- [00:25] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) joined #rest.
- [01:45] Split (~split@84.34.147.60) left irc: Quit: Changing server
- [02:45] Hakon|mbp (~hakon1@130.82-134-26.bkkb.no) joined #rest.
- [03:38] whartung (~whartung@wsip-98-189-78-118.oc.oc.cox.net) left irc: Ping timeout: 248 seconds
- [04:07] mhausenblas_ (~mhausenbl@wg1-nat.fwgal01.deri.ie) joined #rest.
- [04:09] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) left irc: Ping timeout: 240 seconds
- [04:09] mhausenblas_ -> mhausenblas
- [04:44] Blazeix_ -> Blazeix
- [04:45] pezra (~Adium@184-96-219-53.hlrn.qwest.net) joined #rest.
- [05:21] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) joined #rest.
- [05:26] pezra (~Adium@184-96-219-53.hlrn.qwest.net) left irc: Quit: Leaving.
- [05:29] <mk_> do you think we're allowed to claim that HAL has 'native hypermedia support' ?
- [05:29] <darrelmiller> Seems reasonable to me.
- [05:30] <darrelmiller> Steve and I were doing some major hal pimping at MS yesterday :-)
- [05:36] <mk_> sweet as a nut! :D
- [05:37] <mk_> darrelmiller: what's the story there then ? are they looking at having standard renderers built into the framework ?
- [05:38] <darrelmiller> No, completely different project and they have almost re-invented hal for another product.
- [05:38] <mk_> oh, terrific.
- [05:38] <mk_> that is very unlike them to do something like that.
- [05:38] <mk_> I'll take that as an inverted compliment.
- [05:39] <darrelmiller> Yeah, this group is way ahead. Stating goals like "evolving the client and server independently"
- [05:39] <darrelmiller> using link relations to convey semantics.
- [05:39] <mk_> very cool :)
- [05:39] <mk_> are they aware of HALs existence ?
- [05:39] <darrelmiller> how they managed to get to where they are without hearing about hal is surprising.
- [05:39] <mk_> :)
- [05:39] <darrelmiller> They are now.
- [05:39] <mk_> lol
- [05:40] <mk_> what was their feedback ?
- [05:40] <mk_> is this a private mailing list of some kind or were you actually in meetings with them?
- [05:40] <darrelmiller> There wasn't any during the meeting. They just took the link and said they would look.
- [05:40] <mk_> oh ok awesome
- [05:40] <darrelmiller> Was a Live meeting kinda think.
- [05:40] <darrelmiller> s/think/thing
- [05:40] <mk_> well that is pretty cool, I'll be interested to see what they built on top
- [05:41] <mk_> in terms of some tooling for documenting rels
- [05:41] <mk_> I'm still stacked with normal work atm so I can't work on much HAL related stuff atm which is a pain
- [05:41] <darrelmiller> Yeah, super top sekrit stuff where I had to sign my life away so that I could give them free consulting.... :-\
- [05:41] <mk_> I really want to do some proper examples of how I would document a rel
- [05:42] <darrelmiller> yeah. That would be very valuable.
- [05:42] <mk_> it takes me ages to do stuff like that though
- [05:42] <mk_> I can't write proper articles to save my life
- [05:42] <darrelmiller> I hear you. I was talking the web api mailing list about the Error response media type we were discussing a while back.
- [05:42] <mk_> so I think if I do it I will just do as mamund suggested and create a dummy app
- [05:43] <mk_> possibly a link relation registry
- [05:44] <mk_> :)
- [05:44] <darrelmiller> :-) How meta.
- [05:44] <mk_> right, I actually started a project with mhausenblas a while back to do that
- [05:44] <darrelmiller> I recall.
- [05:44] <mk_> but never got round to it, been stupid busy for a whlie now :|
- [05:45] <mk_> I barely have time to talk poop here/on the list
- [05:45] <mk_> sad times.
- [05:47] <darrelmiller> Yeah, this open source public domain stuff is awesome. If only someone would pay us to do it.
- [05:47] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) left irc: Quit: Leaving.
