15:00:48 RRSAgent has joined #xproc 15:00:48 logging to http://www.w3.org/2006/04/27-xproc-irc 15:00:48 ok, Norm; that matches XML_PMWG()11:00AM 15:00:56 zakim, ? is richard 15:00:56 +richard; got it 15:01:11 Norm has joined #xproc 15:01:23 zakim, who's on the phone? 15:01:23 On the phone I see Alex_Milowski, Alessandro_Vernet, Norm, PGrosso, richard 15:02:25 Meeting: XML Processing Model WG 15:02:25 Scribe: Norm 15:02:25 ScribeNick: Norm 15:02:25 Date: 27 Apr 2006 15:02:25 Chair: Norm 15:02:26 Agenda: http://www.w3.org/XML/XProc/2006/04/27-agenda.html 15:02:32 +MoZ 15:02:42 zakim, who's on the phone? 15:02:42 On the phone I see Alex_Milowski, Alessandro_Vernet, Norm, PGrosso, richard, MoZ 15:03:07 MSM? 15:03:46 zakim, who's talking 15:03:46 I don't understand 'who's talking', Norm 15:03:50 zakim, please call Micahel-Office 15:03:50 I am sorry, MSM; I do not know a number for Micahel-Office 15:03:52 zakim, who's talking? 15:03:56 Zakim, mute me 15:03:56 MoZ should now be muted 15:04:02 Norm, listening for 10 seconds I heard sound from the following: Norm (90%), MoZ (15%) 15:04:07 zakim, please call ht-781 15:04:07 ok, ht; the call is being made 15:04:09 +Ht 15:04:09 zakim, please call Michael-Office 15:04:09 ok, MSM; the call is being made 15:04:11 +Michael 15:04:54 zakim does time-sharing? 15:05:02 zakim, who's on the phone? 15:05:02 On the phone I see Alex_Milowski, Alessandro_Vernet, Norm, PGrosso, richard, MoZ (muted), Ht, Michael 15:05:31 I *believe* it means that Z heard input from Norm during 90% of its time slices, and from MoZ during 15% of them 15:05:44 Present: Alex, Alessandro, Norm, Paul, Richard, Mohamed, Henry, Michael 15:05:49 Regrets: Murray, Andrew 15:05:55 Topic: Accept this agenda? 15:05:55 -> http://www.w3.org/XML/XProc/2006/04/27-agenda.html 15:06:05 Accepted 15:06:08 Topic: Accept minutes from the previous teleconference? 15:06:08 -> http://www.w3.org/XML/XProc/2006/04/20-minutes.html 15:06:17 Accepted 15:06:18 Topic: Next meeting: 4 May telcon 15:06:18 Any regrets? 15:06:29 yes 15:06:53 yes 15:07:27 Zakim, unmute me 15:07:27 MoZ should no longer be muted 15:07:37 Face-to-face meeting? 15:08:06 Registration page: http://www.w3.org/2002/09/wbs/38398/XProcFTF2/ 15:08:47 Topic: Issue 3096: Are components side-effect free? 15:08:47 -> http://www.w3.org/Bugs/Public/show_bug.cgi?id=3096 15:09:55 Norm proposes: 15:09:56 I propose that we say that all components are non-functional. That is, 15:09:56 a pipeline implementation must behave as if it evaluated a component 15:09:56 every time it occurs. "Must behave as if" is spec-ease for 15:09:56 "implementations that are clever enough to determine with certainty 15:09:56 that a component is, in fact, functional are free to cache the 15:09:58 intermediate results because by golly if it is, no one will be able to 15:10:41 Richard: This doesn't preclude adding a mechanism later to allow authors to assert that a step or component is functional 15:10:45 Norm: Yes. 15:11:24 -Alessandro_Vernet 15:11:47 Richard: Does this address the converse case? Producing output side-effects and behaving the same way for given inputs 15:12:12 +Alessandro_Vernet 15:12:46 Norm: This is the "functional" aspect, not the side-effect aspect 15:13:01 Richard: Side-effects are like hidden outputs, functionality is like hidden inpus 15:13:04 s/inpus/inputs/ 15:14:09 Alex: This is a good place to start, register a new issue about functional components? 15:14:21 ACTION: Alex to create such an issue 15:14:42 s/such an issue/an issue about the possibility of functional components/ 15:15:00 Proposal accepted. 15:15:11 Issue 3113: Are components side-effect free? 15:15:11 -> http://www.w3.org/Bugs/Public/show_bug.cgi?id=3113 15:15:25 s/Are components side-effect free?/Does the pipeline engine act as a resource manager? 15:16:10 Norm: One aspect of this question is, does the pipeline engine provide the sort of URI-stability that XSLT, for example, gives the document function 15:16:56 Richard: I strongly disagree with this as a requirement; it requires a degree of intimacy between the engine and the components that may not always be available 15:16:59 +Murray_Maloney 15:17:17 Alex: Is this something that might be "at user option" 15:17:41 Norm: I'd like to avoid that if at all possible 15:17:44 q+ to push back 15:18:00 ack ht 15:18:00 ht, you wanted to push back 15:18:23 Henry: I need some information; in my current state of knowledge I think it's a bad idea for pipeline engines 15:18:33 Alessandro has joined #xproc 15:18:55 Henry: Especially when you are running a pipeline engine as a server, you do not want to flush the cache everytime you run a pipe because it's useful to keep things around. 