IETF
HTTPAPI
httpapi@jabber.ietf.org
Friday, March 12, 2021< ^ >
Rich Salz has set the subject to: HTTPAPI at IETF 109
Room Configuration
Room Occupants

GMT+0
[06:38:19] Glen (AMS IT) joins the room
[06:39:13] zulipbot leaves the room: Disconnected: closed
[06:39:19] zulipbot joins the room
[06:41:15] zulipbot leaves the room: Disconnected: closed
[06:41:27] zulipbot joins the room
[06:48:28] zulipbot leaves the room: Disconnected: closed
[06:49:48] zulipbot joins the room
[06:50:31] Glen (AMS IT) leaves the room
[06:58:48] zulipbot leaves the room: Disconnected: closed
[07:10:42] Glen joins the room
[07:10:52] zulipbot joins the room
[07:11:30] Glen leaves the room
[07:13:57] glen joins the room
[07:14:01] glen leaves the room
[07:29:15] zulipbot leaves the room: Disconnected: closed
[07:29:21] zulipbot joins the room
[07:29:43] Glen (AMS IT) joins the room
[07:29:48] <Glen (AMS IT)> Test
[07:30:05] <zulipbot> (Glen) Test
[07:30:29] Glen (AMS IT) leaves the room
[15:24:13] Darrel joins the room
[15:35:48] Darrel leaves the room
[15:37:02] Meetecho joins the room
[15:50:03] Erik Wilde_web_100 joins the room
[15:50:03] Alan Ford_web_771 joins the room
[15:50:03] Chris Lemmons_web_836 joins the room
[15:50:03] Julian Reschke_web_394 joins the room
[15:50:03] Alessandro Amirante_web_198 joins the room
[15:50:03] Rich Salz_web_227 joins the room
[15:50:03] Eric Kinnear_web_953 joins the room
[15:54:27] Francesca Palombini_web_663 joins the room
[15:54:48] Kenneth Murchison_web_814 joins the room
[15:55:15] francesca joins the room
[15:55:33] <francesca> hello!
[15:55:49] Michael Breuer_web_901 joins the room
[15:56:38] Mary Barnes_web_917 joins the room
[15:56:51] Jonathan Lennox_web_922 joins the room
[15:57:03] Robert Stepanek_web_423 joins the room
[15:57:09] Darrel Miller_web_781 joins the room
[15:57:14] <Rich Salz_web_227> hiya.  do you want to say anything at the start?
[15:58:02] <Erik Wilde_web_100> hello everybody!
[15:58:02] Cullen Jennings_web_774 joins the room
[15:58:26] <Erik Wilde_web_100> "HTTPAPI at IETF 109" in the chat window title...
[15:58:45] Barry Leiba_web_157 joins the room
[15:58:51] Sanjay Dalal_web_619 joins the room
[15:58:54] <Rich Salz_web_227> someone on jabber has to change the topic.  about 50% of the sessions i was in this week had it wrong.
[15:59:31] <francesca> Rich: no, it's fine :)
[15:59:33] Jiankang Yao_web_910 joins the room
[15:59:33] Rich Salz joins the room
[15:59:34] <zulipbot> (Darrel Miller) I'm using Zulip and I don't see that title.
[15:59:40] Gabriel Sullice_web_494 joins the room
[15:59:43] Rich Salz has set the subject to: HTTPAPI at IETF 110
[15:59:44] francesca has set the subject to: HTTPAPI at IETF 110
[15:59:45] <Mary Barnes_web_917> Yeah.  We traveled back to IETF 108 for STIR WG.
[15:59:52] Waqar Zia_web_916 joins the room
[16:00:04] Frode Kileng_web_101 joins the room
[16:00:11] Jiri Novotny_web_124 joins the room
[16:00:17] Hiroyuki Goto_web_527 joins the room
[16:00:22] <Sanjay Dalal_web_619> hello all
[16:00:23] francesca has set the subject to: HTTPAPI at IETF 110 - https://codimd.ietf.org/notes-ietf-110-httpapi
[16:00:31] Gabriel Sullice_web_494 leaves the room
[16:00:38] Gabriel Sullice_web_386 joins the room
[16:00:39] Mike Amundsen_web_881 joins the room
[16:00:50] Jeffrey Yasskin_web_761 joins the room
[16:00:51] Mark Nottingham_web_889 joins the room
[16:01:12] <Gabriel Sullice_web_386> hello!
