#rest on irc.freenode.net 2012-01-10 (txt)

[00:15] darrelmiller (~darrelmil@65.94.30.184) left irc: Ping timeout: 240 seconds
[00:27] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) left irc: Remote host closed the connection
[00:27] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) joined #rest.
[00:31] kennethreitz (~kennethre@c-24-127-96-129.hsd1.va.comcast.net) left irc: Quit: Computer has gone to sleep.
[00:32] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) left irc: Ping timeout: 252 seconds
[01:27] caaakeeey (~caaakeeey@gatekeeper-ext.zeus.com) joined #rest.
[01:42] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[01:43] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[03:52] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) joined #rest.
[04:56] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[04:57] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[05:19] zazi (~zazi@p579D3696.dip.t-dialin.net) joined #rest.
[05:20] <zazi> I'm happy that the current discussion on rest-discuss is getting into the right direction ;)
[05:22] <mikekelly> lol :)
[05:23] sbanwart (~sbanwart@72.240.63.122) joined #rest.
[05:38] <zazi> mikekelly, could you reply on this message http://tech.groups.yahoo.com/group/rest-discuss/message/18396 ?
[05:42] <mikekelly> zazi: yeah thanks for reminding me - I left it this morning cos I was reading it on my mobile and i'm crap at writing emails on my phone :)
[05:43] <mikekelly> I'll try and respond in a bit, gotta sort a few things first
[05:47] darrelmiller (~darrelmil@70.24.176.12) joined #rest.
[05:49] <zazi> cool, no problem, take your time
[06:09] <darrelmiller> Wow, just read some of that rest-discuss stuff. I'm shocked that we are debating the value of LE vs LO.
[06:10] <mikekelly> I don't see why you would not determine that via a relation
[06:10] <darrelmiller> Sure. I would.
[06:10] <darrelmiller> But that didn't seem to be the debate.
[06:11] <mikekelly> well I think mamund was suggesting that didn't count
[06:11] <darrelmiller> The debate seemed to be that the client doesn't need to know the difference.
[06:12] <mikekelly> ok I took the discussion to be that a media type for machines doesn't need a hypermedia control for both
[06:12] <mikekelly> (since you can delegate it to the relation)
[06:12] <darrelmiller> ahhh, ok maybe.
[06:12] bigbluehat (~bigblueha@adsl-74-177-98-159.gsp.bellsouth.net) joined #rest.
[06:12] <mikekelly> I think mike was saying that's not 'real' LE
[06:13] <mikekelly> and my point was really just to figure out whether or not HAL's "framing" or wahtever the word is
[06:13] <darrelmiller> Actually, I was thinking about that the other day. What's the difference between a HAL Link element and a Resource with no content?
[06:13] <mikekelly> :)
[06:13] <mikekelly> a resource with no content is basically how I want to build ESI (and auto-transclusion) into HAL
[06:14] <darrelmiller> Seems reasonable. By using a Link you are explicitly stating that you don't want to embed any content.
[06:14] <mikekelly> right
[06:14] <darrelmiller> but you could easily do HAL without using links at all.
[06:15] <mikekelly> yeah HAL originally was just links
[06:15] <mikekelly> either self-closing or not
[06:15] <darrelmiller> right. That rings a bell.
[06:15] <mikekelly> back in a bit
[06:26] <mikekelly> that thread is monstrous..
[06:27] <darrelmiller> and the Yahoo interface to it sucks. And my Gmail is not much better.
[06:27] <mikekelly> yeah I much prefer google groups, personaly
[06:27] <darrelmiller> agreed
[06:39] sbanwart (~sbanwart@72.240.63.122) left irc: Ping timeout: 240 seconds
[06:40] caaakeeey (~caaakeeey@gatekeeper-ext.zeus.com) left irc: Ping timeout: 240 seconds
[06:40] caaakeeey (~caaakeeey@gatekeeper-ext.zeus.com) joined #rest.
[06:41] <darrelmiller> What am I not understanding? Why do people use the term protocol when talking about media type details?
[06:42] <darrelmiller> What do <img> and <a> have to do with protocol? Aren't we using HTTP as our application protocol?
[06:43] <darrelmiller> I would say <img> and <a> have a "processing model" that defines how a user-agent interacts with the HTTP protocol.
[06:45] mikekelly looks to mamund..
[06:48] <darrelmiller> mikekelly: mamund is still sleeping :-)
[06:48] <darrelmiller> We will abuse him later. muhahahaha
[06:48] <mikekelly> wtf it's almost 3pm..
