ISSUE-223: Various Items to Consider for the CT Guidelines

Jo's CT Shopping List

Various Items to Consider for the CT Guidelines

Guidelines for Web Content Transformation Proxies
Raised by:
Jo Rabin
Opened on:
In the process of editing the CT Problem Statement I have been musing on various things - some, but not all, of them are already suggested by the Problem Statement. If there is any desire to discuss these, it may be better to create individual ISSUEs for those that people choose to discuss.

1. Under what circumstances do user's intentions and choices override author's intentions and choices, and vice versa.

2. How should users signal their choices - if they user has a choice of browser settings how should the browser communicate those settings, expecially the "render as mobile" vs "render as desktop options".

3. What assumptions are legitimate for the owner of a transcoding proxy to make about the default experience.

4. To what extent should the Guidelines talk about HOW transformation is carried out? Do we think that this altogether an area for proprietary differentiation or do we comment on good and bad techniques? Should we, for example, comment on preserving document order, taking into account absolute positioning? Do we think that servers should add clues as to how to paginate and so on? Is this stepping into the land of UWA and DIAL? What repertoire of processing instructions do we anticipate that proxies are able to act on - do we think that they will understand and interpret XSLTransform intended for client side transformation, for example? A better example would be scripting. What about flash?

Is there a vocabulary of functions that a transforming proxy might offer and that a user or origin server might select: e.g. pagination, linearization. image resizing ...

5. To what extent is it permissible for a transforming proxy to step in and correct invalid syntax, even if the origin server says not to transform its content? Is there a difference between "no-transform" - i.e. don't mess with my content even if it is wrong and "no-trasnform-but-tidy-allowed".

6. Testing

As things stand, you can only test the operation of your server in tandem with some transforming proxy by having someone access your Web site from the coverage area of a network that implements that version of the proxy. We should make strong noises about the desirability of making testing practical for content developers. Given that operators/proxy providers were willing to do this, what infrastructure would be needed to facilitate it? i.e. how would you find out about the test facilities and so on?

7. Is there a question here about identifying which DDR the components get their information from, as if you think a proxy is using a deluxe chrome plated one, you might be more inclined to trust it than if you think it is using a nasty plywood one.

8. Do we expect transforming proxies to be aware of DPE servers and the like and should we say so?

9. What part does the mysterious HTTP 300 status play in all this? Where does it fit in with helping origin servers and proxies provide information about alternative representations (with link rel = or something else, to enumerate options). How does this fit in with allowing the user to select their choice as well as helping provide a reasonable default exprience.

10. Do we think that transforming proxies should be able to comply with mobileOK Basic 1.0?

11. To what extent do proxies need to be powder aware and powders need to be proxy aware? Is powder a sensible mechanism for identifying alternative representations, and the author's intent in creating them?

12. If something is mobileOK, does that mean it is not open to transformation?
Related Actions Items:
Related emails:
  1. Content Transformation Guidelines 1m (Rev 13) nearly now all 'ship shape and Bristol fashion' (from on 2008-07-22)
  2. Content Transformation Guidelines 1m (Rev 13) [was Re: Content Transformation Guidelines 1l (Rev 12) and Change List] (from on 2008-07-22)
  3. RE: Content Transformation Guidelines 1l (Rev 12) and Change List (from on 2008-07-18)
  4. [minutes] CT Call Tuesday 15 July 2008 (from on 2008-07-15)
  5. RE: [agenda] CT Call Tuesday 15 July 2008 (from on 2008-07-15)
  6. [agenda] CT Call Tuesday 15 July 2008 (from on 2008-07-15)
  7. Content Transformation Guidelines 1l (Rev 12) and Change List (from on 2008-07-11)
  8. [minutes] BPWG F2F in Sophia, day 1 (from on 2008-06-17)
  9. Content Transformation Guidelines - preparation for the F2F (from on 2008-06-13)
  10. Re: [agenda] CT Call Tuesday 10 June 2008 (from on 2008-06-10)
  11. [agenda] CT Teleconference Tuesday 11 March 2008 (from on 2008-03-10)
  12. Seoul BPWG F2F - report (from on 2008-03-09)
  13. [minutes] Seoul F2F Day 2 - Tuesday 4 March 2008 (from on 2008-03-04)
  14. [minutes] Minutes of the Content Transformation Task Force, 2007-11-20 (from on 2007-11-20)
  15. ACTION-581 Promote discussion on list, and produce editor's draft of Guidelines (from on 2007-11-19)
  16. FW: ISSUE-223 (Jo's CT Shopping List): Various Items to Consider for the CT Guidelines [Content Transformation Guidelines] (from on 2007-10-03)
  17. ISSUE-223 (Jo's CT Shopping List): Various Items to Consider for the CT Guidelines [Content Transformation Guidelines] (from on 2007-10-03)

Related notes:

We've covered items 1-6 on this list at the f2f March 3-4 in Seoul.

Daniel Appelquist, 4 Mar 2008, 09:08:04

For the F2F in Sophia:
- points 1 to 6 (included) were addressed during F2F in Seoul
- point 7: Scope for Future Work? Don't see how this could be done without new mechanisms.
- point 8: Scope for Future Work? Add links to DIAL, OMA-DPE, ...?
- point 9: Scope for Future Work? Automation of using the HTTP Status 300 is not well-implemented (RFC2295) and cannot be used at that time.
- point 10: Good point. Do we want to link to mobileOK?
- point 11: Scope for Future Work? No dependency on POWDER for the time being.
- point 12: don't think so. Resolution?

François Daoust, 13 Jun 2008, 08:54:22

7 to 9, 11: scope for future work
RESOLUTION: On point 10 of ISSUE-223 (whether CT proxies should be MobileOK) we should remain silent.

RESOLUTION: On point 12 of ISSUE-223 (whether CT proxies should refrain from transforming MobileOK content) we should allow for the possibility that CT proxies should be able to transform MobileOK content but that they should take into account MobileOK-ness as part of the heuristics involved in determining whether content is mobile-friendly (but remain silent on how you check if something is mobileOK)

Dominique Hazaël-Massieux, 16 Jun 2008, 11:38:04

Display change log ATOM feed

Jo Rabin <>, Daniel Appelquist <>, Chairs, Dominique Hazaël-Massieux <>, François Daoust <>, Staff Contacts
Tracker: documentation, (configuration for this group), originally developed by Dean Jackson, is developed and maintained by the Systems Team <>.
$Id: 223.html,v 1.1 2011/01/10 15:19:44 dom Exp $