This is the recent traffic on the #SPF-council IRC channel on irc.pobox.com. Anyone may join the channel, but only council members can talk.

If you do not have access to IRC, you may view the recent traffic at: http://www.schlitt.net/spf/spf-council/now/irc_log.html.

This log can be can be viewed at: http://www.schlitt.net/spf/spf-council/2005/07/18_irc_log.html.

IRC nicknames:
freesideMeng Weng Wong
gconnorGreg Connor
grumpyWayne Schlitt
JulianJulian Mehnle
MarkKMark Kramer (asarian-host.net)

--- Mon Jul 18 13:21:42 UTC 2005 ---
13:21<gconnor_l>hey guys, I'm just getting ready to drive to work, I should be online again in 40 min
14:09<gconnor>Hi guys... I'm here
14:09<gconnor>anyone else online?
14:47<grumpy>hi
14:47<grumpy>am I late?
14:49<Julian>I'm here.
14:49<Julian>Can we start?
14:50<Julian>We're 45min late for the informal session.
21:21<freeside>moo.
21:38<Julian>Today's meeting agenda: http://moongroup.com/pipermail/spf-council/2005-July/000373.html
21:52<freeside>mmm.
21:54*grumpy pokes his head in
21:55<gconnor>5 minutes to go
21:55<freeside>moo.
21:58<MarkK>good evening
21:59<grumpy>Uh, are we all here arely?
21:59<Julian>
21:59<grumpy>Ok, well, lets start!
22:00<Julian>Beep.
22:00<grumpy>First up: item 1: minutes...
22:00<grumpy>Julian?
22:00<Julian>Yeah, sue me. ;-)
22:00<grumpy>Ok
22:00<grumpy>Next up: Chairman's report: Wayne
22:00<Julian>I promise to catch up on writing the minutes within the next two or three weeks.
22:00<freeside>i'm here
22:01<grumpy>Uh, last week I spent mostly preparing for, going to, returning from, and recovering from the Email Authenetication Summit.
22:01<grumpy>It was a *very* good show.
22:01*freeside listens with gr8 interest
22:01<grumpy>Greg Connor and I are editing a summary of it that we should send to the list SOON.
22:02<grumpy>Any Minute Now.
22:02<Julian>No.
22:02<grumpy>no?
22:02<Julian>No, I object.
22:02<grumpy>ok
22:02<grumpy>why/what?
22:03<Julian>gconnor said he wanted to send it "aroung 15:00 UTC". That's ok, but I need to give some input before he sends it.
22:03<grumpy>ok.
22:03<Julian>It is not 15:00 "Any Minute Now", so...
22:03<gconnor>Understood. I will wait.
22:03<grumpy>"Any Minute Now" means "within the next year".
22:03<grumpy>"Real Soon Now" can be decades...
22:03<Julian>Yeah, sure.
22:03<grumpy>;-)
22:04<Julian>So what was "15:00 UTC" all about then?
22:04<grumpy>Ok, anyway, unless someone objects, I think I should just go with the stuff gconnor is editing rather than give a duplicate report.
22:04<Julian>*sigh*
22:04<grumpy>freeside: the first copy was sent to spf-private. You should have seen it.
22:04<grumpy>Julian: ?
22:04<Julian>Why did the informal session have to be private in the first place?
22:04<gconnor>Julian- Just wanted to get everyone's feedback sooner rather than later. I will hold for your comments.
22:04<grumpy>it didn't.
22:04<grumpy>that's hindsight though...
22:05<Julian>gconnor: I assumed you were going to wait until 15:00 UTC, which is in about 17h, so why the hurry now?
22:05<gconnor>Agreed. I wanted to make sure everyone felt free to discuss anything that needed it. And because we were not intending it as an "official" action
22:05<grumpy>gconnor: did you mean 15:00 PDT rather than UTC?
22:06<Julian>Ok, so now we're stuck with the community not yet knowing what we're talking about for the next 17h.
22:07<gconnor>Sorry the 15:00 UTC was a typo on my part, I meant to say 22:00. But regardless, since we didn't say from the beginning that we were logging, I want to get permission from everyone there. Go ahead and get your comments to me asap... I will hold till I have all comments.
22:07<Julian>Yes, please do. Thanks.
22:07<gconnor>No problem.
22:07<grumpy>Ok, so we are going to try and get the report out asap.
22:07<Julian>So can we get on with the meeting right now?
22:07<grumpy>yeah,
22:08<grumpy>anyone have any other questions for the temp-chair?
22:08<grumpy>type ... quickly
22:08<MarkK>Julian: I had set the channel to 'private', thinking an informal talk with gconner did not constitute a council session, 'private' or otherwise. I thought we were just a bunch of guys talking. :)
22:08<grumpy>MarkK: I agree, we were just a bunch of guys talking. However, things like the Email Auth report would be very good to publish
22:08<grumpy>Ok, let's move on...
22:08<MarkK>grumpy: yes
22:08<gconnor>that's what I was thinking too. If it didn't contain stuff about the summit I probably wouldn't have proposed to email it out.
22:08<grumpy>Next up: ED report: Meng
22:09<grumpy>Meng: go
22:09<Julian>For the record, I don't agree with Frank that we cannot have official but private discussions. But we were definitely discussing council business today, so it should be public unless there's a explicit requirement for privacy.