[06:48] <mikekelly> laaazyyyy..
[06:49] <darrelmiller> What can I say... <whispers>he's old, you know</whispers>
[06:51] <mikekelly> bwaha.
[06:53] zazi (~zazi@p579D3696.dip.t-dialin.net) left irc: Ping timeout: 255 seconds
[07:10] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) joined #rest.
[07:50] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) joined #rest.
[08:03] quest88 (~quest88@c-98-207-205-137.hsd1.ca.comcast.net) left irc: Quit: quest88
[08:13] mamund wakes up, grabs his walker, and heads to the day room w/ the other crazies
[08:13] <darrelmiller> mamund: Sleep well :-)
[08:13] <mamund> LOL, not really<g>
[08:14] <mamund> so, i guess i annoyed the folks on the list over the last 12 hrs, eh?
[08:14] <darrelmiller> I don't think anyone is annoyed.
[08:14] <mamund> had a couple chances to "end it" but didn't
[08:14] <mamund> well, that's nice, i guess
[08:19] DracoBlue (~Adium@ip-vlan-obckunde-02-217-66-60-14.pixelpark.net) left #rest.
[08:23] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) left irc: Read error: Connection reset by peer
[08:23] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) joined #rest.
[08:25] <mikekelly> huh ?
[08:25] <mikekelly> I don't think anyone's annoyed mamund :P
[08:26] <mikekelly> I should know, I'm the expert.
[08:27] <mamund> heee
[08:27] <mamund> well,that thread draged on for quite a while; not sure if it provdied much new info, etc.
[08:27] <mikekelly> yeah sorry about that
[08:27] <mamund> LOL
[08:28] <mikekelly> I did try my hardest to start it off on the right foot
[08:28] <mamund> both you and darrelmiller added good new data
[08:28] <mamund> i thought of rel="stylesheet", but left it out
[08:28] <darrelmiller> mamund: Well it pointed out to me that some poeple still don't realize the significance of LE and LO.
[08:28] <mikekelly> but quickly got distracted by people telling me I "don't get it" :|
[08:28] <darrelmiller> That's good to know.
[08:28] <mamund> i _did_ _not_ think to pursue the state transition POV, tho.
[08:29] <mamund> yes, darrelmiller, i think you are right.
[08:29] <mikekelly> mamund: was my point about rel="stylesheet" even clear?
[08:29] <mikekelly> or did I miss the point ?
[08:29] <mamund> to me, it was.
[08:29] <darrelmiller> +1
[08:29] <mamund> as was the inline resource idea w/ HAL & Atom.
[08:29] <mikekelly> ok cool
[08:29] <mamund> all good stuff, IMO.
[08:29] <mikekelly> yeah - is that worth making distinct or not ?
[08:29] <mikekelly> I always get confused when we talk about 'embed'
[08:30] <mamund> it was for me.
[08:30] <mamund> now i am not sure about teh frame stuff.
[08:30] <mikekelly> I dunno why I picked frame
[08:30] <mamund> i thought about that quite a bit when forming teh factors.
[08:30] <mikekelly> stupid word :|
[08:30] <mamund> i decided not to include it
[08:30] <mamund> well, i think i know what you mean, tno
[08:30] <mamund> boundary, etc.
[08:30] <mamund> right?
[08:30] <mikekelly> yeah
[08:30] <mamund> ypoe
[08:30] <mamund> yep
[08:30] <mikekelly> I was thinking maybe
[08:30] <mamund> <g>
[08:30] <mikekelly> active/passive transclude
[08:30] <mikekelly> or something
[08:31] <mamund> yeah, i think the boundary notion also touches on the state change point from darrelmiller
[08:31] <mamund> maybe not
[08:31] <mamund> anyway
[08:32] <darrelmiller> mikekelly: Are you trying to distinguish between a LE and a representation that embeds other resource representations directly in the response?
[08:32] <mikekelly> yeah
[08:32] <mikekelly> exactly
[08:32] <mikekelly> to me those are two distinct things
[08:32] <mikekelly> 'things'
[08:32] <mikekelly> :)
[08:32] <mamund> i think Markus continues to think of HTTP as a file-style protocol. and that messages are simply data moved about.
[08:32] <darrelmiller> Hmmm. I don't know if the distinction is necessary to formalize though.
[08:33] <mikekelly> darrelmiller: it's definitely helpful when analysing a media type
[08:33] <mikekelly> it's something a *lot* of media types do
[08:34] <darrelmiller> yes, but is the distinction necesary to be able to consume the representation?