[16:01:21] Mike Bishop_web_963 joins the room
[16:01:22] Alessandro Ghedini_web_337 joins the room
[16:01:46] Jon Peterson_web_222 joins the room
[16:02:05] Alex Huang Feng_web_142 joins the room
[16:02:18] Ali Begen_web_958 joins the room
[16:02:19] Tommy Jensen_web_583 joins the room
[16:02:30] Alan Frindell_web_702 joins the room
[16:02:31] Alexander Mayrhofer_web_544 joins the room
[16:02:32] Stephan Emile_web_501 joins the room
[16:02:39] Alex Huang Feng_web_142 leaves the room
[16:02:43] Alex Huang Feng_web_854 joins the room
[16:03:06] Phil Archer_web_685 joins the room
[16:03:14] Phil Archer_web_685 leaves the room
[16:04:49] Phil Archer_web_654 joins the room
[16:05:31] mnot joins the room
[16:06:12] Ulysse Carion_web_100 joins the room
[16:07:05] <Gabriel Sullice_web_386> Thank you for quickly explaining terms like "agenda bashing" Rich! I really appreciate it :pray:
[16:07:47] Rick Taylor_web_377 joins the room
[16:08:35] Stephan Emile_web_501 leaves the room
[16:11:46] <Alexander Mayrhofer_web_544> I'll give Deprecation a review with fresh eyes, new to this WG
[16:11:59] <Chris Lemmons_web_836> Thanks! That's really helpful.
[16:12:11] <mnot> +1 - I'll take a fresh look at it if we think it's getting close.
[16:12:21] <Julian Reschke_web_394> same here
[16:12:26] <Erik Wilde_web_100> apologies if i was jumping the queue here... ;-)
[16:12:39] <Erik Wilde_web_100> *waving*....
[16:12:43] <francesca> *waves back*
[16:12:59] <Mike Amundsen_web_881> :wave::wave:
[16:13:05] <Gabriel Sullice_web_386> I will post a comment on the repository, I have one small concern about the recommendation to use the `successor-version` link relation given that I am already using that rel for a document revisioning, but otherwise I am very enthusiastic about it! Looks very good overall.
[16:13:22] <Mike Bishop_web_963> "a few weeks" seems very optimistic.
[16:13:24] <Erik Wilde_web_100> thanks, @gabriel!
[16:13:38] <Erik Wilde_web_100> will do.
[16:14:09] Cullen Jennings_web_774 leaves the room
[16:14:25] <Chris Lemmons_web_836> My main question, which I can put on the mic if we're going to have conversation about it, is what we expect a client to actually do with this header? Is it designed for people or should things happen automatically based on it?
[16:15:05] <Gabriel Sullice_web_386> +1 to that question
[16:15:32] <mnot> Straightforward editorial suggestions are best as pull requests
[16:17:32] <Ulysse Carion_web_100> Paragraph 2 of Section 3.1?
[16:17:55] <Julian Reschke_web_394> +1 to mentioning logging
[16:20:06] <Chris Lemmons_web_836> I particularly worry that naïve implementations will do bad things with successor-version links.
[16:21:56] <Waqar Zia_web_916> Hi, cant find the link to slides in meeting material (there are 3x slides there), would be good to have the link. Thanks!
[16:22:44] <Erik Wilde_web_100> here are mine: http://localhost/github/lectures/ietf-110/
[16:22:52] Philipp Tiesel_web_719 joins the room
[16:23:00] <mnot> Not helping, Erik :)
[16:23:08] <Erik Wilde_web_100> ;-)
[16:23:24] <Julian Reschke_web_394> Chris: https://github.com/ietf-wg-httpapi/deprecation-header/issues/5
[16:23:28] <Erik Wilde_web_100> http://dret.net/lectures/ietf-110/
[16:23:33] <Ulysse Carion_web_100> If the draft says that the deprecation rel tells the client what docs the client developer should read, it does seem weird that the server would also send further information in the headers that those docs would presumably cover.