22:09<grumpy>freeside?
22:10*grumpy wonders if freeside has horrible lag again....
22:10<Julian>The ping time in Japan must be inversely proportional to the band-width.
22:10<grumpy>hmmm....
22:11<grumpy>Uh, next item is "elect new chair", which I don't want to do w/o freeside
22:11<grumpy>could we quickly skip to item 5: status of website?
22:11<Julian>Why not.
22:11<grumpy>s/could/should/?
22:11<Julian>Sure, sure.
22:11<freeside>ergh
22:11<freeside>sorry.
22:11<freeside>keep going.
22:11<grumpy>Hi freeside!
22:11<freeside>i have nothing new to report really.
22:12<grumpy>ok.
22:12<Julian>???
22:12<Julian>freeside: No news from the emailauthenticationsummit?
22:12<freeside>had lunch with wayne ...
22:13<freeside>everything seems on track to me
22:13<Julian>How's karma going?
22:13*Julian is sorry for having missed the Karma BoF at the MAAWG conference.
22:14<Julian>If you don't want to talk about it, please say so.
22:14<grumpy>freeside?
22:14<Julian>I'll shut up then.
22:15<grumpy>freeside?
22:15<MarkK>freeside?
22:16<MarkK>Ok, we definitely need Meng for electing the new chair
22:16*grumpy agrees
22:16<Julian>I hate this.
22:17<grumpy>Uh, let's move on tothe "website report" until Meng returns....
22:17<grumpy>Julian: go
22:18<Julian>...
22:20<Julian>Ok. I have spent a lot of hours trying to package the website software -- we should use packaged software wherever we can in order to reduce maintenance efforts (like upgrading, re-installation, etc). Unfortunately, I was having trouble with the installer module, Module::Build. I just found out how to use a special but important feature of M::B about 20 minutes ago.
22:20<freeside>sorry, i'm neither here nor there, let me try to be more here in about 5 minutes.
22:20<Julian>So I should be able to get packages done and installed on spf.schlitt.net tomorrow.
22:21<Julian>...
22:21<grumpy>Julian: (answer whenever) which software are you using? the same wiki as you used for the spf council?
22:22<Julian>Then the website will be set up with a basic design, a council sub-section (with all the stuff moved from http://spf.mehnle.net), and a publicly editable sub-section, so we will finally be able to start creating the new site.
22:22<Julian>Yes, the same, but in a cleanly packaged form.
22:22<Julian>Any specific reasons you're asking?
22:22<grumpy>no, just curious...
22:23<gconnor>I like wikis :)
22:23<Julian>Ok, that's about it.
22:23<grumpy>that was what I though, but I thought that software was already packaged...
22:23<Julian>Not really. There's a UseMod package in Debian, but not for the extended version.
22:23<grumpy>Ok, anyone else have questions for Julian?
22:24<grumpy>Well, ok, if that's done, let's move on to item 6: domain name (wayne)
22:24<gconnor>no, good job, let us know if you need a hand with anything
22:24<grumpy>Uh, I'm still in the process of getting the new domain name set up.
22:25<Julian>What's up with the domain transfers?
22:25<grumpy>As you know by my couple of messages to spf-private, I'm talking with James.
22:25<MarkK>yes, let us know; I know installing these packages can give yiu quite a run for your money :)
22:25<grumpy>James has confirmed that he *will* transfer the domains.
22:25<Julian>Cool.
22:25<Julian>Very nice of him.
22:25<grumpy>I haven't heard back from him since last Friday(?).
22:25<MarkK>decent chap!
22:25<grumpy>yep
22:26<Julian>What's holding the transfer up?
22:26<grumpy>I'm also working with JohnP about getting all the other associated domains transfered.
22:26<MarkK>Wasn't he on godaddy too? It should be real easy then
22:26<grumpy>Julian: hang on.
22:26<Julian>grumpy: I hope you're aware that you (as the chosen trustee) will have to pay for all domains transferred.
22:27*Julian just tripped on a mine.
22:27<grumpy>The open-spf.{org,com,net,info} and the openspf.{com,net,info} domains are currently in the process of having their primary name servers switched over to dns.schlitt.net
22:27*Julian should be playing Tetris instead.
22:27<grumpy>Julian: yeah, I just realized that today, but it isn't any problem.
22:27<Julian>With "transfer" I mean owner-c and admin-c transfer.
22:27<freeside>moo. okay, i'm here more.
22:28<grumpy>I just checked before the meeting and johnp's domains are still in the process of being updated.
22:28<grumpy>we wanted to get it working before the transfer in case the transfer takes a while.
22:28<grumpy>Ok, back to openspf.org and James.
22:29<Julian>"we wanted to get it working before the transfer in case the transfer takes a while" -- Good thinking.
22:29<grumpy>James has requested, and I think it is fair, to put a note on the website thanking the domain name owners for their donations and have a pointer to their new websites.
22:29<grumpy>This is similar to firefox.com
22:30<MarkK>yes, that sounds only fair
22:30<grumpy>Now, it appears that firefox.com is not really used as firefox's primary website, so they have a large blurb on the front page.
22:30<grumpy>I don't think we can do that for James/JohnP
22:30<Julian>Uh, I hope James doesn't mean a front-plate like on firefox.com, does he?