[08:34] <mikekelly> the reason I chose 'frame' over boundary is because e.g. in HAL it's not only a boundary but it has semantic meaning that resets the context URI of the containing info
[08:34] <mikekelly> darrelmiller: no, but I thought H Factors were more a framework for analysing hypermedia ?
[08:35] <darrelmiller> IMHO a H Factor is only significant if it provides some information to the client.
[08:35] <mikekelly> mamund ..?
[08:35] <mikekelly> :)
[08:35] <mikekelly> he doesn't know.
[08:35] <mamund> sorry, was backgrounding this convo
[08:36] <mikekelly> he's probably got confused or fallen asleep again
[08:36] <mamund> 'sup?
[08:36] <mikekelly> oh there he is
[08:36] <mikekelly> wake up.
[08:36] <darrelmiller> Does the client care whether the resource is embedded or just part of a bigger representation.
[08:36] <mamund> was playing mahjong w/ the old ladies
[08:36] <mamund> hmmm
[08:36] <darrelmiller> I suppose you could allow a client to "refresh" the embedded resource.
[08:36] <mamund> pretty sure that's context dependent, right?
[08:36] <mamund> as in "yes, i might"?
[08:36] <mamund> why ask?
[08:36] <darrelmiller> That is only possible because it is an embeded resource.
[08:37] <mamund> think of Jon Moore's example where the media type actually has options to allow the client...
[08:37] <mamund> to 1) just render the link
[08:37] <mamund> 2) resolve the link client side as an agument to the representation
[08:37] <darrelmiller> mikekelly: But HAL already provides this distinction by using the <resource> element. Are you trying to elevate this concept to a H-Factor?
[08:38] <mamund> 3) or the server sends the entire "augment" w/o a link
[08:38] <mikekelly> darrelmiller: yeah elevate it to an h factor
[08:38] <darrelmiller> Ahhh.
[08:38] <darrelmiller> You probably need to discuss that with some authority on H-Factors ;-)
[08:39] <mikekelly> I was trying to do that on rest-discuss but the grumpy old bastard ignored me
[08:39] <mamund> mikekelly: what is the hypermedia factor you want to identify
[08:39] mamund whacks mikekelly w/ his cane
[08:39] <mikekelly> steady on.
[08:39] <mamund> :)
[08:39] <mikekelly> :D
[08:39] <mikekelly> erm
[08:40] <mikekelly> it's to have 2 separate h factors for this "pre-fetched"/passive transclusion vs client-fetched/active transclusion
[08:41] <mikekelly> it's possible to do neither, either, or both
[08:41] <darrelmiller> Do you need to distinguish between passive partial transclusion and passive full transclusion?
[08:41] <mikekelly> and w/ the same or two controls for either
[08:42] <mikekelly> darrelmiller: no I wouldn't go that far in the h factors
[08:42] <mikekelly> I don't have a scientific explanation as to why though :)
[08:42] <mikekelly> infact, I don't have a scientific explanation for anything
[08:42] <darrelmiller> See, going back to my previous example of doing a refresh. You couldn't do a refresh on a partial.
[08:42] <darrelmiller> All you could do is convert a partial into a full.
[08:42] <mikekelly> or refresh the containing resource ?
[08:43] <darrelmiller> ..because it is the container resource that decides how much of the embedded resource to include.
[08:43] <mikekelly> yeah, totally
[08:43] <mikekelly> well that is the case with HAL anyway
[08:44] <mikekelly> you could have a different type which says it must be full
[08:44] <mikekelly> but I'd say that was a bad design
[08:44] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) joined #rest.
[08:44] <mikekelly> given that Content-Location doesn't work that way in HTTP
[08:44] <mikekelly> and it's kind of a similar conundru,
[08:44] <mikekelly> m
[08:45] <darrelmiller> Well my official position on the subject is, I don't know.
[08:45] <mikekelly> for me the 'passive transclusion' (do we need a better term?) is an optimisation thing
[08:46] <darrelmiller> LEP "Link Embed Passive"
[08:46] <mikekelly> is it Jon Moore that did the solution that can have clients deal with going from active to passive ?
[08:46] <darrelmiller> yes
[08:46] <mikekelly> that just makes me think leprecy
[08:46] <darrelmiller> PLE Passive Link Embed
[08:47] <darrelmiller> ooh and ALE for Active Link Embed
[08:47] <mikekelly> nice.
[08:47] <mikekelly> PALE ALE
[08:47] <darrelmiller> Passively Active ?
[08:47] <mikekelly> PAssive Link Embed
[08:48] <darrelmiller> Ahhh...
