2015-03-13 UTC
# 00:00 acegiak chromium has a separate process for each tab and iframe and it enrages me
dariusdunlap and KevinMarks_ joined the channel
# 00:23 gRegor` acegiak: Chrome's task manager / "stats for nerds" might be what you want, though not as convenient as 'top'
# 00:24 acegiak gRegor`: Stats for nerds doesn't display in conky :P
scor and pwcc joined the channel
# 00:36 pwcc Perfectly possible I've broken something
# 00:37 pwcc In the dashboard, the text is showing correctly.
KevinMarks_, electronicmaji and wolftune joined the channel
# 00:52 acegiak pwcc which version of the plugin are you running?
# 00:54 pwcc acegiak: Webmentions 2.4.0-beta || Webmentions for comments 1.0.0 || Semantic linkbacks 3.0.5
# 00:54 GWG pwcc: You might want to go to stable
# 00:55 pwcc I'm poking around comment_text_add_cite in semantic linkbacks to see if i can figure it out,
# 00:56 GWG pfefferle tested with 4.2 beta, where a patch has been deployed. We're both running, I assume, 4.1.1
# 00:57 GWG When a webmention is noted as a reply, Semantic Linkbacks changes the comment type from webmention to "" . A blank type is a comment.
# 00:58 GWG He used to do it at the database level
# 01:01 pwcc GWG: I think I see. So should I comment out the condition starting on L131?
# 01:05 GWG The solution is for pfefferle to either revert till 4.2 is released or put in a check
# 01:07 pwcc Speaking of which, 4.2 beta 1 was released a couple of hours ago.
# 01:08 GWG pwcc: Beta testing is an alternative solution
# 01:10 GWG pwcc: Or manually revert this change..
# 01:10 pwcc GWG: appreciate the help - knowing what's going on is 90% of it
# 01:11 pwcc GWG: yeah, just saw that - might hack it when I relaunch this weekend. Eitehr WP core or the plugin. or both
# 01:11 GWG global $wp_version; if (version_compare($wp_version, '4.2', '>=')) { // version is 2.7 or higher }
# 01:11 GWG Just put a version check in there. I might do it and submit a PR
# 01:15 pwcc thinks wistfully that hacking WP core would be more fun but knows he will do it the more robust way
benwerd_, tantek, scor and tantek_ joined the channel
# 01:22 pwcc GWG: just changed the comment types in the DB - worked a charm - thank you
electronicmaji and mdik joined the channel
# 01:57 tantek benwerd: "ship it or shut it" is growing on me
# 01:57 Loqi tantek: snarfed left you a message 1 day, 7 hours ago: happy birthday tantek!
# 01:57 Loqi tantek: aaronpk left you a message 1 day, 6 hours ago: ah thanks. she hasn't imported those yet (they had to be done manually) and I think you're getting redirected based on fancy wordpress slug matching magic
# 01:58 tantek !tell snarfed thank you for the birthday wishes!
# 01:58 Loqi Ok, I'll tell them that when I see them next
snarfed joined the channel
# 01:58 tantek !tell aaronpk can you bring packs of IWC stickers to Cambridge? I'm near out!
# 01:58 Loqi Ok, I'll tell them that when I see them next
snarfed joined the channel
# 02:21 tantek deliberately authored that way because I know my auto-ellipser will be smart enough to break at the "Continued: " and put a permalink to the original
# 02:22 tantek but now I'm considering updating my original quotation post to reconstitute the quote into a single block, as I would have authored it if POSSEing were no consideration.
# 02:23 tantek and then the question becomes, is there some way for my POSSE code to *automatically* shorten a quote like that, with a "Continued: " for the sake of higher fidelity POSSEing (rather than just dumb-ish truncation/ellipsing in the middle of the longer quote)
# 02:25 KevinMarks I did a truncator like that before but peoepl don't always realise that there is more without he ellipsis cue