22:30<grumpy>I'm not certain
22:31<grumpy>as per the email sent to spf-private, I asked him about that.
22:31*Julian wonders why he hasn't read that message.
22:31<Julian>What subject does it have?
22:31<Julian>Oh, there it is.
22:31<gconnor>I don't think we have any other acknowledgements (of any kind) on the home page now, do we?
22:32<gconnor>I'm sure we can find an appropriate place for an acknowledgement.
22:32<grumpy>however, I think we can more forward by sanctioning a special "tahnk you" page for the domain owners and a line at the bottom saying something along the lines of "This websit domain name was donated by James and JohnP", with a link to a large web page.
22:32<grumpy>gconnor: no, I don't think we do.
22:32<grumpy>this somewhat gets back to the new website also.
22:33<Julian>I agree that a prominent acknowledgement is not in order.
22:33<grumpy>Mind you, James made it clear that such an acknowledgement is a courtesy, it is not a requirement for the transfer.
22:33<freeside>word.
22:33<MarkK>not as big as that on the firefox page, no
22:33<grumpy>yeah.
22:34<Julian>A line at the bottom with a link to any site he wishes would be ok, though.
22:34<grumpy>but something on the donate.html page, and a link on the front page seems appropriate to me.
22:34<Julian>(any site except goatse.cx)
22:34<grumpy>Julian: that is safe for work now a days
22:34<gconnor>We should probably have an acknowledgements page somewhere on the site. that would be a good place for the thank-you message.
22:35<Julian>grumpy: Can you take care of the acknowledgement for the current site?
22:35<grumpy>yes.
22:35<Julian>k
22:35<grumpy>do we need to have a motion on this?
22:35*grumpy says no, but...
22:35<gconnor>I don't think so, everything we said seems to fall into progress report or existing/ongoing work
22:35<Julian>I wouldn't know about what a motion was required.
22:36<grumpy>ok...
22:36<grumpy>let's move back to the ED report
22:36<freeside>Just Do It :)
22:36<grumpy>I think freeside is here now...
22:36<freeside>ED report, uh, email authentication summit went well, and i'm now curious to hear what people think about PRA/v=spf1 reuse...
22:36<freeside>MS gave a fairly balanced presentation so MAIL FROM was well represented.
22:37<Julian>Interesting.
22:37<freeside>i still think the advantages of reusing v=spf1 for PRA outweigh the disadvantages.
22:37<grumpy>I think we left off on the ED report with Julian asking about the Karma project....
22:37<Julian>freeside: Well, I think we have discussed that at great length in the past. :-)
22:37<Julian><Julian> How's karma going?
22:37<Julian>* Julian is sorry for having missed the Karma BoF at the MAAWG conference.
22:37<Julian><Julian> If you don't want to talk about it, please say so.
22:37<freeside>oh, karma is moving along well, making relationships with the right people.
22:38<freeside>had nice long chats with several ceo types at the auth summit and received lots of good advice from them on how to build it, which was heartwarming.
22:38<freeside>tried to bend esther dyson's ear about it too but she's so busy she said let's wait until we see each other again at foo camp later thi syear.
22:38<freeside>which is fine by me :)
22:39<Julian>Would it make sense to involve the SPF project in any way? Probably not, but I'm just asking.
22:39<grumpy>(if anyone has questions for the ED, go ahead and ask while freeside is typing...)
22:40<grumpy>freeside: are you still typing?
22:41<freeside>i'm done.
22:41<grumpy>oh... ok.
22:41<grumpy>any questions?
22:41<grumpy>if so, type ... quickly
22:41<Julian>$_
22:41<freeside>the SPF project is involved ...
22:41<freeside>but right now the onus is on me to build out the karma system a bit more before we can do real integration.
22:42<Julian>Ok. Call for help if you need it.
22:42<grumpy>Ok, next item: elect a new chair
22:42<grumpy>Uh... I guess we should discuss things....
22:43*grumpy isn't sure how to start this.
22:43<gconnor>What does the chair normally do?
22:43<gconnor>(since I'm new I can ask :)
22:43<grumpy>The chair runs the meeting.
22:43<grumpy>the chair sets up the meeting, makes the agenda, makes rulings on procedures, etc.
22:43<Julian>"The Council Chair presides and moderates the council meetings, and prepares the meeting agendas with input from the council."
22:44<MarkK>And is eminently qualified to create press releases :)
22:44<gconnor>Is it feasible to switch off/rotate chair position?
22:44<Julian>I guess the chair also takes care of personnel issues of the council.
22:44<grumpy>normally, the chair tries to be as impartial as possible and will only vote when the chair's vote would change the outcome.
22:44<Julian>No, I disagree.
22:44<grumpy>Julian: no to what?
22:45<Julian>Chuck chose not to vote unless required to, but this is not a rule.
22:45<gconnor>Well... voting last, you still get a vote. ;)
22:45<MarkK>julian: I agree to the disagree; I think the chair surely has an active vote too
22:45<grumpy>Julian: that is pretty typical of the role of a chair. I've seen it in other organizations and it is, for example, documented in Robert's Rules.
22:45<Julian>In fact, I'd much more like to see the chair cast votes.
22:46<Julian>"it's typical" has never been a good argument for anything.
22:46<Julian>Anyway, the chair can of course choose to abstain.