[08:48] <mikekelly> mamund: so.. when are you going to work in our corrections?
[08:48] <darrelmiller> ROFL
[08:49] <fu-manchu> embedding becomes much more important when your media type gets used outside of HTTP, where you can no longer assume every consumer of a document can follow a link
[08:49] <darrelmiller> Does there exist software that can't do HTTP requests? Yuck, glad I don't have to deal with that :-P
[08:50] <fu-manchu> I'm working on one right now where the media type essentially has to start out as hypermedia and then become gradually less so as it ages
[08:50] <mikekelly> fu-manchu: could you do that with hal ?
[08:50] <darrelmiller> So the ALE becomes PALE over time?
[08:50] <mamund> sorry, was called away
[08:50] <mikekelly> no worries, we fixed H Factors for you while you were gone
[08:51] <mamund> oh, so glad!
[08:51] <mikekelly> you're welcome.
[08:51] <mamund> ha!
[08:51] <mikekelly> ;)
[08:51] <darrelmiller> mamund: You can still update your book right?
[08:51] <mamund> no worries
[08:51] <mamund> btw - about to sign for another book starting this week.
[08:52] <mikekelly> wtf..
[08:52] <darrelmiller> you are a machine!
[08:52] <mamund> easy one this time, tho
[08:52] <darrelmiller> It's not on Web API is it?
[08:52] <mamund> nope
[08:52] <mikekelly> Extreme Zimmerframing ?
[08:52] <mamund> wht?
[08:53] <darrelmiller> I'm sorta talking with Pedro, Pablo and Glenn about doing one on Web API.
[08:53] <mikekelly> was woth a shot.
[08:53] <mamund> https://www.google.com/search?q=Zimmer+Frame&hl=en&sa=X&prmd=imvns&tbm=isch&tbo=u&source=univ&ei=72wMT__HNcL50gH9_9n7BQ&ved=0CDYQsAQ&biw=1301&bih=801
[08:53] <mamund> bwhaaa
[08:53] <mikekelly> saturated market?
[08:53] <darrelmiller> ROFL
[08:53] mamund spit his tea when that came up
[08:53] <mikekelly> is zimmerframe not an american thing?
[08:54] <mamund> never hared of it
[08:54] <darrelmiller> me neither.
[08:54] <mamund> course, i'm losing my hearing, too.
[08:54] <mikekelly> oh so my funny was actually a big culture fail
[08:54] <mamund> that's just hilarious, dude
[08:54] <darrelmiller> the images were awesome.
[08:54] <mikekelly> what's your book about then?
[08:54] <mamund> http://www.find-me-a-gift.co.uk/zimmer-frame-old-mans-costume-inflatable-zimmer-frame.html
[08:54] <mikekelly> you can't tease like that and not spill the means.
[08:55] <mikekelly> means?
[08:55] <mikekelly> beans.
[08:55] <mikekelly> heinz
[08:55] <mamund> http://www.motifake.com/image/demotivational-poster/1012/zimmer-frame-zimmer-frame-demotivational-posters-1292234010.jpg
[08:55] <mamund> sheesh, i could spend hours doing this.
[08:56] <mamund> Z-factors
[08:57] <mikekelly> ha
[08:57] <mamund> me running the half-marathon:http://www.yallaf1.com/wp-content/uploads/2010/10/d10kor1614.jpg
[08:57] <mikekelly> that is
[08:57] <mikekelly> interesting.
[08:57] <mamund> note the redbull
[08:58] <mamund> ha, the text is even better: http://www.yallaf1.com/2010/10/28/ecclestone-gives-finger-to-zimmer-frame-gift/
[08:58] <mamund> ok, ok
[09:04] zazi (~zazi@p579D3696.dip.t-dialin.net) joined #rest.
[09:06] bigbluehat (~bigblueha@adsl-74-177-98-159.gsp.bellsouth.net) left irc: Quit: Leaving.
[09:07] <whartung> o/
[09:08] <darrelmiller> \o/
[09:24] <mamund> darrelmiller: you should _totally_ do the Web API book
[09:25] <mamund> mikekelly: you should _totally_ to the Zimmerframing book
[09:26] <mamund> hey, bigbluehat made the InfoQ!
[09:26] <mamund> http://www.infoq.com/presentations/Why-CouchDB
[09:26] <darrelmiller> mamund: Glenn and Pablo are talking to publishers at the moment. Still not sure if this will go the traditional route of be self published.
[09:26] <mamund> keep me posted.
[09:26] <darrelmiller> I'm looking at doing more the client side stuff.
