W3C

Results of Questionnaire [Call for Objections] Network interaction definition(s)

The results of this questionnaire are available to anybody. In addition, answers are sent to the following email address: team-tracking-chairs@w3.org

This questionnaire was open from 2013-11-30 to 2013-12-18.

8 answers have been received.

Jump to results for question:

  1. Objections to Option A: all requests caused by a user interaction
  2. Objections to Option B: user action, network interaction, and subrequest

1. Objections to Option A: all requests caused by a user interaction

Option A: all requests caused by a user interaction

A network interaction is the set of HTTP requests and responses, or other sequence of logically related network traffic, caused by a user visit to a single web page or interaction with page elements.

Non-normative: This is meant to represent the atomic level of interaction by users with web pages. Page loads, navigation, and clicks are common forms of interaction.

Page load, object mouse-over, image click, hyperlink click, sub-element re-load, and form entries are all examples of valid network interactions.


If you have an objection to this option, please describe your objection, with clear and specific reasoning.

Details

Responder Objections to Option A: all requests caused by a user interaction
Bryan Sullivan Conditionally agreed (not an objection) if it is clarified that "all requests caused by a user interaction" includes anything that the app does automatically (e.g. via XHR, JSONP, etc) in response to some earlier interaction. This includes simply launching the start page of the site/app, as in many current apps all kinds of ongoing automatic transactions can occur simply due to visiting the start "page".
Mike O'Neill The definition is very vague and of no practical use.
David Singer This has serious problems. (a) DNT is attached to a single HTTP header. Is the server supposed to somehow work out which other HTTP requests were for a 'single web page'? (b) Servers know nothing about web pages, only web resources; most servers will not know the end of the 'page load'. (c) many web sites are constructed of independent frames, dynamically generated content and the like, which also makes the concept of 'page' fairly meaningless. As stated, this term in unusable conceptually, unimplementable technically, and unverifiable formally.
Walter van Holst While making sense from a legal perspective, this definition makes little sense from a technical perspective. Under this definition a network interaction can involve a one-to-many relation since loading for example a banner ad is logically related to a page load, but may involved entirely different parties than the publisher of the page.
Rob van Eijk Objection: A network interaction is a single HTTP request, not a dialogue. Network interaction and visiting a single site are not
synonymous. At least not from a TPE perspective. Since the current state of play is centered around the TPE, it makes sens to define the network interaction in a bottom up (i.e., technical) sense. If it were a definition for the TCS, a more top down (i.e., legal) definition would be needed.
Jack Hobaugh In general, I object to the porting of a TCS compliance definition into the TPE. I feel strongly that the TPE should remain a pure protocol and technical specification document. Some have contended that some TCS definitions are needed in the TPE in order for the consumer to understand the choice that the consumer is making regarding the DNT signal. This is simply not the case. A technical specification need only specify the requests and responses necessary for a DNT protocol to be implemented in a scalable and implementable solution across all browsers and the servers called. The intended audience for the TPE is made up of technical implementers such as software engineers, developers, and programmers. To be clear, the TPE should not take on the responsibility of informing consumers and attorneys regarding a policy or compliance choice but instead should inform the technical community on how to implement the technical solution. The compliance specification for the DNT signal should be left to the compliance regime, whether it is a national compliance regime, a W3C-based compliance regime or an industry-based compliance regime. Porting definitions from a particular compliance regime into the TPE only serves to provide an incomplete and confusing picture to the software engineers, developers, and programmers who will be tasked with implementing the technical protocol.

Although I object to any compliance definitions being ported from the TCS into the TPE, I would prefer the Option A definition for network interaction over Option B.
Shane Wiley While the current TPE drafts don’t carry the term “Network Interaction” it was my understanding we want this definition to suggest its only in these cases that DNT be communicated. In that context, I believe this is the more succinct and complete definition of the two. It provides a logical approach to request and response pairing through user interaction and subsequent activities. Easy to understand and very straight-forward.
Brad Kulick While I don't see network interaction referenced anywhere within the TPE, I understand that we need to scope which network traffic DNT signals are relevant for.

I support this option. It succinctly captures all of the network traffic (requests and responses, both direct and non-direct user initiated) which DNT signals would be applicable for.

This definition of network interaction is complete, simple, and straightforward manner.

2. Objections to Option B: user action, network interaction, and subrequest

Option B: user action, network interaction, and subrequest

A network interaction is a single HTTP request and its corresponding response(s): zero or more interim (1xx) responses and a single final (2xx-5xx) response.

A user action is a deliberate action by the user, via configuration, invocation, or selection, to initiate a network interaction. Selection of a link, submission of a form, and reloading a page are examples of user actions.