- [05:49] <mk_> totally
- [05:49] <mk_> I've got some overlap with my day job atm I just need to get around to the relevant bits on my queue and convince the people who pay me to let me release it
- [05:50] <mk_> they probably will though, which is awesome :)
- [05:52] <darrelmiller> You can thank the Canadian government for most of my OSS efforts. :-)
- [05:52] <darrelmiller> ..oops, I mean research and development work.
- [05:52] <mk_> :))
- [06:38] <mk_> I'm such a generous chap
- [06:38] <mk_> I took the time to teach JJ something about hypermedia
- [06:38] <mk_> positive karma for me.
- [06:39] <mk_> spot the problem with this statement:
- [06:39] <mk_> Basically, hypermedia is a mess. The semantics of links are so poor that you have to create your own markup to use it beyond the trivial "self" and "related". For instance, just for thumbnails, they create a markup that looks like this:
- [06:39] <mk_> <media:thumbnail url='http://i.ytimg.com/vi/MkMXxn7CjNE/0.jpg' height='360' width='480' time='00:01:41'/>
- [06:39] <mk_> we do? really?
- [06:39] <mk_> interesting.
- [06:40] <mk_> I almost pointed out that he wouldn't have made such a fundamental mistake if he'd have been constrained by HAL
- [06:40] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) joined #rest.
- [06:40] <mk_> decided against it in the end, on the basis that he might track me down and stab me.
- [07:01] <mk_> mamund: the whole thing about H Factors not having a way to do boundary embedding, or whatever
- [07:01] <mk_> that's really bugging me :)
- [07:01] <mk_> also - mamund I want your help/feedback on my latest vision for 'forms' in HAL :)
- [07:10] <mk_> "You have been unsubscribed from rest-discuss" :O
- [07:10] <mk_> drama!
- [07:17] pezra (~Adium@webmail.openlogic.com) joined #rest.
- [07:25] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
- [07:27] pezra (~Adium@webmail.openlogic.com) joined #rest.
- [07:42] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) joined #rest.
- [07:50] hdave1 (~hdave@static-71-245-233-130.bstnma.fios.verizon.net) left #rest.
- [08:44] mamund finally shows up
- [08:44] bigbluehat (~bigblueha@nmd.sbx08595.greensc.wayport.net) left irc: Quit: Leaving.
- [08:47] <mk_> mamund: I want your help
- [08:47] <mk_> :D
- [08:47] <mk_> please can I have it?
- [08:49] <mamund> 'sup, dide?
- [08:49] <mamund> 'sup, dude?
- [08:49] <mk_> nah much, you ?
- [08:50] <mk_> :)
- [08:50] <mamund> :)
- [08:50] <mk_> trying to finish off this damn post about self-descriptiveness
- [08:50] <mk_> got distracted by JJ
- [08:50] <mamund> !
- [08:50] <mamund> LL
- [08:50] <mk_> tried to explain the purpose of HTTP's semantics and self-descriptiveness
- [08:50] <mk_> what do you think about this:
- [08:51] mamund pulls up a chair
- [08:51] <mk_> HTTP's semantics; they're there to produce system-wide visibility in the interactions to enable intermediate processing. Applications that operate via HTTP (e.g. OMA) are meant to layer their specific semantics over the top of the protocol's.. so why is it surprising or a problem that OMA model cancelation using DELETE?
- [08:51] <mk_> REST (hypertext) applications are about shared understanding between client and server.. HTTP the application protocol is about having a system that constrain client/server interactions so that applications 'do their thing' within the bounds of some generalised semantics that are useful systemically. A very good (and necessary, in the case of the web) example of "systemically useful, generalised semantics" are the parts of HTTP that enable cachi
- [08:51] <mk_> ng.
- [08:52] <mamund> yeah, i like where this is coming from
- [08:52] <mk_> ok good just want to make sure I'm not veering off dangeoursly into crazy-land
- [08:52] <mamund> actually been working a similar POV, too.