[09:26] <mamund> if you like, i can introduce you to my editor @ O'Reilly.
[09:27] <darrelmiller> bigbluehat: Nice !
[09:27] <darrelmiller> mamund: Apparently Glenn and Pablo have contacts there...
[09:27] <mamund> yep, i think O'Reilly does the MS Press stuff now, too.
[09:27] <darrelmiller> mamund: Did you see my HttpClient post?
[09:27] <mamund> yep
[09:27] <mamund> nice one
[09:28] <darrelmiller> I got more page views on that yesterday than I have in the last six months combined
[09:30] <mamund> very cool
[09:30] theartisan (u4087@gateway/web/irccloud.com/x-dkshgmcocfcjwack) joined #rest.
[09:34] theartisan (u4087@gateway/web/irccloud.com/x-dkshgmcocfcjwack) left #rest.
[09:42] bigbluehat (~bigblueha@adsl-74-177-98-159.gsp.bellsouth.net) joined #rest.
[09:47] dkubb (~dkubb@50.92.131.237) left irc: Quit: Linkinus - http://linkinus.com
[10:12] KevBurnsWork (~KevBurnsW@50.0.103.39) joined #rest.
[10:19] kennethreitz (~kennethre@204.14.152.118) joined #rest.
[10:23] mhausenblas (~mhausenbl@wlan-nat.fwgal01.deri.ie) left irc: Quit: mhausenblas
[10:29] <zazi> RDF (itself as knowledge representation structure/knowledge representation language) doesn't need LE, because it's all about resource _description_ - LE is only important for hypermedia clients that display content for humans
[10:30] <darrelmiller> zazi - It can also be useful in M2M scenarios. LE is just a way of partitioning a resource based on lifecycle of the data. i.e. how long it can be cached for.
[10:32] <zazi> darrelmiller, can you give an example in RDF (choose your preferred RDF serialisation format)
[10:32] <zazi> the image example doesn't work :)
[10:33] <darrelmiller> Not sure why it matters what the media type is.
[10:33] <darrelmiller> here's a example.
[10:33] <darrelmiller> I might get cut off though as a have a conference call starting soon....
[10:33] <zazi> okay, no problem
[10:33] <darrelmiller> My client app loads a root resource from a service.
[10:34] <darrelmiller> There are a set of available databases that are presented as links in the root resource.
[10:34] <darrelmiller> It's a multi-tenant type of app.
[10:34] <Jarda> darrelmiller: made any WPF apps?
[10:34] <darrelmiller> Jarda: Partially WPF and partially winforms.
[10:34] <Jarda> darrelmiller: have you used telerik radcontrols for wpf?
[10:35] <darrelmiller> We are using them on one of the projects I am working on now. I don't have much experience with them.
[10:35] <Jarda> ok
[10:35] <darrelmiller> zazi: One of the links in the root resource is to the user's preferences.
[10:35] <Jarda> I'm having thread safety problems..
[10:35] <Jarda> which sucks on a async http client app..
[10:36] <darrelmiller> jarda: Hell yeah.
[10:36] <zazi> I see no embedded link till now
[10:36] NOKAH (~hakon1@193.137.202.84.customer.cdi.no) joined #rest.
[10:36] <darrelmiller> zazi: The client does a LE on the preferences because inside the preferences, amongst other things is the "preferred" tenant that the user wishes to use.
[10:37] <darrelmiller> So to recap. The client retreives the root resource, then GETs a LE link, determines the preferred tenant and then does a LO to get to the "shell" of the tenant.
[10:37] <zazi> but that's just another resource with hopefully an own URI
[10:37] <darrelmiller> Sure.
[10:38] <darrelmiller> The difference to me is that the client did not "navigate" to the preferences end-point. It merely retreived that data and aggregated it into it;s current state.
[10:38] <zazi> I think, in RDF there are no concepts for root resource etc. - it's just a graph of interconnected resources
[10:38] <darrelmiller> It then navigated to the tenant.
[10:38] <darrelmiller> Sure, that's not really that relevant. You could start anywhere.
[10:39] <zazi> yep, that's why
[10:39] <darrelmiller> The point is from the client's perspective there are two completely different ways to consuming a link.
[10:39] mamund is back
[10:39] <darrelmiller> RDF is all about linking data. It's not about applications. That's the difference.
[10:39] <zazi> I know ;)
[10:40] <darrelmiller> IMHO REST is about exposing an application workflow, not data.