# 02:26 KevinMarks the way to mark it up is with an h-summary around the first bit
# 02:26 pwcc tantek: I'd refactor the post. I've set up my site so I can post different text to the silo if needs be.
# 02:27 tantek the point here is to make it *less* work for the author
# 02:28 tantek I'm thinking of trying to automate a quotation POSSEier for Twitter - e.g. having it automatically excerpt from *inside* what's being quoted, leaving enough room for 1. attribution, 2. hashtag(s), 3. a "Continued:" link to the original quotation post with the whole thing
# 02:29 pwcc tantek: Y. If I can use the same text I do, which is most of the time.
# 02:29 tantek KevinMarks - my CASSIS code knows to count images as links
# 02:30 tantek KevinMarks: aaronpk somewhat does this with his POSSE notes - he puts hashtags automatically at the end of the POSSE copy, and counts them first when ellipsing the rest of the content that precedes the hashtags
# 02:31 tantek "Please enable JavaScript to view the comments powered by Disqus." :)
# 02:32 tantek KevinMarks: yeah, web apps manifest overkill - though better than a JS lib hack
# 02:33 KevinMarks interesting social engineering - they only show it with https and nudge towards offline
# 02:33 tantek that's not social engineering, that's just good incentivizing (is that a word?)
# 02:34 tantek also only show it with https is good *user-centric* "trust" design
# 02:38 tantek I should also extract a bunch of my auto-POSSEing code into small CASSIS functions
# 02:47 GWG acegiak: I thought we both fixed it
# 02:47 GWG acegiak: I have more enhancements planned though
snarfed and wolftune joined the channel
# 02:52 GWG That should do until I make the whole thing a for loop
# 02:53 benwerd_ I didn't get a notification, anyway - what are you webmentioning?
# 02:54 GWG acegiak: What we need is a button to send a webmention
benwerd joined the channel
# 02:57 benwerd Sorry, vanished too early. Got it (and great reply) - but no notification at my end.
# 02:57 benwerd I need to go offline but I can investigate later.
danlyke_ joined the channel
electronicmaji joined the channel
KevinMarks_ joined the channel
wolftune and j12t joined the channel
tantek and colintedford joined the channel
KartikPrabhu, KevinMarks_, squeakytoy and loic_m joined the channel
KartikPrabhu joined the channel
sanduhrs, Jihaisse, KartikPrabhu, KevinMarks__, KevinMarks_, krendil, alanpearce_, interactivist, friedcell and dns53 joined the channel
Sebastien-L, Jihaisse1 and elf-pavlik joined the channel
petermolnar joined the channel
Pierre-O, adactio, nloadholtes, KevinMarks_, scor, interactivist, loic_m, stream7, uhsf, friedcell and csarven joined the channel
frzn and elf-pavlik joined the channel
scor joined the channel
# 12:04 elf-pavlik ben_thatmustbeme, ^
stream7 joined the channel
# 12:17 ben_thatmustbeme basically they believe the easiest, and sanest way to do this is to just use chanserv to give voice to those you wish to display authenticity
# 12:25 GWG petermolnar: Were you looking for me?
# 12:27 petermolnar but I'm still working on some more well-known silo implementations
# 12:28 petermolnar dfq I read "Were you looking for me" as "were you working for me"?
KevinMarks_ joined the channel
# 12:34 GWG petermolnar: My Internet has been down for nearly 9 hours. I'll see if I can have a look when it comes back.
davemenninger joined the channel
# 12:35 petermolnar short story: X worker for a silo collects all the relevant posts - relevant is determined it there is a matching line entry in syndicated_urls meta field
# 12:36 GWG I know. Just didn't want to deal with a net outage on my day off. I find it hard to look at code on my phone
# 12:36 petermolnar after this either from wp cron or manually, in batches, it imports every response (silo specific, comments, favs, etc) as comments as matching comment type ( favorite, comment, etc )
stream7_ joined the channel
friedcell, joskar, wolftune, KevinMarks_, krendil, KartikPrabhu and bupkes joined the channel
# 14:04 aaronpk elf-pavlik: no he does not, in practice we haven't had a need for that
joskar, alanpearce, snarfed, cweiske, verdi_, gRegor` and KevinMarks_ joined the channel
# 14:53 aaronpk i started working on monocle this morning. i'm hoping to make a bunch of progress on the plane this afternoon
KevinMarks_ and joskar joined the channel
# 14:56 cweiske fkooman - you said the token parameter is different when using indieauth.com as auth server. is that really so?