22:46<MarkK>grumpy: I agree this holds true for larger boards; for us 5, though, I think the chair needs a more active role
22:47<Julian>The chair not casting his vote most of the time hampers representativity.
22:47<Julian>At least with small bodies.
22:47<grumpy>well, this is part of the checks and balances. Since the chair controls the agenda, giving the chair too much power can cause problems.
22:47<MarkK>I mean, chuck sure was not shy of voicing his opinion :)
22:47<grumpy>I haven't been shy either as my 1.5 meetings as temp-chair.
22:48<Julian>Anyway, as I said, the chair can choose to vote or not to vote. It's up to him.
22:48<grumpy>it is not that the chair isn't supposed to speak, it is just that the role of the chair requires a certain amount of impartialness
22:48<gconnor>I would agree with Mark. All five of us should be active, and vote if we have opinions
22:48<grumpy>freeside: feel free the voice your opinion, you shy guy... ;-)
22:50<Julian>Also, "typically" the chair leads the preparation of press releases.
22:50<gconnor>while waiting on him, any thoughts on rotating the chair position once in a while? mod 5 the meeting number or something?
22:50<Julian>I don't think we need to do that.
22:50<grumpy>gconnor: we could do that, although our meeting schedule is so inconsistent that it might be tough
22:51<grumpy>MarkK: thoughts on a rotating chair?
22:51<Julian>gconnor: BTW, have you browsed the council website?
22:51<grumpy>freeside?
22:51<MarkK>I am not opposed to rotating chairs
22:51<gconnor>Understood. Does anyone *not* want to be considered? As in, normally don't have time to send agendas, etc? (I am not seeking the position but no reason I can't serve)
22:52<Julian>I already have a position. I don't want to be the chair as long as I'm the secretary.
22:52<grumpy>I'm quite willing to let someone else be the chair as I have plenty of other tasks that I'm working on.
22:53<grumpy>I will warning who ever becomes the chair, I will push for meetings and getting agendas out and such.
22:53<MarkK>gconner: indeed; I wanted you here for the vote: so you could vote, and to be potentially voted. :)
22:53<gconnor>julian- I am not sure if I have browsed the site... I know that it was announced a while ago but I forgot there was one... sorry about that. :)
22:53<Julian>For now, it is still http://spf.mehnle.net -- but soon it will be a part of http://www.openspf.{org,com}.
22:54<grumpy>spf.mehnle.net is a good website. Thanks Julian!
22:54<gconnor>Normal frequency of meetings?
22:54<Julian>Uh, strike .com. We agreed on .org.
22:54<MarkK>grumpy: if, say, you become the chair, a more 'limited' role, as you outlined above, would perhaps hamper your valuable inputs
22:54<grumpy>MarkK: maybe...
22:54<Julian>gconnor: Every other week. But we never really achieved that.
22:55<MarkK>maybe the new chair should be a bit more strict on that :)
22:55<grumpy>gconnor: we started out trying to have meetings every week, but we really usually don't have enough material to need it that often.
22:55<grumpy>we then went to every other week, but that was theory rather than practice.
22:56<grumpy>during the final days of finishing up SPF I-D, we met more than once per week.
22:56<gconnor>I am willing to spend time on it. (My naive estimate would be ~2 hrs /week apart from the actual meetings, but I'm comfortable committing to 5 hrs/week)
22:56<Julian>No, I don't agree to that, Mark. I would like for grumpy to become the chair, and I don't think that this would (or should) restrict the input he can give. If you mean that this costs him valuable time that should better be spent on other things, you _might_ be right, but I'm not sure. Only he can tell.
22:57<gconnor>That's as close as I get to self-promotion. Though being the new guy, I will probably be a bit passive anyway, at first, if passive is desired. Or at least "not likely to abuse chair position"
22:57<Julian>I am not against gconnor being the chair, however.
22:57<grumpy>I really want to move things forward. I have plent to do w/o being the chair, so if someone else wants to take charge of that part, I think that would be great.
22:57<grumpy>what I *don't* want is to have to spend as much time pushing the chair to get agendas out and such as it would take for me to actually do it myself.
22:58<gconnor>Whoever is the chair, if it's not grumpy, I expect grumpy to keep him honest :)
22:58<Julian>Motion: Wayne shall be the new chair. (Feel free to make a different motion.)
22:58<MarkK>julian: no, I meant if he restricts the role of chair to just one who sets the agenda and such, then I think I rather see him remain more active
22:58<Julian>MarkK: Oh, ok.
22:59<grumpy>gconnor: how are you about writing PRs and such?
22:59<gconnor>I'm a good writer. I'm very good at being diplomatic while still telling it like it is.
22:59<grumpy>MarkK is right, writing PRs and giving a voice to the community isimportant.
23:00<Julian>grumpy: I know.
23:00<grumpy>freeside? comments?
23:00<Julian>22:50 <Julian> Also, "typically" the chair leads the preparation of press releases.
23:00<grumpy>Oh, was ityou?
23:00<grumpy>*blush*
23:01*grumpy actually tried to read the scrollback, but gave up.
23:01<MarkK>I am quite content with gconner's dimplomatic skills, as demonstrated in his moderator role
23:02<Julian>So, anybody please second my motion or make another.
23:02<Julian>ping freeside
23:02<grumpy>Well, in reality, PRs should be bounced off of at *least* one other council member, even if the chair just "signs" it...