[16:23:43] <Chris Lemmons_web_836> Thanks, Julian!
[16:24:21] <Chris Lemmons_web_836> I was about to apologize for missing the fact that there was already an issue open for this, then I saw the timestamp on it. :D
[16:24:32] <Julian Reschke_web_394> :-)
[16:24:37] Philipp Tiesel_web_719 leaves the room
[16:25:27] <Chris Lemmons_web_836> The less ambiguously the server conveys it's intent, the more effectively clients can make decisions about what actions they should take.
[16:25:28] <Erik Wilde_web_100> thanks everybody!
[16:25:52] Hiroyuki Goto_web_527 leaves the room
[16:26:43] Hiroyuki Goto_web_602 joins the room
[16:27:45] <Rich Salz> Thanks for raising the issues Julian :)
[16:28:12] <Chris Lemmons_web_836> Who needs linkset when we have rfc1436 directory objects? :D
[16:28:48] Marc Blanchet_web_140 joins the room
[16:28:52] Marc Blanchet_web_140 leaves the room
[16:29:17] Marc Blanchet_web_268 joins the room
[16:29:23] Rick Taylor_web_377 leaves the room
[16:29:40] <Julian Reschke_web_394> The fact that I originally defined these link relations was a motivation...
[16:34:04] Michael Breuer_web_901 leaves the room
[16:34:08] Michael Breuer_web_394 joins the room
[16:34:26] <Phil Archer_web_654> https://id.gs1.org/01/9506000134352?linkType=all
[16:35:03] <Erik Wilde_web_100> risotto!
[16:35:15] Jiankang Yao_web_910 leaves the room
[16:35:19] Jiankang Yao_web_428 joins the room
[16:37:58] <Erik Wilde_web_100> we just define the media types, and not a protocol around it.
[16:38:01] Jiankang Yao_web_428 leaves the room
[16:38:07] Jiankang Yao_web_212 joins the room
[16:38:08] <Chris Lemmons_web_836> Right.
[16:39:48] <Erik Wilde_web_100> let's move on.
[16:40:04] Alexander Mayrhofer_web_544 leaves the room
[16:40:10] <Phil Archer_web_654> Thanks Erik!
[16:41:18] <zulipbot> (Darrel Miller) The question is how long Mark manages to stay awake for.  We appreciate you being up in the middle of night Mark.
[16:41:27] Alessandro Ghedini_web_337 leaves the room
[16:47:42] <Jeffrey Yasskin_web_761> Mark's lighting really emphasizes that we've kept him up late.
[16:49:12] <Marc Blanchet_web_268> both are viable: have the wiki for proposals and "discussions" and "promote" the "good" ones to the IANA registry. IANA registry being the normative one
[16:49:21] <Sanjay Dalal_web_619> +1 Darrel. Completely agree.
[16:50:42] Alan Ford_web_771 leaves the room
[16:51:01] <Julian Reschke_web_394> In WebDAV, problem types (failed preconditions/postconditions) are just URIs
[16:52:22] <mnot> E.g., we manage link relation requests here now: https://github.com/protocol-registries/link-relations
[16:53:39] <Mike Amundsen_web_881> https://www.rfc-editor.org/rfc/rfc8720.html ?
[16:54:08] Julian Reschke_web_394 notes that being a designated expert for a IANA registry is really popular
[16:54:15] <Mike Amundsen_web_881> ;)
[16:54:16] <Chris Lemmons_web_836> I like Specification Required, but not necessarily RFC Required.
[16:56:11] <Erik Wilde_web_100> that's a classic!
[16:56:38] <Chris Lemmons_web_836> Creating a registry like this allows people to host their own.
[16:57:03] Alexey Melnikov_web_861 joins the room
[16:57:52] <Chris Lemmons_web_836> +1 on multiple problems. We should solve this.
[16:58:27] <Marc Blanchet_web_268> the RDAP Bootstrap registries (RFC7484) are hosted by IANA and are used operationally. IANA put it on their CDN provider and no issues have been found with 2 years of operation.