marclaporte, wont_on, wont_on_, ryukafalz, joskar_ and alanpearce joined the channel
joskar_ joined the channel
# 15:14 cweiske so the parameter used by indieauth.com is the legacy one and will be the future one because of BC?
# 15:16 aaronpk If you include a client id like you should then you'll always get a code
# 15:17 aaronpk If you get "token" it's because you're using IndieAuth.com as a shortcut for development, not becAuse you're using it as a user delegating to it
marclaporte joined the channel
# 15:18 aaronpk IndieAuth.com/developers describes using the server from the POV of a developer
# 15:19 aaronpk I wonder how I can better clarify the two roles IndieAuth.com plays
KevinMarks_ joined the channel
KevinMarks_ joined the channel
# 15:29 aaronpk Maybe authorization vs authentication but I think a lot of people glaze over when they see that
KartikPrabhu joined the channel
KevinMarks_ joined the channel
KevinMarks_, joskar, joskar_, friedcell, interactivist, dariusdunlap and pdurbin joined the channel
# 17:01 fkooman aaronpk: maybe a complete spec document for (distributed) IndieAuth would help a bit at least :-)
# 17:04 ben_thatmustbeme i do want to work out a way of direct server-server auth and comms that doesn't open things up to the security issues fkooman pointed out
# 17:09 fkooman ben_thatmustbeme: that was mostly the CSRF right?
friedcell joined the channel
# 17:11 aaronpk what does CSRF have to do with server-to-server auth?
# 17:11 fkooman hmm what is server-to-server auth?
joskar_ and gmack joined the channel
# 17:13 aaronpk yes maybe we should just hold off on this until we solidify the first bit
# 17:14 aaronpk i'm pretty sure that if you trust your authorization server (which you obviously do since you delegated to it) then you could trust it to generate auth codes on your behalf even if you are not in front of the keyboard, which can be used for server-server auth
# 17:14 fkooman oh with server-to-server auth you mean authorization
# 17:14 aaronpk actually now i don't remember. ben_thatmustbeme what was the server-server case again?
friedcell, michielbdejong and j12t joined the channel
# 17:56 ben_thatmustbeme aaronpk, we were creating auth tokens between our servers that would auth without any interaction to allow fetching of private messages.
# 18:01 ben_thatmustbeme so i don't store that the regquest to log in to my site actually came from my site, only that the call back gives me a code that will validate the whoever is in the 'me' param.
# 18:02 ben_thatmustbeme which was what allowed you to just generate a code from your auth provider, send it to my site and thereby get me an auth token to pull data from your site
benwerd joined the channel
# 18:13 GWG My Internet has been down, so I took a walk to somewhere with wifi. So much for writing code..
# 18:14 GWG I was going to see how close I could get to a check in or rsvp before next week
# 18:14 aaronpk I'm gonna be the last one to publish Checkins aren't i
# 18:16 aaronpk For Monocle. Or for any site that wants login with optional micropub
# 18:17 GWG aaronpk: I don't go anywhere. You seem to
alanpearce joined the channel
joskar joined the channel
# 18:21 aaronpk When it makes the request for an access token the response includes the list of scopes granted
# 18:22 aaronpk Because the other thing that might happen is you grant it additional or custom scopes that it doesn't know about but that your site requires
KartikPrabhu joined the channel
# 18:22 aaronpk I'm planning on adding an interface to IndieAuth.com, which will let you un-check requested scopes or add new ones
joskar_ joined the channel
# 18:29 aaronpk That's something not a lot of OAuth servers support right now, but more should
frzn joined the channel
# 18:41 GWG I'm trying to figure out, if I can just decide that a WordPress page with geo data is a venue. That seems easier than creating a new type