23:02<Julian>"I'm fine with whatever you guys decide."
23:03<Julian>All PRs should be approved by the council before release.
23:03<grumpy>Ok, is someone going to second 22:58u, or make another motion?
23:03<Julian>BTW, I still think we can make formal decisions on spf-private or spf-council.
23:03<gconnor>I was waiting to see if you would second ;)
23:04<MarkK>22:58u: seconded
23:04<grumpy>ok, votes?
23:04<freeside>argh, i just dropped off.
23:04<grumpy>freeside: comments?
23:04<freeside>my connectivity sucks, sorry.
23:04<freeside>no comments.
23:04<Julian>2258u: <Julian> Motion: Wayne shall be the new chair. (Feel free to make a different motion.)
23:05<Julian>2258u: yes
23:05<MarkK>22:58u: yes
23:05<freeside>2258u: yes
23:05<gconnor>2258u:yes
23:05<grumpy>motion carried.
23:05<grumpy>ok, well, anything else to discuss about the new chair?
23:05<Julian>grumpy: There you have it. ;-) Do you accept?
23:05<grumpy>if so, type ... uqickly.
23:05<grumpy>oh, yeah, I accept.
23:05<Julian>Thanks.
23:05*grumpy is used to being a BOFH
23:05<grumpy>I'm sure being a BCFH isn't much different.
23:06<MarkK>Is it the exclusive role of the chair to do press releases? I can help there too
23:06<gconnor>Tag me (or any of us) if you want some writing or any other help... chair doesn't mean do all the work ;)
23:06<Julian>Everybody can help writing the PRs.
23:06<MarkK>I think we need to write more, namely :)
23:06<grumpy>Well, the most important thing about the PR is finding someone who can send them to prnewswire, or whoever.
23:06<Julian>The PRs are usually thrown to spf-private before release, so all council members can comment.
23:07<Julian>grumpy: Yeah, that's a real issue. I don't have access to prnewswire.
23:07<grumpy>would someone like to volunteer into finding out exactly how we can get access to prnewsire?
23:07<grumpy>I'm willing to donate money, but not if it is $10k/yr + $500/PR...
23:07<Julian>I think it costs several hundred dollars per year, if one doesn't have access through 3rd party channels.
23:08<gconnor>I think I can find out relatively easily
23:08<grumpy>I think I can deal with $2-500/yr...
23:08<grumpy>I think we have a volunteer....
23:08<grumpy>thanks gconnor!
23:08<Julian>:)
23:08<MarkK>appreciated
23:09<grumpy>(and we wonder why freeside keeps quiet... ;-)
23:09<grumpy>Ok, anything else about the new chair?
23:09<grumpy>if so, type ... quickly
23:09<grumpy>next item: 7) how to proceed with regard to MS and SID?
23:10<Julian>Now it gets real interesting.
23:10<grumpy>I've thought about this subject a lot over the last week or two.
23:10<grumpy>my answer is:
23:10<grumpy>I hav eno clue.
23:10<freeside>can we have a summary of your thinking?
23:10<grumpy>I think that MS *was* very even handed in their handling of mfrom vs pra.
23:11<grumpy>not just at the Email Auth Summit, but also with their updated wizard.
23:11<grumpy>I think that the re-use of SPFv1 records is abusive and opt-out is from the evil world of spammers.
23:12<grumpy>that said, I think the SPF community would do much better by promoting the positive aspects of the mfrom and spf-classic rather than trying to fight MS and the PRA.
23:12<grumpy>I talked with several folks from MS at the Email Auth Summit. They are looking at things in the persepctive of the 80/20 rule.
23:13<Julian>Fighting the PRA and fighting the abuse of v=spf1 for the purposes of the PRA are two different things.
23:13<Julian>I don't think we need to do the first.
23:13<grumpy>They want to tackle the 80% of the problem first, and then deal with the problems in the remaining 20%.
23:13<grumpy>At that level, the damage from re-use becomes irrelevant.
23:13<Julian>So their current plan seems to be to ignore the PRA problems, of which they are mostly inaware in the first place anyway.
23:13<grumpy>I strongly disagree with that approach, but as long as MS has that mindset, it will be almost impossible to convince them to change.
23:14<Julian>I would very much like to see a compromise with MS, but I agree to your assessment that this is not going to happen. They're just too ignorant.
23:14<grumpy>Julian: Well, say ELNK announced tomorrow that they were going to use v=spf1 records for checking against the References: header.
23:15<grumpy>There isn't anything we can do to stop them, other than make it very clear that doing so is a Really Dumb Idea.
23:15<MarkK>like I saod earlier, a 20% loss/false positives in email is a staggering, totally unacceptable margin of error
23:15<Julian>Then we contact them and tell them it is bullshit.
23:15<Julian>Yes, we just make things clear to everyone who wants to listen.
23:15<MarkK>even 1% is probably too much
23:16<grumpy>MarkK: I think 1% is way too much. The current goal is 0.1%, and many people want far less than that.
23:16<grumpy>Ok, that's my view.
23:16<Julian>Well, the forwarding-not-done-right problem causes more than 1% false positives already. But we don't need to double that.
23:16<grumpy>Julian: agreed.
23:17<Julian>So, do we want to take any further steps toward a compromise with MS?
23:17<Julian>Does it make sense to contact them any further?