[16:58:49] Alex Huang Feng_web_854 leaves the room
[16:58:51] <zulipbot> (Darrel Miller) Having multiple problems would definitely be valuable in my personal opinion
[17:00:54] Alan Frindell_web_702 leaves the room
[17:01:16] <mnot> Marc: How many clients?
[17:01:55] <mnot> We're talking about potentially very many clients here, and some of them may not be disciplined - or easily contactable.
[17:02:23] <mnot> +1 to chris
[17:03:36] Julian Reschke_web_394 is reminded about status code 207 (and now ducks)
[17:03:55] Alexey Melnikov_web_861 leaves the room
[17:05:01] <Erik Wilde_web_100> currently, the warning draft proposes 2xx responses.
[17:05:04] Ali Begen_web_958 leaves the room
[17:05:41] <mnot> If you embed in another format, the status code can be whatever it needs to be
[17:05:57] <Marc Blanchet_web_268> Mark: nowadays, I don't see this as an issue. if you think a typical web page today, you would be downloading various JS framework files that are often much larger than what we are talking about here. So to me that issue of "downloading standard-related files" is moot (given that you put them on a CDN).
[17:06:18] <mnot> Marc: Ask W3C how much bandwidth they burn on this :)
[17:06:45] <zulipbot> (Darrel Miller) Julian:  I see lots of usage of 207
[17:07:04] <Julian Reschke_web_394> so do I :-)
[17:07:12] <mnot> The Warning header is deprecated
[17:07:31] <Erik Wilde_web_100> the warning draft proposed a different header because of that.
[17:07:41] <Marc Blanchet_web_268> Mark: historically maybe, but I can't see this as a big issue now. again, a single web page would load JS frameworks from a CDN.
[17:11:12] Michael Breuer_web_394 leaves the room
[17:11:51] <Erik Wilde_web_100> i like the "truncation warning" idea!
[17:12:25] <Chris Lemmons_web_836> And yeah, I'm using "warning header" to mean "whatever header the Warning draft is defining".
[17:13:04] <Erik Wilde_web_100> it's "Content-Warning", technically, as mentioned by darrel.
[17:13:12] <Chris Lemmons_web_836> Right.
[17:15:20] <Erik Wilde_web_100> i see XSD Part 1 coming up... ;-)
[17:17:44] Hiroyuki Goto_web_602 leaves the room
[17:19:01] <mnot> RUN AWAY
[17:19:18] <Chris Lemmons_web_836> I think this smells a lot like CBOR tags.
[17:19:20] <Julian Reschke_web_394> GTFO?
[17:19:29] <Erik Wilde_web_100> now XSD Part 2 is creeping up...
[17:19:38] <mnot> SAVE YOURSELVES
[17:19:56] <francesca> Chris +1
[17:19:58] <Gabriel Sullice_web_386> I think your characterization of JSON Schema does not do it justice. The definition of an "object" and its properties and the ability to mint an ID for that definition makes it possible to create an "address" type
[17:19:59] <Julian Reschke_web_394> wrong WG?
[17:20:19] <Chris Lemmons_web_836> Can we just map CBOR tags into JSON and save ourselves a lot of duplicated work? :)
[17:20:45] <mnot> First you have to educate 99% of the developers in the world what CBOR is, and get over the innate NIH
[17:20:58] <Gabriel Sullice_web_386> Those definition IDs are URIs
[17:21:03] <Chris Lemmons_web_836> Heh. There are also mapping differences that make that not trivial, too.
[17:22:07] <mnot> That was in response to XSD reference :)
[17:22:15] <Ulysse Carion_web_100> Can you provide an example beyond linkset of what could go in this registry?
[17:22:18] <Julian Reschke_web_394> if this happens in httpai, please add a separate mailing list :-)
[17:22:27] <mnot> ^^ julian ++_
[17:22:32] <Chris Lemmons_web_836> But I do find that every time I have a CBOR tag, I wind up making a JSON version of it for clients that Accept: json instead.