michielbdejong1 joined the channel
# 18:49 GWG Isn't that the right command? I forget.
# 18:52 GWG I thought it was supposed to indicate the last time someone was seen
# 18:54 kylewm I haven't seen Loqi do that particular trick. that'd be a nice one though
# 18:57 GWG Kylewm, I guess I was mixing up things
# 18:57 GWG Kylewm, do you have a link to a good example of a venue?
# 18:58 kylewm GWG: mine don't have microformats yet, is that what you're looking for?
# 19:15 bret i got a firefox os flame phone yesterday.
alanpearce, wont_on, wont_on_, KevinMarks__, KevinMarks_ and tantek joined the channel
KartikPrabhu joined the channel
joskar and gRegor` joined the channel
tantek joined the channel
Gu_______, joskar_ and KevinMarks_ joined the channel
Gues_____, wont_on and friedcell joined the channel
joskar_ joined the channel
# 20:57 Loqi Got it! There are now 19 spammers blacklisted
j12t and KevinMarks_ joined the channel
benwerd joined the channel
# 21:01 kylewm tantek: aaronpk: were you guys going to show something with Woodwind at the f2f? Is there a particular day I should be careful not to break it :)
# 21:01 tantek kylewm I would except I don't have micropub support on my site :(
tilgovi joined the channel
# 21:02 tantek but yes - someone *with* micropub support on their site should demo woodwind as a client!
# 21:02 kylewm I thought it was discussed for realtime pubsubhubbub updates
# 21:02 tantek I have a feeling that might come up during Federation discussions
benwerd joined the channel
# 21:07 aaronpk Made some more progress on Monocle on the plane :-)
snarfed and dariusdunlap joined the channel
# 21:09 aaronpk Hey so I was thinking about the problem of syndicating to IndieNews
# 21:09 aaronpk Since obviously pre-constructing the IndieNews url has been challenging
# 21:10 aaronpk And it struck me that it's actually more similar to a regular webmention than I originally realized
# 21:10 aaronpk If you send me a home page webmention your text in the link is probably my name
# 21:11 bear kylewm - oh! you made it a Flask plugin!
# 21:11 aaronpk My site *could* send a reply to your webmention saying "here is the url of your comment"
# 21:11 aaronpk At which point this is exactly what is needed for IndieNews
# 21:11 aaronpk You'd send a webmention to news.indiewebcamp.com and it would create the post
# 21:12 aaronpk Then in the response could be the IndieNews url and you could optionally update your syndication URL accordingly
# 21:13 aaronpk And then submitting things to IndieNews becomes really easy since you just have to mention news.indiewebcamp.com in your post
# 21:13 bear kylewm++ nice work - I'll compare it to the manual stuff I was doing and then switch over to it
# 21:14 kylewm bear: that'd be awesome, let me know if anything is noticeably missing
# 21:14 bear that is handed by flask forms - you shouldn't have to do anything special
# 21:14 bear (well, other than init the forms properly)
# 21:14 bear but yea, I'll look at that this weekend
# 21:16 ben_thatmustbeme i only have a few months (at most) to really get this worked out, I need to make a micropub client that works on my ios and android that can track diaper changes for 2 kids.... figure my girls will have every piece of data we collect saved to a private site for them
# 21:17 kylewm the micropub servers section says it is sorted by deployment date,but it is actually sorted alphabetically (except for adactio)
# 21:18 kylewm wait, maybe just one of them is out of date order
# 21:18 bear yea, just *using* flask-wtf you get CSRF
Gues_____ joined the channel
# 21:28 tantek from analyzing the API Candidates page, nothing other than micropub has anyone in the WG running anything on their own personal client/server site
# 21:29 tantek except I know from inference that Evan's site supports pump.io
# 21:29 aaronpk hmm should I add Monocle even though it's not finished yet? ;-)
# 21:29 Loqi tantek meant to say: aaronpk, what site or service is "finished"? ;)
# 21:30 aaronpk Are personal websites not a priority for the Social WG?
# 21:30 aaronpk It seems like everyone is so focused on corporate social networks
# 21:33 Gues_____ @tantek I'm just reading http://indiewebcamp.com/Micropub for the first time (ima noob). do you think that the accepted Social API solution/candidate should specify how to request/respond with JSON and not just formencoded? I do.
# 21:34 tantek Gues - no. plumbing based requirements are out of scope.
# 21:34 tantek bengo: no. plumbing based requirements are out of scope. This is why we have documented user stories for API Candidates to support, not acronyms.