- [08:52] <mk_> ooh
- [08:53] <mk_> I can't spell dangerously apparently
- [08:53] mamund notes great minds think alike
- [08:53] <mk_> or I'm spying on you and stealing all your awesome ideas
- [08:53] <mk_> nice hair cut, btw
- [08:53] <mamund> i've been thinking that a very cool way to get people to think about this whole thing is to focus on the _network_ itself
- [08:54] <mamund> shared understanding across the network
- [08:54] <mamund> shared understanding that lets parties "do their thing" on the network
- [08:54] <mamund> that's your "systematically useful, generalized semantics" i think.
- [08:55] <mk_> yeah those semantics relate to that definition of visibility I shared the other day
- [08:55] <mamund> yeah
- [08:56] <mk_> I say share, I really mean brain puke'd
- [08:56] <mamund> see, you got me thinking about this again
- [08:56] <mamund> LOL
- [08:56] <mk_> "Visibility is the degree of semantic assertion an intermediary can make about an interaction"
- [08:56] <mamund> hmmm....
- [08:56] <mk_> so HTTP is a set of generalised semantics that help intermediaries make semantic assertions
- [08:56] <mamund> as long as the "semantic" is about the network, right?
- [08:57] <mk_> about interactions
- [08:57] <mamund> yes
- [08:57] <mamund> see, that's the "extra" thing that i think needs more description
- [08:57] <mamund> the word "semantic" is the new "REST"<g>
- [08:57] <mk_> well I just mean ;meaning'
- [08:57] <mamund> sooo many overladed defs for that word
- [08:57] <mamund> :)
- [08:57] <mamund> ok
- [08:58] <mk_> degree of meaning intermediaries can assert/assume about an interaction
- [08:58] <mk_> maybe assume is a better word, actually
- [08:58] <mamund> could be
- [08:58] <mamund> you see what i'm on about, tho
- [08:58] <mamund> it's a _very_ small point, tho
- [08:58] <mamund> just about clarifying the prose
- [08:58] <mk_> well I get your point about avoiding poncy words
- [08:58] <mamund> yeah
- [08:59] <mk_> I don't even know why I've picked that up
- [08:59] <mamund> LOL
- [08:59] <mk_> I hate using words like that
- [08:59] <mk_> I'm a total dick.
- [08:59] <mk_> noooOOOOOO
- [08:59] <mk_> it's happened to me
- [08:59] mamund looks away
- [09:00] <mamund> :)
- [09:00] <mamund> anyway, i like what you are saying here
- [09:00] <mamund> it's something i've been mulling quite a bit lately
- [09:00] <mk_> it's good to make sense.
- [09:00] <mk_> (tm)
- [09:01] <mamund> fwiw, i think it can make a big dent in the whole picture of REST, dist net apps, etc.
- [09:01] <mamund> this visibility/self-desc. has been vague for a very long time
- [09:02] <mk_> shall I let you go first ?
- [09:04] <mamund> no, no
- [09:04] <mamund> you take the lead here
- [09:04] <mamund> i got my hands full right now anyway
- [09:06] <mk_> there's a real danger I will totally fuck it up and send everyone off on a wild goose-chase.
- [09:06] <mk_> (where I'm the stupid bloody goose)
- [09:08] <mamund> LOL
- [09:08] <mamund> no danger here
- [09:08] <mamund> this is, IMO, very good stuff
- [09:08] <mamund> if you're "unsure", do the trick i do..
- [09:09] <mamund> announce an "experiment" and then prattle on
- [09:09] <mamund> works for me every time<g>
- [09:13] KevBurnsJr (~KevBurnsW@50.0.103.39) joined #rest.
- [09:23] <mk_> well that was unexpected - he agreed with me
- [09:29] <mamund> interesting....
- [09:30] <mk_> he is such a weird dude.
- [09:30] <mamund> LOL
- [09:40] mhausenblas_ (~mhausenbl@wg1-nat.fwgal01.deri.ie) joined #rest.
