RFC SERIES OVERSIGHT COMMITTEE (RSOC) August 23, 2018 RSOC Meeting Reported by: Cindy Morgan, IETF Secretariat ATTENDEES --------------------------------- Sarah Banks (Chair) Nevil Brownlee Heather Flanagan (RSE, non-voting) Tony Hansen Bob Hinden Cindy Morgan (Scribe, non-voting) Adam Roach Robert Sparks (Lead) Martin Thomson Portia Wenze-Danley REGRETS --------------------------------- Joel Halpern MINUTES --------------------------------- 0. Review of minutes The minutes of the 19 July 2018 meeting were approved. 0a. Agenda bash The following new items were added to the agenda: - RFCs at ISI - rfc6635bis - Publishing copies of RFCs that incorporate errata 1. Format work update Heather Flanagan reported that that the svgcheck tool is waiting for an update to the document with one last round of modifications. Rfclint is complete and will just be responding to bug reports and then the round of edits needed from the update of the RFC. Xmldiff still needs the source windows to be complete so that all source files are listed. Robert Sparks asked if the v3 text formatter has been tested yet. Heather Flanagan replied that it has, and that Alice Russo has sent some bug reports to Henrik Levkowetz. Tony Hansen noted that he has had some conversations with Henrik Levkowetz about making the tools available on the website. Heather Flanagan replied that they are not looking for community testers yet, as they would like to make sure that the tools work well enough for the RPC before opening them up for community testing. Tony Hansen suggested that they could be put on a hidden page or the experimental page. Robert Sparks noted that anyone who downloads the most recent version of XML2RFC will have the tools available, so he does not think there is an issue with having them linked from the experimental page. 2. IAB and RSOC changes Robert Sparks reported that the IAB ran out of time to talk about the RSOC appointments at their most recent meeting, but that he thinks there is a high chance the IAB will finish this up in the next couple of weeks. 3. Other in-progress items 3.1. Crossref and DOIs Heather Flanagan reported that the invoices from Crossref for assigning DOIs were going to Ray Pelletier's old email address instead of the generic iad@ietf.org address, so payment was past due and Crossref sent a shut-off notice. Portia Wenze-Danley has all of the old invoices and has updated the billing address, so DOIs should start being assigned again shortly. The RFC Editor has not held up publication of RFCs while this issue is being sorted out. 3.2. RFC Editor priorities Heather Flanagan reported that she is working on the list of RFC Editor priorities and is reviewing that with the RPC. She expects that this will be the main topic at the next RSOC meeting. 3.3. IPJ article Heather Flanagan reported that she is working on an article for the IP Journal about the RFC Series celebrating 50 years of RFCs. 3.4. Streams and statuses Heather Flanagan reported that she has started working with Dan York and Greg Wood on outreach regarding streams and statuses. 4. RPC update 4.1. GitHub experiment Heather Flanagan reported that she sent email to the IESG with the RFC Editor's experience using GitHub during AUTH48. She noted that while she can see why authors might like using it, it is not a particularly efficient tool for copy editing. They still need to determine how to revise the experiment before proceeding with the JSEP draft. updated statistics regarding AUTH48 times during this experiment are on the wiki at . Heather noted that if the RFC Editor does proceed with using GitHub as a collaborative editing tool, that will further specialize the skill set needed by editors, which will mean a longer ramp-up time when new editors come on board. Nevil Brownlee asked how the RPC feels about Eric Rescorla's suggestions for changing the experiment. Heather Flanagan replied that some of them would be easy to implement, but others, such as using Markdown or a different editing tool would only add extra complications. Bob Hinden said that it sounds like the experiment was not successful from the production center side, and that he does not think they should be forced to use GitHub. Robert Sparks suggested that it might be worth exploring what the impact on the RPC work flow would be if the editors' changes were separated out between text changes and whitespace changes. Heather Flanagan replied that that would be difficult because that is not how copy editing works; she talked about this with Alice Russo and Lynne Bartholomew from the RPC and they could not come up with a way to make that work. Heather suggested that Robert chat with them to see if he can bring in a different perspective that would make that work. Tony Hansen suggested that there is a front-end tool into GitHub that might make it easier for the RPC to use, but he is not familiar enough with that tool or with GitHub to say whether that would be useful. Heather Flanagan said that the RPC is also not familiar with that tool, and that she is hesitant to add another tool to the tool chain without knowing how it would affect the editors' work flow. Adam Roach said that if using GitHub becomes standard operating procedure, he suspects that some sort of tool on top of the existing GitHub UI will be required in order to make it work. Heather Flanagan replied that in such case, the would need to be a conversation about it with someone who is familiar with the GitHub universe and someone who is familiar with the copy editing universe. Heather said that she may try to find someone to have such a conversation with Alice Russo at the next IETF meeting. 4.2. Format tool testing Heather Flanagan reported that format tool testing is in process. The next big push for testing will happen once Henrik Levkowetz returns from vacation. Tony Hansen asked where the ancillary tools can be found. Robert Sparks replied that right now the RPC is still testing them, so they have not been announced to the community yet. 5. RFCs at ISI Heather Flanagan reported that the hard copies of early RFCs that are being stored at ISI have been moved to a physically better location, and that Bob Hinden asked if the question of getting the Computer History Museum to take those RFCs from ISI should be revisited. Heather Flanagan noted that when she has tried to broker the conversation in the past, she has not been able to make much progress because ISI does not want to give the RFCs up. Nevil Brownlee asked if ISI would make the RFCs available to anyone who would like to see them, like the Computer History Museum would. Tony Hansen asked if ISI would be amenable to having the RFCs scanned. Bob Hinden asked if anyone who isn't on the RSOC even knows that the physical copies of early RFCs are being held at ISI. Heather Flanagan said that she would ask for answers to all of those questions. 6. draft-ietf-iasa2-rfc6635bis Bob Hinden reported that he has a few more things to fix in draft- ietf-iasa2-rfc6635bis, "RFC Editor Model (Version 2)". The goal of this update is to bring the document up to date with IASA 2.0 rather that to make other changes. Nevil Brownlee pointed out that a similar update should probably be made to RFC 6548, "Independent Submission Editor Model." 7. Publishing copies of RFCs that incorporate errata Adam Roach reported that the IESG has begun discussing ways to make errata that has been approved more visible to implementers. He said that while the IESG is still early in these conversations, the direction seems to be leaning less towards publishing new documents and more towards making tools changes that would allow for the errata to be displayed inline. Robert Sparks said that displaying errata inline might be difficult because a large portion of errata reports are fairly general and do not include specific line changes. He suggested pulling up the 50 most recent errata and trying to see from those what that sort of automation would look like. He also noted that trying to force that errata reports be submitted in a format that could result in inline changes to the text might make it more difficult for people to call out problems, so the problems will end up not being reported. Martin Thomson suggested that changing the errata report into something that could result in inline text changes could be part of the verification step rather than part of the reporting step. Tony Hansen suggested that errata could be appended one per page to the end of a document. Adam Roach said that it sounds like there might be two different formats for this, where some errata are displayed inline and others are called out in the header of the document. Sarah Banks asked if displaying the errata inline would result in a versioning problem when responding to subpoenas. Heather Flanagan replied that she does not think that would make a difference; the original text is still there, and errata are currently available publicly anyway. Bob Hinden observed that there seems to be a version of this today already with the HTML RFCs on tools.ietf.org, and asked why that isn't sufficient. Adam Roach replied that it is not very visible as it is, and it would be better to have the errata appear at the point of the document that is affected. Heather Flanagan added that one of the things they learned during the RFC++ BOF is that people don't always pay attention to the headers of documents.