23:17<Julian>Any compromises would have to be reached ASAP.
23:18<grumpy>I think it makes sense to talk with MS and tell them concisely what our technical objections are
23:18<MarkK>julian: yes; and I agree with grumpy, earlier today, that we are NOT going to play the delay game
23:19<grumpy>and to make sure it is clear to them that we aren't doing this just to bash MS, but rather because we are doing The RIght THing.
23:19<gconnor>I'd like to hear from freeside as well on this...
23:19<grumpy>yes
23:19<MarkK>If the door is open, fine; but I do not feel like dragging this out forever with them
23:19<grumpy>and I would like to hear from gconnor and MarkK too.
23:19<MarkK>grumpy: you just did :)
23:19<grumpy>MarkK I don't think they want to drag it out also.
23:19<grumpy>freeside?
23:20<grumpy>gconnor?
23:20<Julian>I'm not even sure that MS is currently trying to play some sort of delaying game. From what I gather, they just don't have a clue about any of the procedures. They're not even aware of why they're promoting the PRA over MAIL FROM.
23:20<Julian>(except for the patent license, of course.)
23:20<gconnor>freeside- you said that there is potential benefit to having MS actively promote v=spf1 records. (when you get back to the keyboard) please elaborate?
23:20<freeside>the benefit is, well, they tell peopel to do the sam ething we've been telling people to do =-- publish spf1 records
23:20<gconnor>while waiting on freeside, I'll say what I'm thinking so far.
23:20<gconnor>(never mind, go ahead)
23:21<freeside>and we can still control a lot of the messaging on that, it's not like it's 100% PRA and no mail from.
23:21<freeside>and i don't think there's any point trying to convince somebody else that what they're doing is wrong.
23:21<freeside>too much of that effort is one of the hallmarks ofa kook.
23:21<MarkK>grumpy: I did not gather from your respective reports (meng and you), that MS has much of a clue, or is even interested in getting one. If they did not even bother to test MAIL FROM, then it does not seem they even really grasp the issue -- strange as that may sound.
23:22<freeside>i trust that MS has its reasonf or doing what it's doing.
23:22<Julian>I think MS promoting v=spf1 is a red herring. If they do it, this means promoting PRA checking against v=spf1. Any other option that is useful for either them or us (like the compromise mentioned above) does not involve the promotion of v=spf1.
23:22<freeside>and the way competition works is, you don't tell your competitors what to do. you wait for them to do the wrong thing and when they screw up you say, sorry too bad.
23:23<grumpy>MarkK: I think you are right.
23:23<Julian>Why is this a competition between MS and us? I don't get it. This is not a battle.
23:24<grumpy>agreed
23:24<gconnor>How much value do you think is potentially there? Do you think the gain in sheer numbers of v=spf1 records would outweigh the fraction that have conflicts between mfrom usage and pra/2822 usage?
23:24<grumpy>and I said something very similar to that to Craig... The battle is with the spammers/phishers, not between us.
23:24<Julian>It's not as if we were trying to get v=spf-the-real-one adopted over v=spf1-the-one-from-ms.
23:24<gconnor>I don't think it should be a competition. I very much like the "lots of tools in the box" approach rather than an "us vs. them" approach.
23:25<grumpy>I think that some tools are far more useful than others. I personally don't see much use for the PRA.
23:26<Julian>The thing is that MS (and Ted Hardie) don't accept the fact that v=spf1 was there before they decided to use it.
23:26<grumpy>yep, and I don't think there is anything we can do to make them accept that fact.
23:26<gconnor>I would agree with that grumpy, but my vision of "many tools in the box" allows for there to be bad tools :)
23:27<gconnor>freeside- did you see the previous question?
23:27<Julian>grumpy: Problably. But that doesn't mean that we have to accept the "fact" that they are trying to redefine history.
23:27<MarkK>Julian: but the IETF should have cared about that, and have asked MS not to use v=spf1 records defined by a 'competing' spec
23:27<Julian>gconnor: v=spf1 is our only tool. If we allow it to become bad, we're dead.
23:28<Julian>So, do we want to take any further steps toward a compromise with MS?
23:28<MarkK>(Julian: in reply to your "[01:26] <Julian> The thing is that MS (and Ted Hardie) don't accept the fact that v=spf1 was there before they decided to use it.")
23:29<MarkK>Julian: if they are serious, yes; otherwise we are wasting our time with them
23:29<grumpy>I say, we make sure we tell MS what the technical objections that we have with re-use are, and we politely inform them that we intend to make sure people are aware of the advantages to doing mfrom checking.
23:29<Julian>I can't really judge if they are serious. At the moment, Wayne has the most accurate impression, I guess.
23:30<Julian>grumpy: Your suggestion doesn't involve trying to get an agreement with MS, right?
23:30<MarkK>MS's strategy is clea, imho. Of course they care about those 20%. But they think like a business: "First aquire 80% of the market; then we can set the other 20% to our hand."
23:30<gconnor>I have a couple more thoughts but I was waiting for freeside to answer the question.
23:30<freeside>you forgot the "muhahaha" bit.
23:30<freeside>what question did i miss?
23:30<grumpy>Julian: no, not really...
23:31<gconnor>How much value do you think is potentially there? Do you think the gain in sheer numbers of v=spf1 records would outweigh the fraction that have conflicts between mfrom usage and pra/2822 usage?
