Re: Things that used to be clear (was Re: Evolving Documents (nee "Living Documents") side meeting at IETF105.)

[Michael StJohns <[e-mail]>]

On 7/10/2019 1:18 PM, Eric Gray wrote:
>
> Richard,
>
> A few things that people who suggest a complete replacement of tools 
> should think about:
>
>  1. frequently introducing yet another learning curve doesn’t shorten
>     the overall learning curve for anyone.  If there is any benefit in
>     it for anyone, it is that it tends to level the playing field. I
>     doubt there is any way to determine definitively whether that is a
>     good thing, or a bad thing.  YMMV.
>  2. wholesale replacement of tools has a direct analog in many other
>     areas.  Imagine for instance if, in trades (wood working, metal
>     working, etc.), someone suggested obsoleting the tools used by
>     experienced craftsmen (ignore any seeming gender specificity) in
>     order to (theoretically) simplify the learning curve for
>     apprentices.  The theory behind any such move would be that this
>     would allow apprentices to more quickly become qualified tradesmen
>     – but that theory would require time to validate and would likely
>     severely impact skilled production in the meantime.
>  3. in this particular work (i.e. – standardization) – as with a
>     number of other areas – “tools” is actually the area with the
>     least significant learning curve.  A far more significant learning
>     curve is associated with learning the many things that have been
>     tried, and determined to be unworkable, over a time period far
>     greater than the tools learning curve for even the most casual
>     tools user.
>
> I suspect that a prolonged discussion of the benefits of introducing 
> new tools is misplaced effort.
>

*applause from a long time Emacs user*

Mike


> --
>
> Eric
>
> *From:* ietf <[e-mail]> *On Behalf Of * Richard Barnes
> *Sent:* Wednesday, July 10, 2019 11:17 AM
> *To:* [e-mail]
> *Cc:* Julian Reschke <[e-mail]>; IETF Rinse Repeat 
> <[e-mail]>
> *Subject:* Re: Things that used to be clear (was Re: Evolving 
> Documents (nee "Living Documents") side meeting at IETF105.)
>
> You're absolutely correct that any tool comes with a learning curve.  
> The trick is "bringing the mountain to Mohammed" as it were; using 
> tools for which many people have already climbed the curve.  By 
> definition, anything that is IETF-custom is brand-new learning curve 
> for a newcomer.
>
> --Richard
>
> On Wed, Jul 10, 2019 at 10:12 AM Adrian Farrel <[e-mail] 
> <mailto:[e-mail]>> wrote:
>
>     Is it really that hard?
>
>     Even crusty old idiots like me have worked out how to use an
>     editor to make XML that is acceptable to XML2RFC. I doubt that I
>     am clever or more talented than Fellows of major engineering
>     organizations.
>
>     All tools (even github) require to be learned.
>
>     Replace tools with better tools, by all means.
>
>     But don’t make changes for personal preference: that way lies
>     unending debates about whose preference is best.
>
>     Thanks,
>
>     Adrian
>
>     *From:* ietf <[e-mail]
>     <mailto:[e-mail]>> *On Behalf Of *Richard Barnes
>     *Sent:* 10 July 2019 14:57
>     *To:* Randy Bush <[e-mail] <mailto:[e-mail]>>
>     *Cc:* Julian Reschke <[e-mail]
>     <mailto:[e-mail]>>; IETF Rinse Repeat <[e-mail]
>     <mailto:[e-mail]>>
>     *Subject:* Re: Things that used to be clear (was Re: Evolving
>     Documents (nee "Living Documents") side meeting at IETF105.)
>
>     I'm glad it works for you.  The non-IETF-habituated new authors
>     I've worked with have found it mystifying.  Including everyone
>     from junior engineers to Fellows of major engineering organizations.
>
>     As Christian says, our continued attachment to bespoke tools is a
>     barrier to getting new work in the IETF, and thus detrimental to
>     the long-term health of the organization.
>
>     --RLB
>
>     On Wed, Jul 10, 2019 at 9:47 AM Randy Bush <[e-mail]
>     <mailto:[e-mail]>> wrote:
>
>         >> Tooling is just one of the problems with XML2RFC. The real issue is
>         >> that XML2RFC is completely specific to the IETF. This
>         translate into
>         >> training ...
>         > It also has been optimized for the production of RFCs. Also
>         note that
>         > many changes in the v3 vocab just align the language with
>         HTML (lists
>         > and tables come to mind).
>
>         from an xml non-lover:
>
>         xml2rfc rocks!  it produces the baroqe, designed by committee,
>         internet
>         draft format from input which i can easily produce in my text
>         editor.
>
>         and the support, maintenance, and responsiveness of the tools
>         team is
>         simply stunning.
>
>         [ and for the poster who wished for a gooey, there is one ]
>
>         thank you!
>
>         randy
>