# 21:35 bengo I see. That makes sense for consensus building.
# 21:35 bengo Isn't a deliverable of the WG a Social API specification that I as an implementor could follow pretty specifically
# 21:37 bengo When I've seen "Social API", I always read 'web service api'
# 21:37 bengo not 'client-side API that lets developers embed and format third party information such as social status updates inside Web applications'
alanpearce joined the channel
# 21:40 tantek micropub can and is used client-server or server-server
# 21:41 tantek then again I don't understand most of those API candidates anyway as they don't seem to be related to any citable real world uses of anything - very abstract
# 21:41 tantek some of them appear to border on architecture astronomy
# 21:49 kylewm maybe just me but that description of the Social API deliverable is confusing
# 21:49 kylewm embed and format third party information inside web applications
# 21:50 tantek we had a bunch of follow-up clarification at the F2F last fall at TPAC / Santa Clara
# 21:50 tantek about what should a social API generally cover
# 21:51 tantek then went (premature) to a long list of specific requirements (over engineered)
# 21:51 tantek then backed off and proposed user stories instad of abstract API requirements
# 21:51 kylewm bengo: so micropub is something you could as an implementor could follow pretty specifically right now
# 21:54 bengo @kylewm I agree! I'll take a stab at it soon. Was just ruminating on whether there's some work to do to groom it further for the WG, where it seems like because the WG wants the candidate API to speak ActivityStreams2, there should be a AS/JSON-flavored spec-extension for MicroPub
# 21:54 kylewm ok, why are people talking about CERN scientific data in the social wg?
# 21:56 bengo @tantek nice link to the voted user stories. I'll peruse that.
KevinMarks_ joined the channel
# 22:03 tantek bengo - it's not clear that a Social API has to speak AS2
# 22:03 tantek note that none of the user-stories require AS2
# 22:04 tantek it may be that AS2 needs to be simplified to handle Social API requests
# 22:04 tantek AS2 is a WD. Everything in a WD is subject to change.
# 22:06 tantek to note how those requirements have been superseded by the user-stories
# 22:09 tantek bengo thanks for the heads-up about that out of date wiki page
# 22:13 tantek bengo - really appreciate the close "new eyes" review of the wiki pages
# 22:13 bengo :) Then I won't hold back as I spend more time around here
# 22:13 tantek bengo, asking good questions is *really* appreciated by the community here
# 22:14 tantek we try to document the answers one way or another, good or bad
# 22:14 bear constant review by people who have no learned assumptions is the only way to keep the community new-person friendly
# 22:16 bear tantek++ because having the senior folks support new folks asking probing questions helps reinforce that as a good community behaviour
squeakytoy2 joined the channel
# 22:17 tantek ben_thatmustbeme: I believe the identity faking going on in socialwg IRC is perhaps a strong use-case for POSSEing to IRC with permashortlinks
# 22:18 bengo With a sufficiently complete indieweb, is IRC even necessary?
marclaporte joined the channel
# 22:19 bear bengo, IMO IRC will always be necessary until own-your-site reaches 100%
# 22:19 bear IRC is the stepping stone to a indieweb identity federated chat environment
# 22:21 bengo I didn't say anything was wrong with it, just that the web may compete with it
# 22:21 bear for me IRC stays until we get a curated feed federation with topic filtering
nloadholtes joined the channel
# 22:22 tantek bengo - we may very well end up with web-based replacmenets for IRC. The challenges fall into two big categories IMO.
# 22:23 tantek 1. design. UI of IRC clients is so simple and information dense that it is greatly superior for back/forth discussion as compared to anything even remotely similar, even Twitter @-replies
# 22:23 tantek 2. latency. typical IRC clients have apparently "instant" latency (far less than 200ms), and there are no web-based posting / reading solutions that have that (yet)
# 22:23 tantek those are the two big challenges if you want to replace IRC
# 22:24 tantek I've brought this up before but I'm not sure where to wiki it
# 22:24 pdurbin But can we back up and ask why we would replace IRC in the first place?
# 22:24 tantek ben_thatmustbeme: since you're actually working on replacing your use of IRC clients, are either of those itches for you?
# 22:24 tantek and where would you document them, indiewebwise?
# 22:24 bengo @pdurbin, there are a lot more web developers than irc developers
# 22:25 tantek pdurbin: it's the principle that everything you write/create you should own (ownyourdata), on your own domain, and anything in a "shared space" should be either peer-to-peer copies, or POSSE copies (e.g. to an IRC channel)
# 22:25 tantek pdurbing - no, it is like saying *an IRC server* is a silo, just like *a* website itself is a silo
# 22:25 bengo web developers don't want to learn protocols if they don't have to. If the Social WG also produces sufficient work to inspire chat servers, then web developers will have a choice of which protocol to learn. They'll choose the new one.