[10:40] <zazi> that's why there can't be LE
[10:40] Hakon|mbp (~hakon1@193.137.202.84.customer.cdi.no) left irc: Ping timeout: 240 seconds
[10:40] <zazi> yep, that's why we can combine both (hopefully) elegantly
[10:41] <darrelmiller> What do you mean "can't be LE" ?
[10:43] <zazi> I don't insist that the embedded link feature is applicable in a RDF M2M media type
[10:45] <zazi> i.e. the control flow you described would not fit into a new tag in that media type, e.g., <tenant>
[10:46] <zazi> more verbose: <tenant rel="user-profile" href="http://example.com/bobs-user-profile"; />
[10:50] <zazi> I think, the user's preferences resource also wouldn't be linked in that root resource with database links (endpoints)
[10:55] <zazi> anyway, it makes probably sense to differentiate between "navigation links" and "links that are utilised to process a resource and continue in the application flow"
[10:57] <zazi> maybe it's "embedded links" is just a bad naming for this case
[10:58] <zazi> it's like a parameter request
[11:00] <zazi> so a "navigation link" can be interpreted as "push application into a next state", e.g. "submit-order"
[11:01] <zazi> and an "embedded link" can be interpreted as "deliver information to be able to push application into a next state"
[11:02] <zazi> this could maybe also "include", or?
[11:05] <zazi> maybe the example you've describe is more LT than LE then, or? - i.e. retrieve the resource and run a (SPARQL) query to get a specific piece of information (the preferred tenant)
[11:06] <mamund> zazi: fwiw, i _am_ listening to your thoughts here...
[11:07] <mamund> this is interesting to me and any/all comments about whether some of the
[11:07] <mamund> H-Factors are vague/opaque, mis-named, etc. is helpful
[11:07] <mamund> thanks
[11:07] <darrelmiller> zazi: You bring up an interesting point. For me an LT would have to be either an LE or LO as well.
[11:07] <zazi> np, I'm thinking just out loud
[11:07] <mamund> yep, thanks.
[11:08] <zazi> thanks for listening ;)
[11:08] <zazi> what about "transclusion link"
[11:08] <darrelmiller> Your definitions of "navigation link" and "embedded link" make sense to me.
[11:08] <darrelmiller> That's how I understand LO and LE.
[11:09] <zazi> darrelmiller, I derived them from my understanding of your example
[11:09] <darrelmiller> Phew, I would have looked dense if I had disagreed :-)
[11:09] <zazi> :)
[11:09] <darrelmiller> the term "include" and LE are pretty much synonymous for me also.
[11:10] <zazi> okay, maybe I just had problem with embedded in the context of the image example then
[11:12] darrelmiller has gotta scrum. Be back soon.
[11:13] <zazi> the tipping point, there are actions for LEs which are beyond "include the resource" into the "root" resource, e.g., processing a query (as it is the case from the example above)
[11:15] <zazi> for the example above, a tag could be looking like this: <XYZ rel="user-profile" action-rel="retrieve-tenant" href="http://example.com/bobs-user-profile"; />
[11:18] <zazi> 'retrieve-tenant' could be http://example.com/retrieve-tenant and by resolving that URI one retrieves a (SPARQL) query that asks for the preferred tenant (sidenote: SPARQL queries could be described via SPIN, see http://spinrdf.org/ - an RDF format for SPARQL queries)
[11:20] <zazi> in such action description one could also include information such as (HTTP) method, or accept/enc-type (anthing else?)
[11:23] <zazi> btw, there already exist SPIN descriptions that are attached to usual RDF class (etc.) descriptions
[11:27] <zazi> btw2, in they are already able to define/describe functions
[11:29] <zazi> (however, in that case these are SPARQL functions)
[11:47] <zazi> mamund, maybe I could recycle the RDF Affordance Vocabulary you co-designed with danja ;)
[11:47] <mamund> zazi: i'd like to see more of that
[11:47] <mamund> danny knows this stuff (RDF design) _much_ better than i
[11:48] <mamund> i am most interested in what affordances look like in RDF formats.
[11:48] <mamund> and _which_ affordances make sense in that environment.
[11:48] <mamund> anything you can add to that is most welcome
[11:50] <zazi> I think, maybe one does not have to describe everything in RDF for the hypermedia case (see my sketches above)
[11:52] <zazi> e.g. another name proposal for 'action-rel' could be 'affordance'
[11:53] <zazi> (ps: I'm quite into RDF design as well ;) )
[11:54] <mamund> zazi: _names_ are not too interesting to me; IOW,that's fine.
