Re: [Tools-discuss] .txt? [I-D Action: draft-xxx.txt]

[Eric Rescorla <[e-mail]>]

On Tue, Jun 29, 2021 at 6:55 AM Warren Kumari <[e-mail]> wrote:

> On Mon, Jun 28, 2021 at 5:48 PM Brian E Carpenter
> <[e-mail]> wrote:
>
> > 3. SVG. Hard work to use, but well worth it.
>
> Again, see #1. I believe that requiring that complex things be
> explainable using ASCII art was a feature - if your protocol is
> sufficiently complex that it cannot be easily explained using words
> and ascii art, it's probably too complex.
>

I'm sure I'm going to regret this, but....

Obviously things *can* be explained using words and ASCII art. The question
is whether it would be easier to explain with diagrams that were better
looking. I -- as I imagine you have -- have read a number of books
documenting network protocols and nearly universally they use high quality
diagrams, not ASCII art. I would suggest that this is a strong indication
that our format reduces clarity.

There are a number of cases that are particularly burdensome
- State machine diagrams
- Math of any kind
- Ladder/swimlane diagrams in which there is nontrivial amounts of stuff
happening on each step

People who make the kind of argument you are making -- a related one being
that engineers should have to use slow computers or tiny displays in order
to keep complexity down -- usually seem to have in mind that this will
result in the output being better along some axis. Is that in fact your
thesis? If so, so you have any evidence that forcing people to use ASCII
art reduces output complexity?

-Ekr


> 4. Using tools that are currently maintained.
>
> Tools like the XML2RFC XMLMind plugin were working just fine with the
> V2 format; every few years I'd release a new version when XMLMind
> updated their major version, but it was literally just a version bump.
> Same thing for my "copy the last RFC, change the title and words". #4
> sounds very much like "people weren't keeping the tools maintained the
> way we liked, so we made a breaking change to force them to abandon
> what they were doing, and use what we think is better.
>
> Anyway, I realize that this ship has sailed, and I'm doing a good
> impression of "Old Man Yells at Cloud"... but someone asked why more
> stuff wasn't being submitted in v3 format, and this is at least one
> set of reasons why...
>
> W
>
> >
> > I've had no trouble at all editing XML in v3. The only real
> > annoyance is that the converter can't do anything sensible with
> > <vspace blankLines="1"/> which I used to use a lot. Everything
> > else is trivial.
> >
> > (The converter insists on inserting lots of format="default"
> > and toc="default" which are just noise and can be deleted.)
> >
> > Regards
> >    Brian
> >
> > On 29-Jun-21 07:38, Toerless Eckert wrote:
> > > I only converted the last rfc i was editor for during the final
> revisions to go to rfc,
> > > and i only did this so i could use the only one new feature of v3 that
> i felt i wanted
> > > to use, name the Contributors tag. I still don't know what else i as
> an author
> > > would benefit from in v3.
> > >
> > > I did find the conversion sufficient for that last mile to RFC editor,
> but not persuasive to
> > > suggest it to authors if/when they want to continue doing mayor edits
> to the document. This
> > > is primarily beceause the v3 ended up with a tag-verbose XMLv3 than
> the v2 i had
> > > edited for years. This specifically included inlining the rfc/draft
> references as opposed
> > > to keeping the references, but also several other tags that where
> written out more verbosely
> > > and with a lot of default parameters (unnecessarily).  Hope i remember
> this all correctly.
> > >
> > > This v2->v3 conversion process feels a bit like attempting to have a
> good idea but then
> > > outsource the conversion cost.  Reminds me of linux. Great new
> SDK/Library, but now all
> > > the third-party apps developed against an older version have to be
> rewritten. In comparison,
> > > in Windows i can have 10 versions of the same core SDK co-installed
> and all the old but
> > > still useful programs will still run. But no linux distributions do
> not support such slotting
> > > or do not compile all old library versions, and library developers
> don't care about supporting
> > > multi-slotting... *sigh*
> > >
> > > Chers
> > >     Toerless
> > >
> > > On Sat, Jun 26, 2021 at 10:40:40PM -0400, John R Levine wrote:
> > >>>> Among the many things on the to-do list is to redo the I-D
> submission page
> > >>>> to make it clearer that you only need to submit one version of a
> draft,
> > >>>> and that we'd appreciate the XML version if you have one.
> > >>>
> > >>> Excellent. Is there any reason not to run the v2 to v3 converter
> automatically?
> > >>
> > >> We really want people to stop using v2.  It's obsolete and missing
> some
> > >> semantic features of v3.
> > >>
> > >> Regards,
> > >> John Levine, [e-mail], Taughannock Networks, Trumansburg NY
> > >> Please consider the environment before reading this e-mail.
> https://jl.ly
> > >>
> > >> ___________________________________________________________
> > >> Tools-discuss mailing list - [e-mail]
> > >> This list is for discussion, not for action requests or bug reports.
> > >> * Report datatracker and mailarchive bugs to:
> [e-mail]
> > >> * Report tools.ietf.org bugs to: [e-mail]
> > >> * Report all other bugs or issues to: [e-mail]
> > >> List info (including how to Unsubscribe):
> https://www.ietf.org/mailman/listinfo/tools-discuss
> > >
> >
> > ___________________________________________________________
> > Tools-discuss mailing list - [e-mail]
> > This list is for discussion, not for action requests or bug reports.
> > * Report datatracker and mailarchive bugs to:
> [e-mail]
> > * Report tools.ietf.org bugs to: [e-mail]
> > * Report all other bugs or issues to: [e-mail]
> > List info (including how to Unsubscribe):
> https://www.ietf.org/mailman/listinfo/tools-discuss
>
>
>
> --
> The computing scientist’s main challenge is not to get confused by the
> complexities of his own making.
>   -- E. W. Dijkstra
>
> ___________________________________________________________
> Tools-discuss mailing list - [e-mail]
> This list is for discussion, not for action requests or bug reports.
> * Report datatracker and mailarchive bugs to: [e-mail]
> * Report tools.ietf.org bugs to: [e-mail]
> * Report all other bugs or issues to: [e-mail]
> List info (including how to Unsubscribe):
> https://www.ietf.org/mailman/listinfo/tools-discuss
>