# 22:26 bear what I would love to see is an IRC server that allows for webmentions for my nick
# 22:26 bear then I can POSSE to IRC like other places
# 22:26 bear and would also have a feed endpoint for a channel
# 22:27 tantek pdurbin, imagine if your own website was your UI to IRC conversations
# 22:27 tantek and everything you said had its own permalink on your own server
# 22:28 bear right - that's why I was saying a feed api - so I could go to bear.im/topics/indieweb and see a stream from freenode/#indiewebcamp because that's how I curate that topic
# 22:28 ben_thatmustbeme Tantek. Somewhat. My interest was more in direct messaging. But I realized that chat goes hand in hand with that. My interests for now are to start making the tools to store and recall the necessary info then just obscure the pipe. I don't care if it's irc or done other thing at that point
KevinMarks_ joined the channel
# 22:29 pdurbin tantek: that's not at all how I use IRC but I'm trying to imagine it.
# 22:29 tantek pdurbin - right, it's also not at all how people in general today use the social web (which is by picking a silo first, then finding friends there)
# 22:30 tantek which is precisely the user model we are shifting with POSSE
# 22:30 tantek people using IRC today first go to a server, then a channel, then find people to talk to
# 22:30 pdurbin Right. I'm on IRC because that's where the open source people are. Especially the freenode silo.
# 22:31 tantek pdurbin - right, and in the indieweb perspective, that would be motivation to *POSSE* to the freenode silo
# 22:31 tantek same reason we choose to POSSE to silos - because our friends are there
# 22:31 tantek pdurbin yes it is potentially spammy if abused
# 22:31 tantek POSSE is also potentially spammy if done indiscriminately
# 22:32 tantek but that's why a key aspect of POSSE is the recommendation to *only* POSSE to destinations where you *know* friends read you
# 22:32 Loqi IRC is an abbreviation for Internet Relay Chat and is the primary discussion forum for the indiewebcamp community, in particular the #indiewebcamp channel on the irc.freenode.net server https://indiewebcamp.com/IRC
KevinMarks__ and marclaporte joined the channel
# 22:34 pdurbin tantek: if you can get it to work and not be spammy, more power to you.
# 22:36 bengo Good social servers will have good solutions for the 'spammy' problems
# 22:37 bengo e.g. restricting clients and using spam detection techniques
marclaporte and electronicmaji joined the channel
# 22:39 pdurbin Seeing how people here POSSE to Twitter makes sense to me. I'm just trying to decide how well I think it would work for IRC. I'm a little skeptical.
squeakytoy2 joined the channel
# 22:39 aaronpk we already see POSSE to IRC, with Loqi bringing in #indiewebcamp tweets
# 22:40 aaronpk the only time it's spammy is when it's actually spam
# 22:47 pdurbin I guess the way I think of it is that as long as my IRC logging bot is in the channel, I'm posting both to my own site ( http://irclog.greptilian.com ) and the channel itself. I feel like I'm done already. I own the data. :)
sammachin joined the channel
# 22:47 aaronpk pdurbin: that's true for you, but not for everyone else in this channel
# 22:48 pdurbin It's not even true for me in *this* channel. I don't log it. But I trust Loqi to log it. :)
# 22:49 tantek pdurbin: your bot logging from IRC is PESOS, not POSSE ;)
# 22:50 pdurbin tantek: for a split second I suppose it goes through a silo. Meh.
# 22:51 tantek pdurbin, hence people are practical about using Instagram -> OwnYourGram -> personal site.
# 22:51 tantek PESOS isn't all bad. It's just not as good as POSSE.
# 22:52 tantek the strongest argument I've heard for PESOS is using it for likes
# 22:53 aaronpk I don't have PESOS likes set up for instagram yet and I've found myself no longer checking instagram because I know that if I see something I'll want to like it and then it won't be on my site
# 22:53 aaronpk maayyybe that's a good thing to add to ownyourgram
# 22:54 pdurbin If an IRC channel isn't logged, you're aren't even publishing anything. So it isn't PESOS.
tantek joined the channel
# 23:03 tantek aaronpk: agreed on adding PESOS of likes (and heck, why not comments too) to ownyourgram
# 23:04 pdurbin ... and it's interesting/weird to think that the moment I send *my* IRC logging bot into a channel it's suddenly considered "publishing" and hence PESOS. Hmm.
