fluid-work IRC Logs-2008-10-06

[07:05:39 EDT(-0400)] * phiggins (n=dante@c-68-34-199-67.hsd1.tn.comcast.net) has joined #fluid-work
[07:05:39 EDT(-0400)] * jacobfarber (n=Jacob@142.150.154.106) has joined #fluid-work
[08:01:51 EDT(-0400)] * anastasiac (n=stasia@h69-128-111-235.mdsnwi.dedicated.static.tds.net) has joined #fluid-work
[08:07:10 EDT(-0400)] * Justin_o (n=Justin@142.150.154.101) has joined #fluid-work
[08:17:16 EDT(-0400)] * Justin_o (n=Justin@142.150.154.101) has left #fluid-work
[08:53:01 EDT(-0400)] * athena7 (n=athena7@75.58.126.207) has joined #fluid-work
[08:53:01 EDT(-0400)] * jrnorman (n=john@ginger.caret.cam.ac.uk) has joined #fluid-work
[08:53:01 EDT(-0400)] * phiggins (n=dante@c-68-34-199-67.hsd1.tn.comcast.net) has joined #fluid-work
[08:53:01 EDT(-0400)] * jacobfarber (n=Jacob@142.150.154.106) has joined #fluid-work
[08:56:21 EDT(-0400)] * michelled (n=michelle@CPE001310472ade-CM0011aefd3ca8.cpe.net.cable.rogers.com) has joined #fluid-work
[08:58:24 EDT(-0400)] * jacobfarber (n=Jacob@142.150.154.106) has joined #fluid-work
[09:14:27 EDT(-0400)] * Justin_o (n=Justin@142.150.154.101) has joined #fluid-work
[09:23:14 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined #fluid-work
[09:24:17 EDT(-0400)] * anastasiac (n=stasia@72.33.23.226) has joined #fluid-work
[09:25:25 EDT(-0400)] * apetro-_ (n=apetro@72.33.23.236) has joined #fluid-work
[09:47:09 EDT(-0400)] * theclown (n=theclown@guiseppi.atrc.utoronto.ca) has joined #fluid-work
[09:58:47 EDT(-0400)] * michelled_ (n=michelle@CPE001310472ade-CM0011aefd3ca8.cpe.net.cable.rogers.com) has joined #fluid-work
[10:09:56 EDT(-0400)] * Justin_o (n=Justin@142.150.154.101) has joined #fluid-work
[10:15:28 EDT(-0400)] * michelled (n=michelle@CPE001310472ade-CM0011aefd3ca8.cpe.net.cable.rogers.com) has joined #fluid-work
[10:18:42 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined #fluid-work
[10:36:08 EDT(-0400)] * jonathan_h (n=Jon@H151.C207.cci.switchworks.net) has joined #fluid-work
[10:37:40 EDT(-0400)] * EricDalquist (n=EricDalq@static-210-59.vpn.wisc.edu) has joined #fluid-work
[10:53:28 EDT(-0400)] * apetro-_ (n=apetro@72.33.23.236) has joined #fluid-work
[11:01:51 EDT(-0400)] * EricDalquist (n=EricDalq@static-210-59.vpn.wisc.edu) has joined #fluid-work
[11:26:54 EDT(-0400)] * ecochran (n=ecochran@adsl-70-137-144-225.dsl.snfc21.sbcglobal.net) has joined #fluid-work
[11:52:47 EDT(-0400)] <michelled> hey everyone, just a reminder that we have the retrospective in 8 minutes
[11:56:56 EDT(-0400)] <michelled> questions to think about:
[11:56:56 EDT(-0400)] <michelled>
[11:56:57 EDT(-0400)] <michelled> * What went well?
[11:56:57 EDT(-0400)] <michelled> * What didn't go well?
[11:56:57 EDT(-0400)] <michelled> * What did we learn?
[11:56:57 EDT(-0400)] <michelled> * What still puzzles us?