[17:22:32] <Julian Reschke_web_394> you'd need a JSON related WG
[17:22:38] <Julian Reschke_web_394> right
[17:24:29] <Ulysse Carion_web_100> I think this is the last time JSON Schema has approached a WG: https://mailarchive.ietf.org/arch/msg/json/w--4QtcF9remEI3FfGM6ghBFkEQ/
[17:24:34] <Julian Reschke_web_394> problem is that there currently is no JSON WG
[17:24:58] <Ulysse Carion_web_100> ^ yeah "WG" in air quotes. json wg was already done at the time
[17:25:08] <francesca> jsonbis has been terminated ... end of 2017
[17:25:18] <Chris Lemmons_web_836> jsontre when?
[17:27:10] mnot is going to bed. This is an interesting discussion, I look forward to reading more about it on list.
[17:27:47] <Jonathan Lennox_web_922> Take to dispatch?
[17:27:55] <mnot> which dispatch?
[17:27:58] <mnot> dispatchdispatch?
[17:28:00] <Phil Archer_web_654> Can this be done with a JSON-LD context file, i.e. define a class structure? Or am I being very Linked Data-y
[17:28:07] Mark Nottingham_web_889 leaves the room
[17:28:07] <Jonathan Lennox_web_922> ART presumably, this is appsy
[17:28:13] mnot leaves the room
[17:30:49] <Rich Salz_web_227> can't take to dispatch without a draft first
[17:32:27] <Mike Amundsen_web_881> thanks to all!
[17:32:28] <Erik Wilde_web_100> thanks everybody! time to cook dinner!
[17:32:28] <Gabriel Sullice_web_386> Thank you!
[17:32:29] <Sanjay Dalal_web_619> Thank you all
[17:32:30] <Chris Lemmons_web_836> Thank you to the draft authors!
[17:32:33] <Phil Archer_web_654> Thanks everyone
[17:32:35] <francesca> thanks all! Thanks Rich and Darrel
[17:32:37] Barry Leiba_web_157 leaves the room
[17:32:39] <Waqar Zia_web_916> Thanks all!
[17:32:41] Mike Amundsen_web_881 leaves the room
[17:32:41] Tommy Jensen_web_583 leaves the room
[17:32:42] Phil Archer_web_654 leaves the room
[17:32:42] <Erik Wilde_web_100> thanks to the MCs!
[17:32:42] Frode Kileng_web_101 leaves the room
[17:32:42] Rich Salz_web_227 leaves the room
[17:32:43] Sanjay Dalal_web_619 leaves the room
[17:32:43] Waqar Zia_web_916 leaves the room
[17:32:43] Darrel Miller_web_781 leaves the room
[17:32:44] Jiri Novotny_web_124 leaves the room
[17:32:44] <Jon Peterson_web_222> bye
[17:32:44] Mike Bishop_web_963 leaves the room
[17:32:44] Ulysse Carion_web_100 leaves the room
[17:32:45] Jonathan Lennox_web_922 leaves the room
[17:32:47] Francesca Palombini_web_663 leaves the room
[17:32:47] Jon Peterson_web_222 leaves the room
[17:32:47] Julian Reschke_web_394 leaves the room
[17:32:48] Kenneth Murchison_web_814 leaves the room
[17:32:49] Chris Lemmons_web_836 leaves the room
[17:32:50] Mary Barnes_web_917 leaves the room
[17:32:54] <zulipbot> (Darrel Miller) Thank you to everyone
[17:32:55] francesca leaves the room
[17:32:59] Erik Wilde_web_100 leaves the room
[17:33:05] Gabriel Sullice_web_386 leaves the room
[17:33:11] Robert Stepanek_web_423 leaves the room
[17:34:12] Eric Kinnear_web_953 leaves the room
[17:34:12] Alessandro Amirante_web_198 leaves the room
[17:34:12] Jeffrey Yasskin_web_761 leaves the room
[17:34:12] Jiankang Yao_web_212 leaves the room
[17:34:12] Marc Blanchet_web_268 leaves the room
[17:35:37] Rich Salz leaves the room
[17:39:51] Meetecho leaves the room