A subrequest is any network interaction that is not directly initiated by user action. For example, an initial response in a hypermedia format that contains embedded references to stylesheets, images, frame sources, and onload actions will cause a browser, depending on its capabilities and configuration, to perform a corresponding set of automated subrequests to fetch those references using additional network interactions.

If you have an objection to this option, please describe your objection, with clear and specific reasoning.

Details

Responder Objections to Option B: user action, network interaction, and subrequest
Bryan Sullivan Given agreement with my comment on Option A, these two definitions are the same. They have to be, because without the additional clarification to A, the "subrequests" in B are omitted as network transactions - an error since all transactions need to be accounted for in the definition.
Mike O'Neill I agree with the definition of a network interaction.

I do not like the definition of user action because the examples it gives may not be the result of an action by the user, they could have been fooled into clicking a link that takes them to a different server than they imagine or it may be the result of script placed by a server i.e. that uses a JavaScript element.click() or form.submit()method.

Similarly the differentiation of a request from a subrequest by whether it has been directly initiated is vague and not useful.
David Singer
Walter van Holst
Rob van Eijk Network interaction: no objection

User action: no objection. maybe including the previously coined term 'meaningful interaction'.

Subrequest: Objection. I object to defining subrequests. A network interaction must be limited to a single request, and not cascade to multiple get requests. It seems almost synonymous with OPTION A (all requests caused by a user interaction).
Jack Hobaugh In general, I object to the porting of a TCS compliance definition into the TPE. I feel strongly that the TPE should remain a pure protocol and technical specification document. Some have contended that some TCS definitions are needed in the TPE in order for the consumer to understand the choice that the consumer is making regarding the DNT signal. This is simply not the case. A technical specification need only specify the requests and responses necessary for a DNT protocol to be implemented in a scalable and implementable solution across all browsers and the servers called. The intended audience for the TPE is made up of technical implementers such as software engineers, developers, and programmers. To be clear, the TPE should not take on the responsibility of informing consumers and attorneys regarding a policy or compliance choice but instead should inform the technical community on how to implement the technical solution. The compliance specification for the DNT signal should be left to the compliance regime, whether it is a national compliance regime, a W3C-based compliance regime or an industry-based compliance regime. Porting definitions from a particular compliance regime into the TPE only serves to provide an incomplete and confusing picture to the software engineers, developers, and programmers who will be tasked with implementing the technical protocol.

Specifically, I object to this particular set of definitions because it does not add any clarity to the concise definition already offered in Option A and adds two superfluous definitions (“user action” & “subrequest”) that are neither referenced in the offered definition for “network interaction” nor are they referenced in the TPE.
Shane Wiley The current TPE drafts don’t mention any of these terms so I’m challenge with understanding the full intent of the assembly of these terms. While it’s possible the terms could be assembled in a meaningful manner within the definition of Network Interaction, they have not been in this case. There could several combinations or modifiers used in their assembly that I would likely disagree with but initiatively I believe they (user action and subrequest) were set out as limitations within the broader conceptualization of a Network Interaction – which would actually make the two definitions much closer than they currently appear. That said, since these terms were never fully articulated in this manner, I need to object to this definition of Network Interaction as being overly broad.
Brad Kulick While I don't see network interaction referenced anywhere within the TPE, I understand that we need to scope which network traffic DNT signals are relevant for.

I strongly object to this option. While it provides definitions for network interaction, user actions, and subrequests and they could be used in conjunction to provide the appropriate scoping for which network traffic DNT is applicable for, it leaves these elements disjoint. Without tying them together or understanding how they will be used within the TPE, I can only assume the manner in which they we meant to be applied, which simply isn't good enough to proceed forward.

This option provide individual definitions that are useless in there disjoint presentation.

More details on responses

  • Bryan Sullivan: last responded on 3, December 2013 at 02:26 (UTC)
  • Mike O'Neill: last responded on 13, December 2013 at 00:06 (UTC)
  • David Singer: last responded on 17, December 2013 at 22:38 (UTC)
  • Walter van Holst: last responded on 18, December 2013 at 09:17 (UTC)
  • Rob van Eijk: last responded on 18, December 2013 at 18:56 (UTC)
  • Jack Hobaugh: last responded on 18, December 2013 at 22:17 (UTC)
  • Shane Wiley: last responded on 19, December 2013 at 04:00 (UTC)
  • Brad Kulick: last responded on 19, December 2013 at 04:43 (UTC)

Everybody has responded to this questionnaire.


Compact view of the results / list of email addresses of the responders

WBS home / Questionnaires / WG questionnaires / Answer this questionnaire