[rfc-i] Graceful degradation is key, was: Re: draft-hildebrand-html-rfc

Martin Rex mrex at sap.com
Mon Jul 16 13:42:14 PDT 2012


Joe Hildebrand (jhildebr) wrote:
> "Martin Rex" <mrex at sap.com> wrote:
> > 
> >And whenever I run out of screen real estate, I can use a printed copies
> >of RFCs in addition (yes, I have and occasionally use printed copies of
> >some RFCs, especially when implementing one).
> 
> I don't understand your point here.  Are you suggesting that you can't
> print HTML?

I'm not aware of any existing solution to print HTML in any useful way.
I can print the existing ascii TXT rfcs in 2-up fashion easily, and
the page numbering still works and is in perfect sync with the
TXT documents I look at on-screen and in perfect sync with the rfcmarkup
available at http://tools.ietf.org/html/rfcXXXX

If page numbering didn't work 1:1 between printouts and on-screen,
then it would be pretty useless (as substitute).

> 
> >Browser windows on a PC are usually *MUCH* larger, because most stuff
> >in the internet, designed for browser usage, is quite unusable with
> >a browser window that is 507x603 pixels on the outside
> >--you see parts of the branding header, partial ads and parts of
> >the navigation pane, but little contents.
> 
> I also don't understand your point here.  We're not going to be injecting
> ads into RFCs.  You can run a browser in your xterm window, or externally.
>  If you run a dedicated graphical browser, you can turn off as much of the
> chrome as you like.  We need to ensure that we're using screen real-estate
> as best we can in the pages we generate, but that's more of a design issue
> than a format issue.

Browser windows have much larger decorations than an "xterm", the tiny
number of font( size)s accessible through "zoom--" are not really useful
if your objective is to fit a number of windows on one screen,
and creating small browser windows often interferes with regular browsing
because of the window size stickyness of browser windows.

No really, Browsers are made for serialized consumption of content in
a single window.  Many system start browser in full-screen-mode by default,
may users use browsers in full-screen mode by default, and many web pages
don't fit all important information on 1024x768 displays even when using
a full-screen browser window.


I use my web browser primarily for displaying the stuff that can not
reasonably be rendered without a browser, and I don't want to constantly
run around chasing down tools that can convert HTML gimmickry into
readable text.

Today, RFCs can be used even on a pure text console.  I don't see any
benefit in breaking this functionality just for the fun of it.

Similar for the authoring.  Todays I-Ds are just fine with a format
like nroff.  nroff supports output with other page sizes than 72col/56ln
and "groff -Thtml" could be used to create HTML output -- although
it would need additional nroff macros to let HTML rather than author
or nroff do the formatting of enumerations/lists.  (I hardly see
use cases for tables where preformatting would _not_ work).

-Martin


More information about the rfc-interest mailing list