on the list about the objections to the > different encodings. > > We seem to agree that: > xml may not be the right choice because the current trend for APIs in > the browsers is towards json; and if we choose json, as some data > structures PAWS may reuse are encoded in xml, a json encoding for > those would need to be defined (which is not impossible, but requires > some extra work) > > The latter of the two objections will not be valid when we'll send the > document to iesg, as all the encodings will have to be in place at > that time. > So we are left with practically one question: do we want to follow the > current industry trend and use json, or do we want to stick with xml. > > A significant number of people prefer to specify both encodings, but > that may not be agreeable with the iesg. > > This is where we stand now, deadlocked on this not critical issue. > > Any suggestion on how to move forward would be appreciated. > > - Gabor > > > From: ext Zhulei [mailto:[email protected]] > Sent: Monday, September 10, 2012 12:59 AM > To: Bajko Gabor (Nokia-CIC/SiliconValley); [email protected] > Cc: Zhulei > Subject: Re: JSON vs XML > > Hi, > > Actually, no so much comments on this choosing. I just do not think > xml is a problem to embedded devices, in fact xml is well supported by > different sort of devices in my view. The issue may be some power and > bandwidth constrained devices (e.g. some smart objects) to support txt > based information. Let’s ignore this case since we are not to define > binary encoding at the moment. > > Best regards, > Zhu Lei > > 发件人: [email protected] [mailto:[email protected]] 代表 > [email protected] 发送时间: 2012年9月8日 4:41 收件人: [email protected] 主题: [paws] > JSON vs XML > > The chairs discussed with the AD, and we came up with the following > action plan to drive this wg to a consensus on the json vs xml encoding: > we’ll collect an objections list for json and one for xml, listing what > is seen wrong/problematic with that encoding. The chairs and the wg will > go through that list and see if the objections are valid, then decide > which encoding has more support and choose that one. If we end up with > good objections list for both, we may choose to support both encodings, > as that list will justify the decision once the document advances to the > iesg. > > I went through the emails and I found so far the following valid > objections: > > xml: > too verbose, may be a problem to be supported in embedded devices > current trend for APIs in the browsers is towards json > > > > json: > some data structures are encoded in xml, a json encoding for those > would need to be defined (which is not impossible, but requires some > extra work) > > > If you have additional objections, send them to the list asap. > > - Gabor > _______________________________________________ > paws mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/paws _______________________________________________ paws mailing list [email protected] https://www.ietf.org/mailman/listinfo/paws