[11:57:13 EDT(-0400)] * ecochran (n=ecochran@adsl-70-137-144-225.dsl.snfc21.sbcglobal.net) has joined #fluid-work
[12:00:24 EDT(-0400)] * ecochran_ (n=ecochran@adsl-70-137-147-173.dsl.snfc21.sbcglobal.net) has joined #fluid-work
[12:23:30 EDT(-0400)] * EricDalquist (n=EricDalq@static-210-59.vpn.wisc.edu) has joined #fluid-work
[12:25:18 EDT(-0400)] * apetro (n=apetro@72.33.23.236) has joined #fluid-work
[13:20:14 EDT(-0400)] * ecochran (n=ecochran@adsl-70-137-130-162.dsl.snfc21.sbcglobal.net) has joined #fluid-work
[13:56:30 EDT(-0400)] * lessthanzero (n=lessthan@cmr-208-97-79-45.cr.net.cable.rogers.com) has joined #fluid-work
[14:00:15 EDT(-0400)] * anastasiac (n=stasia@72.33.23.226) has joined #fluid-work
[14:01:02 EDT(-0400)] * apetro (n=apetro@72.33.23.236) has joined #fluid-work
[14:03:59 EDT(-0400)] * EricDalquist (n=EricDalq@static-210-59.vpn.wisc.edu) has joined #fluid-work
[14:04:41 EDT(-0400)] * anastasiac (n=stasia@72.33.23.226) has joined #fluid-work
[14:19:57 EDT(-0400)] <Justin_o> jonathan_h: if you look at the progress indicators using IE6 do you see the bar being very thin
[14:21:55 EDT(-0400)] <jessm> over on sakai list sgithens is asking about where to find generated JS API docs
[14:22:07 EDT(-0400)] <jessm> didn't 0.5 include those somewhere or did i make that up?
[14:29:41 EDT(-0400)] <anastasiac> jessm, there are no 'generated' docs
[14:29:46 EDT(-0400)] <anastasiac> our docs are hand-written
[14:29:57 EDT(-0400)] <anastasiac> we have a 'snapshotted' version
[14:30:00 EDT(-0400)] <jessm> ah, right, got it
[14:30:09 EDT(-0400)] <anastasiac> i.e. something that is versioned to 0.5, and won't be updated as trunk changes
[14:40:17 EDT(-0400)] <jonathan_h> @Justin_o: sorry, I missed your message. I don't have IE6. Does it really loook that bad?
[14:40:40 EDT(-0400)] <Justin_o> sorry... should have just said IE at all
[14:40:47 EDT(-0400)] <Justin_o> and it is bad
[14:41:19 EDT(-0400)] <Justin_o> jonathan_h: ^
[14:43:29 EDT(-0400)] * simonwang (n=chatzill@swang.itservices.ubc.ca) has joined #fluid-work
[14:43:33 EDT(-0400)] <Justin_o> jonathan_h: do you know what might be causing it...
[14:43:57 EDT(-0400)] <Justin_o> simonwang: hello haven't seen you in a while
[14:45:33 EDT(-0400)] <simonwang> Sorry, I did not check in for quite a while. I've been very busy in our operation work.
[14:46:06 EDT(-0400)] <Justin_o> simonwang: yah.. understandable. That's what we thought
[14:47:10 EDT(-0400)] * Bosmon (n=Antranig@ginger.caret.cam.ac.uk) has joined #fluid-work
[14:47:30 EDT(-0400)] <Bosmon> Renderer is now going into the main repo... plus a test
[14:49:16 EDT(-0400)] <jonathan_h> @Justin_o: if you put some text inside a cell, the height plumps up.
[14:49:43 EDT(-0400)] <jonathan_h> Maybe a work-around would be to put a period inside one of the cells to get the height to format properly?
[14:50:14 EDT(-0400)] <Justin_o> thanks jonathan_h ... I'll give that a try...
[14:50:34 EDT(-0400)] <simonwang> Thanks to Jen Bourey, Fluid 0.5 in now in uPortal trunk. This morning I updated Fluid uPortal test site with latest uPortal changes.
[14:51:03 EDT(-0400)] <colinclark> Yes!
[14:51:08 EDT(-0400)] <colinclark> That's awesome.
[14:51:14 EDT(-0400)] <anastasiac> Bosmon, excellent news!
[14:51:27 EDT(-0400)] <colinclark> Hey, simonwang, while you're here, do you have time for a question?
[14:51:41 EDT(-0400)] <anastasiac> simonwang, I'm glad you updated the test site, thanks!
[14:51:53 EDT(-0400)] <anastasiac> so the reorderer is present in the site now, reordering the portlets?
[14:54:58 EDT(-0400)] <simonwang> anastasiac: I'm sorry, I did not have a chance to check.