23:31<freeside>yes, i think so.
23:31<Julian>I know freeside's answer to that.
23:31<Julian>Duh. :-)
23:31<grumpy>:-)
23:31<freeside>the fraction of friction ...
23:31<gconnor>the "how much" part should be a number?
23:31<grumpy>gconnor is new here... ;-)
23:32<freeside>it may be an irrational number, but it's still a number :)
23:33<grumpy>gconnor: your thoughts?
23:33<gconnor>ok.. hang on
23:33<gconnor>The "opt out is bad" argument doesn't hold well with me. Spam is about consent, not content, but the analogy doesn't relate well to technologies. A new tool that operates on old equipment is perfectly acceptable. If it becomes a question of "how bad is it" then it would have to be horribly bad, AND sell really well, to damage the prior operator's reputation. I don't forsee BOTH happening.
23:33<gconnor>I think there are cases where using v=spf1 records in PRA context will fail, but I disagree that this will make SPF look bad. I don't think the damage done to SPF's reputation, if any, will be enough to harm SPF in a noticeable way.
23:33<freeside>damn, gconnor types fast.
23:33<gconnor>I don't agree that v=spf1 is our only egg in the basket right now. We have come a long way... MS is using SPF language instead of XML, for one thing. I think "embrace and extend" can work both ways.
23:34<Julian>grumpy is talking about consent by domain owners who published v=spf1 before the idea of applying PRA against v=spf1 came up.
23:34<MarkK>like me, lol :)
23:34<gconnor>I very much like the "lots of tools in the box" approach rather than an "us vs. them" approach. Our actions should be consistent with that.
23:35<gconnor>Right. My point is that we deal with spam every day, but that doesn't mean that "consent" is the same issue in terms of tools.
23:36<gconnor>At least, I'm not going to use consent as my only determining factor.
23:36<freeside>i think those domain owners, being early adopters, can be counted on to change if needed. and i don't think they'll be needed to change anyway.
23:36<grumpy>gconnor: ok, well I disagree with the consent thing. requiring people to opt-out when there is a choice is almost never acceptable.
23:36<Julian>If I follow gconnor's argument, there isn't any need _whatsoever_ to make an agreement with Microsoft.
23:36<gconnor>I don't see a need, unless we have a common goal.
23:36<Julian>So, can we just ignore MS, and go ahead with appealing to the IETF and the IAB?
23:37<gconnor>Uh, that's not what I meant :)
23:37<Julian>I say we do an appeal to the IETF/IAB. I have never done that. It should be fun.
23:37<MarkK>Julian: if we ignore MS, then yes, that would be our route
23:37<gconnor>One final thought...
23:38<Julian>"We have come a long way... MS is using SPF language instead of XML, for one thing." -- That must be a joke, right? We haven't won anything we didn't already have.
23:38<gconnor>as an admin, and potential customer of both technologies, I would *want* to be able to publish one record and have it apply to all identites. I want the option of varying them, but I think for most environments they will be the same. I think we should be pushing for a single record, whether it be v=spf1 or spf3.0
23:39<Julian>I mean, I would have preferred MS using XML. That way, their proposal would have been dead in the tracks and v=spf1 would never have been abused.
23:39<gconnor>No, that was a reference to Marid. Several of us worked hard to get MS to drop xml and go with SPF syntax. That's part of the reason I don't want to beat them for it now.
23:40<Julian>I know what happened during MARID.
23:40<freeside>mmm.
23:40<Julian>It's not as if a million domains just published because MS started promoting Sender-ID.
23:40*freeside agrees with gconnor.
23:40<grumpy>i think we (MARID participants) worked hard to get the MARID proposals tonot have to support both spf and xml syntax.
23:41<freeside>i think a lot of those million domains publihsing had to do with sender id.
23:41<MarkK>gconner: they 'conceded' to not using xml, just so they could facilitate their quick 'embrace and extend' policy. I hardly call that a courtesy on their end
23:41<Julian>freeside: Define "a lot". 10%?
23:41<grumpy>freeside: I disagree. I think a huge percentage is due to AOL requiring it.
23:41<gconnor>I participated in MARID because I believe SPF cannot succeed on its own. If there is a proposal that allows me to do Mfrom and other identites, and another that just does mfrom/helo and nothing else, I will probably pick the first one
23:42<Julian>So, do we want to take any further steps toward a compromise with MS?
23:42<gconnor>What compromise would be best?
23:42<MarkK>gconner: I think probably all of us want to check different scopes and techniques. Fine. What we, or I at least, do specifically NOT want, is MS to check the wrong scope with the wrong record(type).
23:44<grumpy>fyi; my surveyof all .com/.net/.org domains as of last Aug showed 650k publish SPF records. That was before MS was really pushing SPF records.
23:44<MarkK>gconner: really, I am all for checking other identities; more power to them. :) But not using v=spf1 records to check something these records were not published for
23:44<Julian>gconnor: Any compromise that keeps MS from re-using v=spf1 for PRA checking.
23:44<gconnor>Mark- I understand what you mean. I don't think any of us (aside from maybe Julian) are talking in absolutes here... it's a matter of degree. I agree that there are cases where it will fail, but I think that issue is much less important than leaving the door open to possible benefits Freeside described.
23:44<freeside>markk: then publish spf2.0/pra ?all, then they won't check spf1.