- [09:41] mhausenblas (~mhausenbl@wg1-nat.fwgal01.deri.ie) left irc: Read error: Connection reset by peer
- [09:41] mhausenblas_ -> mhausenblas
- [09:48] mhausenblas (~mhausenbl@wg1-nat.fwgal01.deri.ie) left irc: Quit: brb
- [09:55] whartung (~whartung@wsip-98-189-78-118.oc.oc.cox.net) joined #rest.
- [10:02] nkoza (~NKoza@190.244.122.9) joined #rest.
- [10:07] aGHz (~Adium@modemcable153.0-178-173.mc.videotron.ca) left irc: Ping timeout: 244 seconds
- [10:25] bigbluehat (~bigblueha@adsl-98-71-143-231.gsp.bellsouth.net) joined #rest.
- [10:45] nkoza (~NKoza@190.244.122.9) left irc: Quit: Client exiting
- [10:51] <mk_> Oh cool irc works good on my phone :)
- [10:51] <mk_> win!!
- [10:51] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
- [10:54] <mk_> mamund: you clear on why this guy is grouping media types that way?
- [10:56] <mk_> I guess he is jus compiling lists, Im surprised he isn't doing it in terms of h factors
- [10:58] pezra (~Adium@webmail.openlogic.com) joined #rest.
- [11:00] <mamund> the list seems like an innocent idea
- [11:00] <mamund> it's just that it's kind of a rats nest once you get started<g>
- [11:01] <darrelmiller> Why is WADL included in there?
- [11:01] <mamund> no clue
- [11:01] pezra (~Adium@webmail.openlogic.com) left #rest.
- [11:01] <darrelmiller> WADL doesn't support updating the metadata does it?
- [11:02] <mamund> not that i know of
- [11:03] aGHz (~Adium@74.198.165.89) joined #rest.
- [11:03] <mamund> WADL is a read-only format
- [11:03] <mamund> link there could be labled as LO, i guess
- [11:03] <darrelmiller> That's what I thought.
- [12:00] darrelmiller (~darrelmil@70.24.176.12) left irc: Quit: ~ Trillian Astra - www.trillian.im ~
- [12:38] Hakon|mbp (~hakon1@130.82-134-26.bkkb.no) left irc: Quit: Leaving...
- [12:52] pezra (~Adium@webmail.openlogic.com) joined #rest.
- [13:31] <mk_> I have a special group in my hypermedia-categorisation-framework for things like WADL
- [13:31] <mk_> it's basically extended H Factors
- [13:31] <mk_> I just added an S factor
- [13:34] <mamund> interesting thread: http://lists.w3.org/Archives/Public/ietf-http-wg/2012JanMar/0030.html
- [14:06] <mk_> :)
- [14:07] <mk_> getting scope creep on this blog post
- [14:07] <mk_> want to add something about UA's like browsers being a form of intermediary
- [14:07] <mk_> think that be a bit too fruity for some
- [14:07] <mamund> see, that's what happens to me, too.
- [14:08] <mamund> btw - the "intermediary POV" is, IMO, the key to this whole visibility stuff anyway
- [14:08] <mamund> that's why i asked you about this
- [14:08] <mamund> (as you got the intermediary 'chops' here)
- [14:09] <mamund> so, what is the basics of your browser as intermediary thing?
- [14:09] <mamund> s/what is/what are/
- [14:14] <mk_> well I don't have a very nice way of explaining it right now
- [14:14] <mamund> :)
- [14:15] <mk_> but basically all the important mechanisms for a standardised media type revolve around generic mechanisms that take place 'in front' of the actor
- [14:17] <mk_> like javascript/stylesheet/image includes etc...
- [14:17] <mamund> in front of the actor, ok
- [14:17] <mamund> so...
- [14:18] <mamund> the browser is the "intermediary" between the actor (user) and the network(?)
- [14:19] <mk_> I guess the network yeah, I really just think about the webpage responses
- [14:19] <mamund> well, that's an important point that i don't want to mess up when listening toyou.
- [14:20] <mamund> browser as intermediary between the actor and the response
- [14:20] <mamund> better?
