Carsten, > On Nov 28, 2019, at 8:59 AM, Carsten Bormann <cabo@xxxxxxx> wrote: > > On Nov 28, 2019, at 17:33, John C Klensin <john-ietf@xxxxxxx> wrote: >> >> So, even if it is too late to reconsider these decisions, I hope >> we can at least be aware of possible negative side effects. > > It is never too late to fix this bug. > > I think the last few months have uncovered a lot of usages for the plain text form of RFCs, as well as the fact that “nobody will ever look at the plain text now that we have HTML” is maybe true for 80 %, but not for 100 %. I think that is about right, I use both html and txt. Html for reading on the screen, txt for doing comments and markups, excerpts in email, etc. I have one active document that I converted to V3. I note that the txt output of XML2RFC produces headers, footers, and page breaks. The other thing I have noticed, since I now have the latest draft produced by V3 and the previous drafts produced by V2, is that several things changed that make this transition more difficult than needed. The first I have found was when doing the conversion from V2 to V3, the default was to expand references, instead of the one line version. This can be worked around when using the --add-xinclude when doing the conversion. The other was default symbols in lists. These can be worked around by using the --legacy-list-symbols when running XML2RFC. My point is that in both cases I think the defaults should have been to make it closer to V2. The current defaults make it harder to deal with converting an existing in progress draft from V2 to V3. I suspect these are issues that many will run into when starting to use V3. Thanks, Bob
Attachment:
signature.asc
Description: Message signed with OpenPGP