15:19:10 Henry: In their parsed and ready-to-go state (provided they haven't changed) 15:19:47 Alessandro has joined #xproc 15:19:53 Henry: I'm happier saying, "no, you should expect your pipeline to behave in the way of any other web application does" 15:19:57 Henry: Yes, things can change. 15:20:41 Alex: If we step back and look at the web browser case, consider an image embedded 10 times on the page. The browser reuses the image. 15:20:55 s/the page/the same page/ 15:21:23 Alex: The resolution of URI-to-resource is stable for the duration of a page is one reasonable expectation 15:21:48 [I think the fact that browser do or do not re-fetch is an optimization they make, not part of the specification of correct browser behavior - am I wrong?] 15:22:27 Richard: consider other things like XML pipelines, like shell scripts, where "cat foo" twice might not return the same file. 15:22:48 [If ten elements in the same HTML document refer to "my_image.jpg", and that image is served with a lifetime of 0, are correct browsers guaranteed to fetch it only once? What spec says that?] 15:23:29 Some discussion of whether or not browsers actually behave that way 15:23:53 q+ Murray 15:23:54 q+ to say that as an empirical statement, it's not a very strong argument for making the behavior part of our spec 15:23:58 ack msm 15:23:58 MSM, you wanted to say that as an empirical statement, it's not a very strong argument for making the behavior part of our spec 15:24:43 MSM: Implementors will do that for performance reasons regardless of whether a spec requires it or not 15:24:54 Richard: Is there a spec for how you display things in a web page? 15:24:56 Alex: No 15:25:06 MSM: In that sense, it's not clear to me that the browser analogy bears on our decision 15:25:20 Alex: There's a user expectation of some aspect of stability 15:26:05 Richard: I don't think the browser analogy is a good one. The engine is running a collection of potentially independently implemented components. 15:26:09 ack Murray 15:26:47 Murray: I'm relying on my memory, but in HTTP there's a mechanism for specifying time-to-live. So if there's a nano-second TTL, then maybe it would go get the resource again. 15:27:17 Murray: Similarly, if I was getting the time of day from a URI then it might change 15:27:29 Murray: So if you're worried about that, maybe you need a "caching" component. 15:27:39 q+ 15:28:43 ack Alessandro 15:28:45 ack alexmilowski 15:29:04 Norm: I think consensus is coming towards the answer "no" 15:29:18 Alex: I don't agree, I think it's important that URIs are stable for the duration of an execution 15:29:28 -Michael 15:29:49 Alex: If you need to identify unique resources, you can generate unique URIs with query parameters 15:30:15 Alex: We haven't decided if the resources flowing through the pipeline have URIs or not 15:30:18 q+ 15:30:19 zakim, please call Michael-Office 15:30:19 ok, MSM; the call is being made 15:30:20 +Michael 15:30:27 ack richard 15:30:53 Richard: I notice that the bug is actually talking about something produced by the pipeline 15:31:35 Norm: I think those are the same case 15:31:45 Richard: You could provide components that fetch and store URIs stably. 15:34:18 q+ Murray 15:34:37 Norm describes the situation where an XSLT needs to get an ancillary resource by URI 15:35:02 Alex: I really want some URIs to be stable throughout the duration of a pipeline 15:35:14 ack murray 15:35:23 Murray: I'm not convinced that we don't need a resource manager 15:35:47 Murray: I'd like to posit the existence of a component that is a proxy server or something of that ilk 15:36:14 Murray: That component knows if requests should always send things back from the cache 15:37:00 [I agree fervently that as users we need resource managers, and that implementations of our language will be more usable if they use good resource managers. But we also need character sets. We don't specify a character set as part of our spec to meet that need, and the same should probably hold for resource management. Separate problem, separate spec.] 15:37:12 Murray: I think it's the case that sometimes you're going to want the documents to remain stable and sometimes you're going to want to get current results 15:37:15 yes! 15:38:06 Richard: But I may be using components that don't know how to use a proxy server 15:38:54 Murray: I thought once you setup a proxy, then all requests went through that proxy. 