[11:54] <mamund> what interests me are whether adding affordances to RDF makes sense
[11:54] <zazi> (pps: RDF designers sometimes tend to overdesign/describe everything in RDF, which is not always useful)
[11:54] <mamund> i.e. will this make a diff in building clients?
[11:54] <mamund> and if yes, _which_ affordances are the ones that will make the most noticeable diff
[11:55] <mamund> in coding clients.
[11:55] <zazi> names are interesting in the context of understanding ;)
[11:55] <mamund> sure, but in the end, they are a string of chars.
[11:55] <zazi> i.e. one should what other persons in the group speak about
[11:55] <mamund> pciing ones that provide the most accurate understanding in RDF-world is fine w/ me
[11:55] <mamund> yes, yes
[11:55] <zazi> +know
[11:56] <mamund> so, any/all efforts that help clarify the impact affordance designs for RDF is most welcome.
[11:56] <mamund> my recent discussions (yours and danja excepted) hve led me to think
[11:56] SvenDowideit (~SvenDowid@foswiki/developer/SvenDowideit) left irc: Excess Flood
[11:57] <mamund> that RDF community is not very interested in adding affordance to responses.
[11:57] <mamund> that's fine
[11:57] <mamund> but i am still quite interseted in the possibliites
[11:57] <zazi> well it depends, I think there are some that are interested in that topic
[11:57] SvenDowideit (~SvenDowid@203-206-171-38.perm.iinet.net.au) joined #rest.
[11:57] <mamund> sure, i think you (and djana) prove that
[11:58] <zazi> one the other side, there are many that like to described services, especially REST services (which is big misunderstanding from my POV, but that's another story ... ;) )
[11:59] <mamund> well, not all media types need affordances.
[11:59] <zazi> I think, mhausenblas is also in the boot
[11:59] <mamund> mhausenblas has been most encouraging and curious about my stuff; much apprecited.
[12:00] <zazi> and I think markuslanthaler most recently as well, or? ;)
[12:01] <zazi> and nathan, ...
[12:01] <mamund> well, i got the impression from Markus that the affordances i discussed were not important for his work.
[12:01] <mamund> yes, nathan has been interested, too.
[12:01] <mamund> ok, well.
[12:01] <mamund> fine.
[12:01] <mamund> do you have access to edit the wiki?
[12:01] <mamund> (affordances stuff)
[12:01] <zazi> yep, gtg now, and do ... something ... different
[12:01] <zazi> yep
[12:01] <mamund> :)
[12:02] <mamund> feel free to add to the wiki in any way you like
[12:02] <mamund> i don't think danny will mind at all. i know it would help me quite a bit
[12:02] <zazi> anyway, I think that I have now for the first time an impression how I could connect the different points in a useful and pragmatic manner
[12:03] <mamund> very cool
[12:03] <mamund> hey, if you want to email me directly on this (mamund@yahoo.com) feel free
[12:03] <zazi> okay
[12:03] <mamund> thanks again for the time
[12:04] <mamund> anything that increases shared understanding improves overall value, IMO.
[12:04] <mamund> i get a lot out of this kind of stuff.
[12:05] <zazi> np, (it's somehow curious that this topic catches me so much appr. one year after the first period of time)
[12:05] <mamund> :)
[12:05] <mamund> stuff like that happens to me quite often
[12:07] <mikekelly> RDF is basically about data
[12:08] <mikekelly> which is fine, but it's definitely got a tonne of baggage most people just don't need
[12:17] <mamund> i continue to think (hope?) that the sweet spot is
[12:18] <mamund> hypermedia for describing protocol interactions ...
[12:18] <mamund> and semantic markup for describing problem domain details.
[12:18] <mamund> we'll see.
[12:18] <darrelmiller> In the same representation?
[12:18] <mamund> i _know_ that hypermedia can ve used to share understanding; it happens all the time today
[12:19] <mamund> not so sure about the semantic markup (i.e. how much is really shared here)
[12:19] <mamund> darrelmiller: either in the same representation or in the same response (i.e. headrs may be needed)
[12:19] <mamund> and the representation might be "compound"...
[12:20] <mikekelly> :|
[12:20] <mamund> as in the say CSS and XSLT are included as pointers in representations today
[12:20] <darrelmiller> I see the following arch. UA <-> HAL/XHTML/etc <-> App Server <-> RDF/JSON/XML <-> Data store.
[12:21] <mamund> darrelmiller: so you see RDF as a server-side tech
[12:21] <mikekelly> :S
[12:21] <darrelmiller> Yeah. Or if you have a completely batch/crawler type of UA.
[12:21] <mikekelly> I think people care way too much about machines.