[14:55:19 EDT(-0400)] <anastasiac> but it was an update from trunk?
[14:55:35 EDT(-0400)] <anastasiac> at this url? http://build.fluidproject.org/uPortal/render.userLayoutRootNode.uP
[15:00:12 EDT(-0400)] <simonwang> Yes, this is the Fluid uPOrtal demo URL. The uportal code is from uportal's trunk HEAD. The Fluid inside uPortal came from uPortal.
[15:00:32 EDT(-0400)] <anastasiac> excellent, thanks, simonwang
[15:00:36 EDT(-0400)] <simonwang> colinclark: sure, what are the questions?
[15:00:52 EDT(-0400)] <anastasiac> athena7, our demo site, http://build.fluidproject.org/uPortal/render.userLayoutRootNode.uP, is using the latest uPortal
[15:01:05 EDT(-0400)] <anastasiac> do you observe the bookmarks portlet problem on that site?
[15:01:45 EDT(-0400)] <athena7> you need to add a mapping for the bookmarks portlet to mod_jk or whatever you're using
[15:01:53 EDT(-0400)] <athena7> right now the bookmarks.js isn't loading
[15:02:05 EDT(-0400)] <athena7> all the forms that are visible in the bookmarks portlet should be hidden when you load up the portlet
[15:02:14 EDT(-0400)] <athena7> brb, phone . . .
[15:02:18 EDT(-0400)] <anastasiac> ah! interesting...
[15:02:47 EDT(-0400)] <simonwang> I will fix it.
[15:02:49 EDT(-0400)] <athena7> yeah - i think most of the portlets won't load up quite right w/o it
[15:02:57 EDT(-0400)] <colinclark> simonwang: We chatted awhile back about my ideas for a new build system for Fluid.
[15:03:06 EDT(-0400)] <colinclark> I'm wondering if you'd be interested in leading that up for Infusion 0.6.
[15:03:25 EDT(-0400)] <colinclark> In short, the idea is to make our JavaScript builds more flexible, allowing users to choose what dependencies are included in the build.
[15:03:36 EDT(-0400)] <colinclark> So for example, they could choose to build only the Reorderer and Inline Edit.
[15:03:51 EDT(-0400)] <colinclark> And it would be great to have the choice of including jQuery or not, for example.
[15:04:23 EDT(-0400)] <colinclark> simonwang: Ideally, we'd even have a "builder" website with an easy-to-use interface for creating a build.
[15:04:47 EDT(-0400)] <colinclark> simonwang: Is this something you're interested in lending a hand with?
[15:05:23 EDT(-0400)] * EricDalquist (n=EricDalq@static-210-59.vpn.wisc.edu) has joined #fluid-work
[15:06:34 EDT(-0400)] <simonwang> I've been working on this task on and off. The including jQuery or not is easy and has been done.
[15:09:10 EDT(-0400)] <simonwang> Because of the current Fluid code tree, to build selected component have to pick files from different directories is not easy. Do you have any suggestions?
[15:09:27 EDT(-0400)] <Bosmon> colinclark, michelled, all: I noticed this comment appeared on fluid.wrap:
[15:09:30 EDT(-0400)] <Bosmon> * Deprecated: please use jQuery to wrap an object - this will be removed in 0.6
[15:09:30 EDT(-0400)] <Bosmon> * Wraps an object in a jQuery if it isn't already one.
[15:09:30 EDT(-0400)] <Bosmon> *
[15:09:56 EDT(-0400)] <Bosmon> I am thinking that this is not a method we can actually deprecate, since it does serve a useful function, which is to have "sensible" behaviour on being asked to wrap nothing
[15:10:03 EDT(-0400)] <Bosmon> Do you think this is reasonable?
[15:10:36 EDT(-0400)] <michelled> The last time I looked at jQuery - which is admittedly a long time ago, I think they were doing the same thing
[15:10:51 EDT(-0400)] <michelled> no, wait, I'm confused
[15:10:52 EDT(-0400)] <Bosmon> Hmm... that doesn't agree with what I generally find
[15:10:55 EDT(-0400)] <michelled> I misread what you wrote
[15:11:01 EDT(-0400)] <Bosmon> Their behaviour on being sent nothing is generally dreadful
[15:11:13 EDT(-0400)] <Bosmon> And amounts to usually randomly returning the document node
[15:11:21 EDT(-0400)] <michelled> right, I remember - it's completely insane
[15:11:24 EDT(-0400)] <Bosmon> Even if you had previously asked for a "scoped" query
[15:11:54 EDT(-0400)] <michelled> Colin had actually pinged me on why I deprecated it but I think we got side tracked.