15:39:01 That's implementation and operating-system dependent 15:40:24 Richard: Proxies do give a degree of generality that seems nice 15:40:34 q+ 15:42:51 ack msm 15:43:37 MSM: I'm not sure I'm understanding everything going on here. I agree that being able to cache and being able to gaurantee up-to-date resources are good things 15:43:57 MSM: But lots of these things seem to be not terribly closely related to pipelines any more than we need a character set. 15:44:15 MSM: We just rely on getting character sets from lower layers. 15:44:47 q+ 15:44:48 MSM: Building it into the pipeline engine strikes me as a breach of orthogonality. 15:45:37 MSM: At least for the components that we require an implmentation provide, we can say what the answers are or say that they're implementation defined 15:46:26 Murray: I think you're thinking of it in terms of the pipeline language and not the overall processing model. If you're processing large volumes of XML, you may want a proxy server that has access to pipeline descriptions so that all your documents can be passed through. 15:46:55 Beware of assuming that everything comes through HTTP. What if they're just files? 15:47:14 Indeed. The proxy has to handle file: URIs as well. 15:48:15 MSM: It should be orthogonal. If I've got a caching proxy installed, I want my pipeline engine to use that one, not one that it felt it needed to build in. 15:48:28 ack alexmilowski 15:49:05 Alex: The document function in XSLT gets the resource through the local environment that might use a local cache 15:49:26 q+ 15:49:38 MSM: The only thing the XSLT language says is that if you call the document function with the same URI, you'll get the same document 15:49:58 Alex: You want to be able to compare the objects you get back from the document function. 15:50:17 Alex: Do we really have the requirement that things behave this way across components? 15:51:17 ack richard 15:51:53 Richard: I think that Alex has drawn attention to an important point. XSLT can do this because it only says the document function behaves this way. 15:52:11 Richard: Are we really going to say that if the stylesheet is a file: URI then it can't just open it? 15:52:48 Murray: In a shell script, you'd handle this by copying it and then referring to the copy. 15:53:08 Richard: Yes, and if you were using a program that had the name hard coded, then you couldn't make it use the copy 15:53:58 Norm attempts to summarize the consensus which remains "no" 15:54:11 HT: The discussion we've had has been drawn somewhat more narrowly than the first sentenc of the actual issue. 15:54:56 [I wonder if there is consensus on the proposition that in cases like the example given by Norm in raising the issue, it *is* our responsibility to say whether the data stream written to uri Foo is or is not guaranteed the same as the data stream (later) read from uri Foo] 15:55:17 HT: We've discussed in the past the use of pipeline engines as resource managers. 15:55:30 HT: Consider output="#foo" somewhere and input="#foo" somewhere else in a pipeline. 15:55:50 HT: One way to think about that is that the engine is managing those resources. 15:56:04 HT: I don't believe that issue is off the table because of this discussion 15:56:34 Norm: I agree 15:56:50 Alessandro has joined #xproc 15:57:52 I'm a little puzzled / troubled here. If I interpret output="#foo" and input="#foo" as references to resources to be managed by the pipeline, then I suddenly have an ambiguity I didn't use to have: 15:58:07 does the input stream read the ouptut stream? 15:58:22 or is this a pipeline which reads resource #foo, does something with it, and writes it back? 15:59:25 Scribe lost the thread 15:59:27 ht, I wonder if you can expound on how you would propose avoiding this ambiguity 16:00:12 So I think Richard just expressed the dichotomy in an interesting way -- do we name ports, or infosets 16:00:36 +1: Richard's formulation of the question is an acute one 16:00:53 ADJOURNED 16:00:54 -Murray_Maloney 16:00:55 -Ht 16:00:57 -richard 16:00:58 -PGrosso 16:00:59 -Norm 16:01:00 -Alessandro_Vernet 16:01:01 -MoZ 16:01:02 -Alex_Milowski 16:01:03 alexmilowski has left #xproc 16:01:07 -Michael 16:01:08 XML_PMWG()11:00AM has ended 16:01:10 Attendees were Alex_Milowski, Alessandro_Vernet, Norm, PGrosso, richard, MoZ, Ht, Michael, Murray_Maloney 16:01:19 rrsagent, set logs world-visible 16:01:36 rrsagent, draft minutes 16:01:36 I have made the request to generate http://www.w3.org/2006/04/27-xproc-minutes.html Norm 16:04:55 PGrosso has left #xproc 17:58:04 Zakim has left #xproc 20:12:23 MSM? 20:12:29 rrsagent, bye 20:12:29 I see 1 open action item saved in http://www.w3.org/2006/04/27-xproc-actions.rdf : 20:12:29 ACTION: Alex to create such an issue [1] 20:12:29 recorded in http://www.w3.org/2006/04/27-xproc-irc#T15-14-21