[12:21] <mikekelly> WAY too much.
[12:21] <mamund> or maybe better to say you see RDF as a data-tech format
[12:22] <mikekelly> all of this mucking around trying to come up with clever mechanisms just makes makes your application harder to consume for 'normal' people
[12:22] <mamund> mikekelly: i will confess that sometimes i suspect "M2M" is used as a kind of "hand-waving" to avoid some ugly details
[12:22] <mikekelly> :)
[12:23] <mamund> the way some might say "forms" in the same case (as a hand-wave to avoid ugly truths)
[12:23] <mamund> :)
[12:23] <mamund> "just use a form"
[12:24] <mamund> "i'll just emit forms and let the client _choose_"
[12:24] <mamund> :)
[12:27] talios (~textual@akl.smx.co.nz) joined #rest.
[13:02] kennethreitz (~kennethre@204.14.152.118) left irc: Quit: Computer has gone to sleep.
[13:24] kennethreitz (~kennethre@204.14.152.118) joined #rest.
[13:29] bigbluehat (~bigblueha@adsl-74-177-98-159.gsp.bellsouth.net) left irc: Quit: Leaving.
[13:46] zazi (~zazi@p579D3696.dip.t-dialin.net) left irc: Quit: HydraIRC -> http://www.hydrairc.com <- Nine out of ten l33t h4x0rz prefer it
[14:00] kennethreitz (~kennethre@204.14.152.118) left irc: Quit: Computer has gone to sleep.
[14:04] kennethreitz (~kennethre@204.14.152.118) joined #rest.
[14:35] Wombert (~Wombert@dslb-092-075-003-110.pools.arcor-ip.net) left irc: Quit: Wombert
[14:56] SvenDowideit (~SvenDowid@203-206-171-38.perm.iinet.net.au) left irc: Changing host
[14:56] SvenDowideit (~SvenDowid@foswiki/developer/SvenDowideit) joined #rest.
[15:09] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) left irc: Remote host closed the connection
[15:09] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) joined #rest.
[15:14] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) left irc: Ping timeout: 252 seconds
[15:42] Wombert (~Wombert@dslb-092-075-003-110.pools.arcor-ip.net) joined #rest.
[15:59] kennethreitz (~kennethre@204.14.152.118) left irc: Quit: Textual IRC Client: http://www.textualapp.com/
[16:16] mamund is done ignoring important things today.
[16:19] <trygvis> then you can read this: http://weknowmemes.com/2012/01/blowy-as-fuck-man/
[17:12] kennethreitz (~kennethre@173-13-176-158-sfba.hfc.comcastbusiness.net) joined #rest.
[17:24] <darrelmiller> I just looked at my blog stats and wondered why it was going nuts. Then I found this... http://www.asp.net/ hehe. Me and the Gu!
[17:25] <whartung> what the hell are you babbling about darrelmiller?
[17:25] <whartung> oic
[17:26] <darrelmiller> The Gu is MS devs equivalent to Jobs. :-)
[17:29] talios takes a break from $work to read up on RESTAS, a Common Lisp web/rest framework - nice.
[17:30] <whartung> linky so I can read about it tomorrow
[17:30] <whartung> I have to go
[17:30] <talios> http://restas.lisper.ru/en/tutorial/hello-world.html
[17:30] <whartung> ty
[17:50] jhajduk (~jhajduk@S0106001ff3c8f310.ed.shawcable.net) joined #rest.
[17:50] jhajduk (~jhajduk@S0106001ff3c8f310.ed.shawcable.net) left #rest.
[18:27] kennethreitz (~kennethre@173-13-176-158-sfba.hfc.comcastbusiness.net) left irc: Quit: Computer has gone to sleep.
[19:53] NOKAH (~hakon1@193.137.202.84.customer.cdi.no) left irc: Quit: Leaving...
[20:47] dkubb (~dkubb@50.92.131.237) joined #rest.
[20:55] talios (~textual@akl.smx.co.nz) left irc: Quit: Textual IRC Client: http://www.textualapp.com/
[22:11] b4284 (~b4283@60-249-196-111.HINET-IP.hinet.net) joined #rest.
[22:12] b4284 (~b4283@60-249-196-111.HINET-IP.hinet.net) left #rest ("WeeChat 0.3.6").
[22:17] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) joined #rest.
[22:19] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) left irc: Remote host closed the connection
[22:19] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) joined #rest.
[22:22] dreinull (~dreinull@ip-78-94-220-161.unitymediagroup.de) left irc: Read error: Operation timed out
[00:00] --- Wed Jan 11 2012