WUGH BoF Minutes
2017-03-27
Scribe: Paul Wouters

PaulH presenting slides
https://www.ietf.org/proceedings/98/slides/slides-98-wugh-chair-slides-00.pdf

Mark presenting slides
https://www.ietf.org/proceedings/98/slides/slides-98-wugh-draft-nottingham-wugh-services-00.pdf

Bernie Volz: How do you define WG use versus private use?
Mark: if participant needs to use the tool, it is a WG tool. If you
have a hard dependency on a tool, that's a WG use.

Erik Kline: what is recoverability

Mark: If the external service folds or starts asking money

Martin Thomson presenting:
https://www.ietf.org/proceedings/98/slides/slides-98-wugh-draft-thomson-github-bcp-00.pdf

Pete Resnick: [paraphrasing] Let's not stomp on a tool and get locked in again. Lets be agile.

Mike Bishop: frustating there is no agreed way for consensus on many github issues. Mark replied: With WG chair hat on, we are new and doing our best. Your point is valid.

Mike Bishop: Is consensus going to be determined on the mailing list in the future?

Ted Hardy: understand we may use different tools for different WGs and at different points within a WG. Let's not get stuck on using this tool. Evolution
should be part of how we think of this.

Benoit Claise: [explains some automated processing causing yang errors]. Splitting draft and code samples would help me

Martin: we have a way of automating commits to ensure commits have proven module code so we cannot have bad code in the draft.

Dave Crocker: I get the degree of adoption already happened. We stuck with
text only for too long. Accessability is still enormous. Barrier to entry
to github is huge compared to barrier to entry of txt draft. We are not
thinking carefully enough about [the process]. 

Mark: we put a fair amount of thought into that. Please look at our material [on jabber?] We are also lowering the bar because they know github but not [our official tools]

Joe Hildebrand: there is a link to the note well, so you cannot do a PR request
       without having read a note well. and we can track you back on your
       account on github.

Michael Richardson: The loop through the meeting still exists and it can be
jarring for a design team that works quickly, [ when we meet at next IETF
when no one has read the document changes]. I publish a long email with notes
to summarize our commits and that triggers [proper] discusison on the list.

Martin: it is crucial to take the changes back to the list to discuss.

Michael Richardson: pointing to git commits is not same as pointing to
draft versions.

Mark: we don't need design teams as much with this process.

Harald Alvestrand:  we switched from buzilla to github and it was very clear,
  that less is more.

Keith Moore: we want to develop documents to help open source software. Tools
for standards and software can be very different. Standards are better with
less features. opensource software and collaboration leads to more features.

Alexander: Recoverability of github might not be good enough. Mark/Martin: it is.

Cullen Jennings: Many of us are being sued whether or not someone is in violation of XXX, it is harder to match that with github ids. We do need to address these.

Martin: sometimes on the list we only have email addresses. Github is not that different.

Alisson Cooper (AD): Great discussion. There are some questions to explore.
Perhaps IETF can help with best practises and tools help. Tools or mailinglists rule the day for making WG decisions.... We want to help people using the
tools they can and want to work with. Please debate on the list.