- [14:20] <mamund> (response in HTTP today can be "compound" - i.e. HTML, JS, CSS, etc.)
- [14:21] <mk_> basically the reason I think in that way is because I think that the value of a standardized type is in visibility (i.e. enabling intermediary processing)
- [14:21] <mamund> yes, i see this.
- [14:21] <mk_> everything else can actually be separated out using 'extension mechanisms' (like rels, or whatver)
- [14:22] <mamund> it's not the same way i view media types, but i see your POV very clearly.
- [14:22] <mamund> yes, "extensions" "decorations", "annotations", etc.
- [14:22] <mamund> whatever name one thinks fits
- [14:22] <mk_> right - but mixed in on top, rather than embedded in the type
- [14:23] <mamund> yes, yes
- [14:23] <mamund> that's HAL's message, right?
- [14:23] bigbluehat (~bigblueha@adsl-98-71-143-231.gsp.bellsouth.net) left irc: Quit: Leaving.
- [14:23] <mamund> (the "essence" of HAL, i mean)
- [14:23] <mk_> right
- [14:24] <mk_> and if you modela browser as an intermediary it's kind of how html works too
- [14:24] <mamund> yep
- [14:24] <mamund> i like this line of tought
- [14:24] <mamund> tought
- [14:24] mamund shees!
- [14:24] <mk_> :D
- [14:25] <mamund> see if i have it right...
- [14:26] <mamund> visibility on the network means media types that are "domain-free"...
- [14:26] <mamund> but these same types can be "decorated" with specifics needed for a particular use case
- [14:26] <mamund> yes?
- [14:26] <mamund> no?
- [14:27] <mk_> well it doesn't necessitate domain-free media types
- [14:28] <mamund> ok
- [14:28] <mamund> i am proly trying to make a "leap" regarding media types....
- [14:28] <mamund> you didn't bring that up in your initial dicussion here
- [14:28] <mamund> i did
- [14:29] <mk_> visibility on the network does depend on standardization, ubiquity (more so than standardization, really)
- [14:29] <mamund> ubiquity of....?
- [14:29] <mk_> the type
- [14:30] <mamund> here's what i struggle w/ right now...
- [14:30] <mamund> i think the MIME type is meaningless when discussion network-level visibility
- [14:31] <mamund> headers are important
- [14:31] <mamund> body is not
- [14:31] <mamund> for proof? just gzip all bodies; things will work just fine.
- [14:32] <mamund> now it is certainly possible to muck-up visibility using a bad message body model (yes, SOAP, i'm looking at you)
- [14:33] <mamund> but i don't think the opposite it true (i.e. you can't _improve_ network visibility w/ a *better* message body model)
- [14:33] <mamund> but... i'm not sure<g>
- [14:33] <mk_> well there are a small set of things that body's can do
- [14:34] <mk_> ESI is the most obvious example
- [14:34] <mamund> how does ESI affect visibliity on the network?
- [14:34] <mamund> (asking to be dumb, really)
- [14:35] <mk_> it allows intermediary processing by standardising how to represent transclusion in HTML docs
- [14:36] <mk_> intermediary-processed transclusion, specifically
- [14:38] <mamund> well, is this network visibility here?
- [14:38] <mamund> that's my real Q
- [14:39] <mk_> sure, that's why there are implementations on standard proxies like squid and varnish
- [14:39] <mamund> so, refresh me, are there headers that come into play here?
- [14:41] <mk_> http://blog.stateless.co/day/2011/10/22
- [14:41] <mk_> woops
- [14:41] <mk_> http://www.w3.org/TR/edge-arch
- [14:41] <mamund> so it's the headers that provide the visibility on the network, right?
- [14:42] <mk_> well the standard semantics of the ESI mechanism itself too
- [14:42] <mamund> but proxies only "jump in" when the header appears, right?
- [14:42] <mamund> otherwise, they skip attempts to process ESI
- [14:42] <mk_> right
- [14:43] <mamund> so the headers make ESI possible
- [14:44] <mamund> anyway, i'm dragging this off into the weeds, i think<g>.
