- From: Noah Mendelsohn <nrm@arcanedomain.com>
- Date: Wed, 18 Jan 2012 14:46:12 -0500
- To: "Roy T. Fielding" <fielding@gbiv.com>
- CC: Robin Berjon <robin@berjon.com>, Larry Masinter <masinter@adobe.com>, "C. M. Sperberg-McQueen" <cmsmcq@blackmesatech.com>, "www-tag@w3.org" <www-tag@w3.org>
On 1/18/2012 1:30 PM, Roy T. Fielding wrote: > On Jan 18, 2012, at 2:36 AM, Robin Berjon wrote: > >>> On Jan 18, 2012, at 03:10 , Larry Masinter wrote: >>>>> What is too complex and ultimately wrong is to give options to >>>>> those writing specs that make references. >>> >>> I beg to differ. Ultimately, those who write specs are those most >>> informed about how they should handle their own references (if >>> they're not, then we have an entirely different problem). They are >>> the ones who should be making decisions about whether the >>> specifications they reference can be trusted to remain compatible >>> and should therefore be loosely bound, or on the contrary (hopefully >>> only in extreme cases) have a broken upgrade path and need an >>> anchored reference. > That's impossible. The appropriate dependencies for a standard depend > on the social context of its implementation at some future time. Spec > authors don't have a clue what that will be, nor whether it will be > uniform across all implementation contexts. That's why the Web's > orthogonal technologies are supposed to be defined in orthogonal specs, > and pretty much by definition orthogonal specs have loose bindings > (unless they are dead). Roy, please clarify. My impression is that Robin is endorsing latitude for authors, in part so they can go for loose coupling; my impression is that Larry is against loose coupling. Your "that's impossible" statement is physically positioned as being a response to Robin, but the content seems more at odds with what I take to be Larry's position. In any case, it seems pretty clear that you're strongly in favor of allowing loose coupling, at least as a (very common) option, and perhaps as the only option. FWIW: I think loose coupling should be an option, and for the reasons you state I think loose coupling will in many cases be the right choice, risks and all. I also think there are situations in which loose coupling is a bad choice, typically if one doesn't have sufficient bounds on the possible changes to have high confidence that security or interoperability won't be inappropriately compromised. Noah
Received on Wednesday, 18 January 2012 19:46:49 UTC