[15:12:00 EDT(-0400)] <Bosmon> ok
[15:12:07 EDT(-0400)] <Bosmon> Consider yourself repinged
[15:12:09 EDT(-0400)] <michelled> I think you're right, we should remove the deprecation warning
[15:12:27 EDT(-0400)] <Bosmon> ok
[15:12:38 EDT(-0400)] <Bosmon> Do we think this is fine for a NOJIRA:
[15:12:40 EDT(-0400)] <Bosmon> ?
[15:12:42 EDT(-0400)] * athena7 is back
[15:13:03 EDT(-0400)] <michelled> ya, but please put a comment in saying why we need fluid.wrap
[15:13:39 EDT(-0400)] <Bosmon> ok
[15:13:59 EDT(-0400)] * apetro (n=apetro@72.33.23.236) has joined #fluid-work
[15:14:18 EDT(-0400)] <athena7> simonwang: did what i said above before disappearing make sense?
[15:19:56 EDT(-0400)] <simonwang> The Fluid uPortal demo site is a combined work. I set up the current one based on other's work. I will keep that in mind when I fix the issue.
[15:23:11 EDT(-0400)] <anastasiac> simonwang and athena7, I have updated the mod_jk stuff for the fluid build server, and restarted apache
[15:23:26 EDT(-0400)] <anastasiac> now, the bookmarks portlet seems (to me) to be correct
[15:23:31 EDT(-0400)] <anastasiac> athena7, would you agree?
[15:23:38 EDT(-0400)] <simonwang> I noticed this! Thanks
[15:23:55 EDT(-0400)] <athena7> yes! thank you
[15:23:56 EDT(-0400)] <anastasiac> I'm seeing the fields invisible now, and hence appearing in the avatar
[15:24:08 EDT(-0400)] <athena7> yes, that's the behavior i was seeing before
[15:24:17 EDT(-0400)] <athena7> forms suddenly reappear while dragging, then go away when i let go
[15:24:25 EDT(-0400)] <anastasiac> wonderful! Now, maybe we can figure out what's going wrong
[15:24:36 EDT(-0400)] <Bosmon> forms "reappear"?
[15:24:49 EDT(-0400)] <athena7> er, appear
[15:24:49 EDT(-0400)] <Bosmon> Were these then things that were styled as "invisible" in the original portlet?
[15:24:50 EDT(-0400)] <athena7> no re
[15:24:51 EDT(-0400)] <athena7>
[15:24:56 EDT(-0400)] <athena7> yes
[15:25:01 EDT(-0400)] <Bosmon> How had that happened?
[15:25:05 EDT(-0400)] <Bosmon> I think I have a good idea why it is
[15:25:14 EDT(-0400)] <anastasiac> Bosmon, please do share!
[15:25:26 EDT(-0400)] <athena7> forms with css attributes of "display:none"
[15:25:28 EDT(-0400)] <Bosmon> Is it possible that they are styled as invisible by means of some kind of scoped selector?
[15:25:37 EDT(-0400)] <anastasiac> ah...
[15:25:41 EDT(-0400)] <Bosmon> Right, but how was display:none bound to them via a CSS rule?
[15:25:57 EDT(-0400)] <athena7> hm
[15:26:04 EDT(-0400)] <Bosmon> We made a change late on in 0.5 to position the avatar at the base of the document per Jacob's suggestion
[15:26:07 EDT(-0400)] <athena7> looks like they've got a class="hidden"
[15:26:09 EDT(-0400)] * anastasiac is starting to see a bit more clearly
[15:26:20 EDT(-0400)] <Bosmon> Rather than dumping them into the same container managed by the reorderer
[15:26:36 EDT(-0400)] * athena7 then there's #bookmarksPortlet .hidden
[15:26:36 EDT(-0400)] * athena7 display:none;
[15:26:37 EDT(-0400)] * athena7 }
[15:26:42 EDT(-0400)] <anastasiac> so because the avatar is not scoped as the portlet was, the style is not being applied?