# 23:06 tantek pdurbin: the moment there is someone else in the channel and you say something, you are "publishing elsewhere"
# 23:06 tantek when your bot enters and starts logging, then you are syndicating back to your own site.
# 23:06 pdurbin Really? If I send an MMS to a few friends' phones is that publishing?
# 23:07 tantek loosely speaking, that is one of the arguments that has been made
# 23:08 bear if you send anything to someone else it's publishing - even if the audience is one and private
# 23:09 snarfed delayed: re IG PESOS, you can do it for likes, but not really comments, since the API can't get comments by user
# 23:11 tantek bear, even if you send something only to yourself (e.g. create a document), you are publishing to your future self.
# 23:12 tantek let's see when I type something and press return.
# 23:13 Loqi KevinMarks meant to say: I now feel like filming it and fraemstepping
kyleweb joined the channel
# 23:13 tantek problem is - what I don't know is how long it takes for me to type something and have you all see it, vs. my client just "show it" as if it was already seen in the channel
# 23:14 tantek aaronpk - good point that's the round trip test
# 23:14 kyleweb trying the web ui
# 23:14 aaronpk the web chat at ?beta doesn't put it on the page itself, it waits for the regular streaming logs to show it on the page like all other messages
# 23:15 tantek right, that's what I meant by good round trip test
Kevin__Marks joined the channel
# 23:15 kyleweb I've noticed it takes Woodwind quite a bit longer to post a "realtime" update than the latency here
# 23:15 kyleweb that may be where PuSH fat pings would help
# 23:15 Kevin__Marks the delay is acceptable
# 23:16 kyleweb I'd say the delay here is fine but definitely noticeable
# 23:16 aaronpk there's been a ton of research into what delay is acceptable vs noticeable
# 23:16 kyleweb Doherty threshold is 400ms
# 23:17 Kevin__Marks can it be 300ms?
benwerd joined the channel
# 23:18 kyleweb I think the 400ms number is like... any longer and you might get distracted between button pushes :)
# 23:19 kyleweb sorry about that, is there a good way to shut it down without this nick staying in the channel?
# 23:20 aaronpk it responds to several disconnect events and I just can't find the event that is run that is making them linger
# 23:26 KevinMarks nice thing about modern phones is you can do quick and dirty latency testing
# 23:26 tantek another comparable metric, FB IM client vs. FB web view
# 23:27 KevinMarks Realtime is a well-defined software field where computing has to complete or fail cleanly by a deadline, because latency is paramount.
# 23:28 aaronpk tantek: btw the socialwg wiki skin on mobile has all the section headers collapsed by default so none of the fragment links worked while i was reading on my phone :(
# 23:29 tantek yeah that's the darn mediawiki default I think
# 23:29 tantek I wonder if you login to the wiki on your phone if you can change your preferred skin
# 23:29 tantek should I do something similar for the "discuss federation candidates" agenda item on day 2?
# 23:29 aaronpk the only thing i was curious about was the requiement of "personal" website
# 23:30 tantek I figured I'd put "personal website" in there as a think you should link your name to, as what you're using to demo, instead of a demo/test site
# 23:30 aaronpk i meant the first reference to it "showing User Stories on people's actual decentralized (personal) social web sites"
# 23:30 aaronpk yeah i'll be curious to see if anyone pushes back
# 23:31 tantek I want such pushback to have to be explicit so we can openly ask why, and document any exceptions
# 23:31 aaronpk corporate networks are totally a thing, so I suspect Ann will
# 23:32 aaronpk they don't have anything they can demo? I thought they had a lot of things working internally?
# 23:32 tantek KevinMarks: more JS in the wp skin could fix that
# 23:32 tantek aaronpk - they don't have anything they can demo that shows any of the API candidates
todrobbins joined the channel
alanpearce and benwerd joined the channel