[rfc-i] URIs in RFC references, was: feedback on draft-iab-styleguide-01

"Martin J. Dürst" duerst at it.aoyama.ac.jp
Wed Mar 26 19:16:04 PDT 2014


On 2014/03/27 00:32, Ted Lemon wrote:
> On Mar 25, 2014, at 11:47 PM, John R Levine <johnl at taugh.com> wrote:
>> You appear to be assuming that everyone will want the document in the form that's on the landing page.  If that were so, why are we screwing around with multiple document formats?
>
> When you are browsing the web, you almost always want a web page.   When you are printing a document to read on the airplane, you probably want PDF.   When you are trying to render the document according to your specific preferences, you may prefer the XML.   So we have to have multiple formats, but it's not unreasonable to designate one as "most likely the one you want."   And cookies of course can optimize further, for people who really do prefer something else.

I think there are many cases where an alternative format makes sense. 
Some of them have come up in discussion already, and others we don't 
know yet, because technology keeps changing.

But if we do a good job with the HTML on the landing page, it should 
work for a wide range of uses and preferences of a wide range of people.

As an example, while Ted may still want to use PDF for printouts, I'd 
consider it a failure if the HTML on the landing page were not good 
enough for me for print outs (I currently use the tools version and 
expect the new HTML version to look better for printing).

Similarly, while John may still want to use a separate ebook version for 
reading on a dedicated reading device, I'd hope that if not initially 
then eventually, we get the HTML to a point where it's acceptable 
directly to a large percentage of users for reading on an ebook device 
(or that content negotiation can take care of the issue).

Why? The average user doesn't like to click twice, and likewise doesn't 
(want to) think about different versions.

Regards,   Martin.


More information about the rfc-interest mailing list