[15:26:43 EDT(-0400)] <Bosmon> Aha, there you go
[15:27:00 EDT(-0400)] <Bosmon> They are no longer nested within the portlet, whilst in the avatar state
[15:27:12 EDT(-0400)] <Bosmon> I think we are now thinking that this was the wrong thing to do, but in the meantime you will have to workaround
[15:27:41 EDT(-0400)] <anastasiac> right - the avatar creator clones the portlet, but strips the ID
[15:27:47 EDT(-0400)] <anastasiac> so the style is no longer applicable
[15:27:52 EDT(-0400)] <Bosmon> Well, it's not so much that it strips the id
[15:28:05 EDT(-0400)] <Bosmon> It's that when the avatar is "live" it is not positioned within the same overall container
[15:28:32 EDT(-0400)] <anastasiac> I'm not sure it's containment - the style is ID based: "#bookmarksPortlet.hidden"
[15:28:33 EDT(-0400)] <athena7> ah ok, that makes sense
[15:29:12 EDT(-0400)] <Bosmon> This is an issue that we will need to revisit, if we ever mean to solve http://issues.fluidproject.org/browse/FLUID-398
[15:29:41 EDT(-0400)] <Bosmon> anastasiac: not quite
[15:29:47 EDT(-0400)] <Bosmon> it is actually "#bookmarksPortlet .hidden"
[15:29:51 EDT(-0400)] <Bosmon> Note the space which is very important
[15:30:17 EDT(-0400)] <anastasiac> right - it's scoped to something that is inside an element with that id
[15:30:26 EDT(-0400)] <anastasiac> the avatar doesn't have that id
[15:30:38 EDT(-0400)] <Bosmon> right
[15:30:47 EDT(-0400)] <Bosmon> Anyway, the workaround is probably a bit unfortunate
[15:31:18 EDT(-0400)] <Bosmon> We should talk to Jacob whether we really want to reverse the effects of http://issues.fluidproject.org/browse/FLUID-1141
[15:31:25 EDT(-0400)] <Bosmon> This was the JIRA which is responsible for this current problem
[15:31:44 EDT(-0400)] <anastasiac> I don't think that changing where we append it will help the hidden problem
[15:31:52 EDT(-0400)] <anastasiac> doesn't the space imply containment?
[15:32:06 EDT(-0400)] <Bosmon> If we reversed FLUID-1141, the problem would be solved
[15:32:07 EDT(-0400)] <anastasiac> i.e. the space in the style selector?
[15:32:13 EDT(-0400)] <Bosmon> Since we would be placing the avatar within the right scope
[15:32:20 EDT(-0400)] <Bosmon> That is, nested within #bookmarksPortlet
[15:32:26 EDT(-0400)] <colinclark> I've pinged Mr. Farber...
[15:32:36 EDT(-0400)] <colinclark> He should be on his way into the channel shortly.
[15:32:42 EDT(-0400)] <jacobfarber>
[15:32:55 EDT(-0400)] <anastasiac> oh - I didn't remember that the avatar had previously been placed inside the portlet - I thought it was placed adjacent to the portlet
[15:32:57 EDT(-0400)] <Bosmon> Come paint this issue with your golden Farb
[15:33:24 EDT(-0400)] <jacobfarber> Bosmon: i cant tell if thats an insult or a compliment
[15:33:31 EDT(-0400)] <Bosmon> Oh, sorry, I thought the # was the id of the overall container...
[15:33:42 EDT(-0400)] <colinclark> where do I get a golden farb?
[15:33:58 EDT(-0400)] <anastasiac> no, that's the id of the particular portlet in question
[15:33:59 EDT(-0400)] <jacobfarber> Hopefully no-where!
[15:33:59 EDT(-0400)] <jacobfarber> whats the effect your thinking of reversing?
[15:34:06 EDT(-0400)] <Bosmon> Well, FLUID-1141
[15:34:09 EDT(-0400)] <jacobfarber> oh
[15:34:11 EDT(-0400)] <Bosmon> We had a bit of a chat whilst I was there
[15:34:17 EDT(-0400)] <Bosmon> And decided it was "probably wrong"
[15:34:26 EDT(-0400)] <jacobfarber> ahh
[15:34:28 EDT(-0400)] <Bosmon> But it seemed too precipitous to change it back so close to the release
[15:34:28 EDT(-0400)] <jacobfarber> i remember this
[15:34:39 EDT(-0400)] <jacobfarber> yech
[15:35:05 EDT(-0400)] <Bosmon> anastasiac: OK, well if the style weren't scoped to just that individual portlet, it could at least be scoped to something
[15:35:12 EDT(-0400)] <Bosmon> Right now with 1141, it can't be scoped to anything at all
[15:35:17 EDT(-0400)] <anastasiac> agreed
[15:35:21 EDT(-0400)] <jacobfarber> unless!