- [14:45] <mamund> your initial comments were very cool; looking to see more.
- [14:45] <mamund> POST 2 DA BLOGZ AWREDDY!
- [14:45] <mamund> :)
- [14:47] <mk_> another example that's less convoluted
- [14:47] <mk_> is accelerating proxies that prefetch css/javascript and inline it in the page
- [14:48] aGHz (~Adium@74.198.165.89) left irc: Quit: Leaving.
- [14:48] <mamund> yeah, how is that implemented!?
- [14:50] <mk_> the proxies just dip into the page, pull out the script and stylesheet links, fetch the content, shove it into the doc inline and then deliver it to the client
- [14:50] <mk_> it was/is used a lot for mobile clients
- [14:50] <mamund> so, no signalling, this is just "aggresive behavior" of the proxy<g>
- [14:51] nuclearsandwich (~nuclearsa@74-93-3-244-SFBA.hfc.comcastbusiness.net) left irc: Read error: Connection reset by peer
- [14:51] Wombert (~Wombert@dslb-088-067-188-023.pools.arcor-ip.net) left irc: Quit: Wombert
- [14:52] <mk_> yea
- [14:52] <mk_> Surrogate-Control: no-store, content="ESI/1.0"
- [14:52] <mk_> shite
- [14:52] <mk_> Surrogate-Control: no-store, content="ESI/1.0"
- [14:52] <mk_> fs lol
- [14:53] mamund feels so much better now
- [14:53] Wombert (~Wombert@dslb-088-067-188-023.pools.arcor-ip.net) joined #rest.
- [14:53] <mk_> http://www.infosyncworld.com/news/n/5018.html
- [14:53] <mamund> yep
- [14:53] <mk_> so I dunno.. that's why I can't completely discount media types/the message body from visibility
- [14:54] <mamund> well, they can adeversely affect visiblity (IOW, force network participants to "peek in")
- [14:55] <mamund> but i don't think they can _improve_ network visibility.
- [14:55] <mk_> so you don't think HTML improves those kinds of acceleration mecahnism ?
- [14:55] <mamund> (ie. including somethin ghte _body_ makes the mesage "more visible"
- [14:55] <mk_> s/improves/allows
- [14:56] nuclearsandwich (~nuclearsa@74-93-3-244-SFBA.hfc.comcastbusiness.net) joined #rest.
- [14:56] <mamund> i don't think anything particular to HTML makes this better
- [14:56] <mamund> could be wrong
- [14:56] <mk_> those mecahnisms depend on html being standard
- [14:56] <mamund> ahhh
- [14:57] <mk_> 'standard'
- [14:57] <mk_> I don't mean 'standardised'
- [14:57] <mk_> I really mean ubiquitous
- [14:57] <mamund> ok, i see where yo are going here
- [14:57] <mk_> I actually think registries/standardisation processes are basically a distraction
- [14:58] <mk_> it's annoying that Roy included them in the diss so explictly
- [14:58] <mk_> mamund: am I headed somewhere good or somewhere bad ?
- [15:00] <mamund> i think good.
- [15:00] <mamund> just not the same place i was looking<g>
- [15:00] <mamund> ok, i gotta split soon.,
- [15:01] nuclearsandwich (~nuclearsa@74-93-3-244-SFBA.hfc.comcastbusiness.net) left irc: Read error: Connection reset by peer
- [15:01] nuclears1ndwich (~nuclearsa@74-93-3-244-SFBA.hfc.comcastbusiness.net) joined #rest.
- [15:15] mamund is wandering off ......
- [15:17] <mk_> bye
- [15:23] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) joined #rest.
- [15:51] Wombert (~Wombert@dslb-088-067-188-023.pools.arcor-ip.net) left irc: Quit: Wombert
- [15:52] Wombert (~Wombert@dslb-088-067-188-023.pools.arcor-ip.net) joined #rest.
- [16:03] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) left irc: Quit: Leaving...
- [16:27] pezra (~Adium@webmail.openlogic.com) left irc: Quit: Leaving.