23:45<MarkK>freeside: that is yet to be seen
23:45<gconnor>freeside is right. Some of us may have a moral objection to "opt out" as an operating mode, but the opt-out option is there.
23:46<Julian>So, now that Chuck is gone, _I_ am the one talking in absolutes? Hmm.
23:46<grumpy>heh
23:46<freeside>it's a tough job, but somebody has to do it
23:46<MarkK>I should gather real data on whether they keep the opt-out bargain (which is still an ugly option)
23:46<gconnor>sorry , no offense meant :) YOu did say "never" at least once :)
23:47<grumpy>Well, as far as counting, I am part of the "some of us" who find out-out to be just plan wrong.
23:47<Julian>Well, MS is saying "never" all the time, they just use other words.
23:48*grumpy looks to see how to move the discussion forward.
23:48<gconnor>We don't have a specific motion, do we?
23:48<Julian>For a change, we could answer this question: So, do we want to take any further steps toward a compromise with MS?
23:48<grumpy>Ok, let me ask each of you to answer the question: How should we proceed? Answer in on line.
23:49<grumpy>(I hope people are typing...)
23:50<Julian>My answer: There doesn't seem to be a real chance at getting a compromise with MS. Ergo we should go ahead and appeal to the IETF, and to the IAB if necessary.
23:50<MarkK>julian: I think your questuon moves us to the belayed item #8 on the agenda
23:50<grumpy>I say, we make sure we tell MS what the technical objections that we have with re-use are, and we politely inform them that we intend to make sure people are aware of the advantages to doing mfrom checking.
23:51<Julian>MarkK: #7 asks "should we keep trying to get a compromise with MS?". #8 asks "if not, should we make an appeal to the IETF/IAB?".
23:51<grumpy>MarkK? freeside? gconnor?
23:51<gconnor>One possibility might be to drop them a line, or call on the phone, and say "Hey, there is significant objection to using TXT records meant for SPF in PRA context. We are considering appealing the IETF proposal you submitted" and see if they respond, without giving a specific demand for what they should do.
23:51<freeside>we've already told MS what our objections are, and we've already informed them that we intend to keep going with mfrom.
23:51<freeside>i don't think there's any point appealing the IETF proposal.
23:51<grumpy>gconnor: that is basically what I did at the Email Auth Summit.
23:52<gconnor>grumpy, regarding that...
23:52<grumpy>freeside: my point of sending email to MS is more of a courtesy and to make sure they understand our motives.
23:52<Julian>going with mfrom != agreeing to re-use of v=spf1 for PRA.
23:52<gconnor>did you get the feeling that the door might be open to further discussion?
23:52<Julian>Well, appealing the IETF proposal gives the IETF a very good chance to once again ridiculing itself.
23:52<grumpy>I got the impression that the door was open for some limited discussions on some subjects.
23:53<Julian>That's a lot of "some" in one sentence.
23:53<MarkK>freeside: did you happen to get a chance to introduce wayne to those few MS folks you spoke of earlier?
23:53<grumpy>freeside: so, your point is that it wouldn't be productive to even contact MS, and that even a courtesy email would probably do more harm than good?
23:54<MarkK>freeside: in the sense that a phone conversation could be had
23:54<grumpy>Julian: agreed. They don't want to re-open every subject that was covered during MARID. If there is something that we can work together on, however, they would likely be interested in talking.
23:55<grumpy>For example, pointing out things like the broken expedia record is something they would likely be quite willing to help fix.
23:55<grumpy>and they would be willing to have an open dialog about such issues.
23:55<Julian>Ok, assuming we sent them a message offering a compromise, can it realisitcally be expected that they give a solid reply within 10 days?
23:55<grumpy>they *might* also be interested in some inter-op testing and such.
23:55<Julian>s/message/phone call or whatever/
23:56<Julian>The expedia record is completely off-topic (and irrelevant) right now.
23:56<grumpy>no, it isn't. It is something that we can productively talk with MS about.
23:57<Julian>(If they want to show their own ignorance by doing recursive inclusions, who am I to stand in their way?)
23:57<grumpy>and just *having* adialog is going to be important if we ever want to get them to change.
23:57<Julian>Quote from the agenda:
23:57<Julian>7. How to proceed with regard to Microsoft, Sender-ID, and PRA?
23:57<freeside>markk: yes, i introduced grumpy to craig spiezle and harry katz, and we had a productive conversation with them. i think.
23:57<Julian>So you think we should feed them non-issues in order to get them into a dialog in the first place?
23:58<grumpy>Julian: I think that may be needed, yes.
23:58<grumpy>I think we are too late to get the to change on the re-use issue.
23:58<gconnor>My preference would be to do nothing... but if it comes to a choice of trying to contact them directly and making an appeal to ietf, I would prefer the informal contact first.
23:58<grumpy>At least not in the near term.
23:59<Julian>"I think that may be needed, yes" -- I hope you realize that this would be a first-class delaying game?
23:59<grumpy>it would only be delaying if we stopped doing stuff.
23:59<grumpy>and I don't intend to stop promoting the advantages of mfrom.
23:59<grumpy>far from it.
23:59<Julian>No, it would be delaying as soon as it causes us to miss the appeal deadline.

This report was generated at Mon Jul 18 23:59:59 UTC 2005.