- [17:13] <KevBurnsJr> mk_: We control surrogates mostly using custom headers
- [17:14] <mk_> oh ok
- [17:14] <mk_> what kind of stuff are they doing ?
- [17:14] <KevBurnsJr> we use varnish so...
- [17:15] <KevBurnsJr> X-BP-Enable-ESI: 1
- [17:15] <KevBurnsJr> of course that leads our varnish configs to resemble an awful hack
- [17:16] <KevBurnsJr> like we have one
- [17:16] <KevBurnsJr> X-Cache-Edge
- [17:16] <KevBurnsJr> the effect of which is to force max-age to 0 and Expires to the epoch
- [17:17] <KevBurnsJr> if it is false
- [17:17] <KevBurnsJr> because infrastructure doesn't trust app-devs to set headers properly :P
- [17:18] <KevBurnsJr> I often feel uncomfortable using the s-maxage header to control my reverse proxy server
- [17:19] <KevBurnsJr> because of what might lie downstream
- [17:19] <KevBurnsJr> it appears the surrogate control header is designed to differentiate between surrogates and other types of shared caches
- [17:25] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
- [17:26] <mk_> indeed
- [17:26] <mk_> I didn't know about it at all until I made a blog post suggesting it should exist
- [17:26] <mk_> and then someone pointed out it already did.
- [17:28] <KevBurnsJr> hehe
- [17:28] <KevBurnsJr> that tends to happen
- [17:28] <mk_> it's cool you're using headers to control that though
- [17:29] <mk_> instead of gnarly URL pattern matching
- [17:33] <KevBurnsJr> ew ya
- [17:34] <KevBurnsJr> course
- [17:34] <KevBurnsJr> ...
- [17:34] <KevBurnsJr> well
- [17:34] <KevBurnsJr> i was going to say something but i probably shouldn't
- [17:34] <mk_> lol..
- [17:34] <mk_> was it racist ?
- [17:34] <KevBurnsJr> no
- [17:35] <mk_> so it can't be that bad
- [17:36] <KevBurnsJr> I did just implement a konami code to our codebase without telling anyone
- [17:37] <KevBurnsJr> u,u,d,d,l,r,l,r,b,a,start
- [17:37] <KevBurnsJr> http://snaptortoise.com/konami-js/
- [17:37] <mk_> oh.. win.
- [17:38] <KevBurnsJr> Director of engineering called me out on it in a meeting today. Apparently he reads the commit logs.
- [17:39] <mk_> hahaha
- [17:40] <mk_> 1% always downpressing us
- [17:40] mk_ shakes fist
- [17:51] aGHz (~Adium@modemcable153.0-178-173.mc.videotron.ca) joined #rest.
- [17:54] Wombert (~Wombert@dslb-088-067-188-023.pools.arcor-ip.net) left irc: Quit: Wombert
- [17:58] <KevBurnsJr> cheers
- [17:58] KevBurnsJr (~KevBurnsW@50.0.103.39) left irc: Quit: 6:00
- [19:03] pezra (~Adium@184-96-219-53.hlrn.qwest.net) joined #rest.
- [19:04] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) left irc: Read error: Connection reset by peer
- [19:05] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) joined #rest.
- [19:27] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Remote host closed the connection
- [19:28] pezra (~Adium@184-96-219-53.hlrn.qwest.net) left irc: Quit: Leaving.
- [20:10] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) joined #rest.
- [20:20] sbanwart (~sbanwart@99-177-126-136.lightspeed.bcvloh.sbcglobal.net) left irc: Read error: Operation timed out
- [21:33] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) left irc: Ping timeout: 240 seconds
- [21:33] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) joined #rest.
- [21:44] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) joined #rest.
- [23:16] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) got netsplit.
- [23:16] Blazeix (~Blazeix@71.74.190.197) got netsplit.
- [23:17] scudco (~scudco@cpe-75-85-13-152.socal.res.rr.com) returned to #rest.
- [23:17] Blazeix (~Blazeix@71.74.190.197) returned to #rest.
- [00:00] --- Wed Jan 18 2012