[15:35:25 EDT(-0400)] <Bosmon> UNLESS!!
[15:35:33 EDT(-0400)] <anastasiac> unless?
[15:35:33 EDT(-0400)] <jacobfarber> unless we build a little fake tree of css
[15:35:35 EDT(-0400)] * anastasiac leans in
[15:35:37 EDT(-0400)] <jacobfarber> which would be crazy
[15:35:51 EDT(-0400)] <Bosmon> EXELEUTHEREOSTOMIZE, to the court, Mr. Farber!
[15:36:14 EDT(-0400)] <jacobfarber> hold on, googling "EXELEUTHEREOSTOMIZE"
[15:36:33 EDT(-0400)] <Bosmon> Sorry, I was using the British spelling
[15:36:40 EDT(-0400)] <jacobfarber> No definitions were found for EXELEUTHEREOSTOMIZE.
[15:36:42 EDT(-0400)] <Bosmon> I am seriously impressed that Google can figure out how to correct that....
[15:36:56 EDT(-0400)] <jacobfarber> No definitions were found for EXELEUTHEROSTOMISE
[15:36:58 EDT(-0400)] <Bosmon> "Did you mean: EXELEUTHEROSTOMISE?" is what I got
[15:37:05 EDT(-0400)] <Bosmon> You must be using a different Google to me...
[15:37:05 EDT(-0400)] <colinclark> lol
[15:37:25 EDT(-0400)] <jacobfarber> please put me out of my misery?
[15:37:30 EDT(-0400)] <Bosmon> Anyway, we are getting off topic
[15:37:34 EDT(-0400)] <jacobfarber> right
[15:37:35 EDT(-0400)] <jacobfarber> so
[15:37:42 EDT(-0400)] <jacobfarber> i was thinking a bit about this one
[15:37:44 EDT(-0400)] <Bosmon> You don't really think it is reasonable to build a little fake tree of css, right?
[15:37:49 EDT(-0400)] <Bosmon> That was just you horsing about?
[15:37:53 EDT(-0400)] <anastasiac> what would the fake tree look like?
[15:37:58 EDT(-0400)] <jacobfarber> no, it would be a huge problem
[15:38:02 EDT(-0400)] <jacobfarber> like you said way back
[15:38:03 EDT(-0400)] <Bosmon> ok
[15:38:09 EDT(-0400)] <jacobfarber> replicating the entire computed style
[15:38:14 EDT(-0400)] <Bosmon> Gah
[15:38:16 EDT(-0400)] <jacobfarber> would take a large amout of time
[15:38:23 EDT(-0400)] <anastasiac> narf!
[15:38:32 EDT(-0400)] <Bosmon> Anyway, yes, if we want to solve http://issues.fluidproject.org/browse/FLUID-398 we will have to do this all very differently anyway
[15:38:40 EDT(-0400)] <Bosmon> We will have to have a "roving avatar"
[15:39:03 EDT(-0400)] <Bosmon> In the meantime, we need at least something semi-sane to suggest to athena7 for her problem
[15:39:17 EDT(-0400)] <anastasiac> a work-around could be this:
[15:39:18 EDT(-0400)] <Bosmon> My suggestion is that she falls back to class-based scoping for this style
[15:39:19 EDT(-0400)] <jacobfarber> bosmon: what was her problem?
[15:39:29 EDT(-0400)] <Bosmon> Things that were formerly invisible in her portlet became visible in the avatar
[15:39:35 EDT(-0400)] <jacobfarber> oh
[15:39:37 EDT(-0400)] <athena7> right, we could ask eric to change the portlet
[15:39:44 EDT(-0400)] <jacobfarber> she could fix that easily, like you said
[15:39:45 EDT(-0400)] <athena7> i suspect this will come up in other places though
[15:39:54 EDT(-0400)] <Bosmon> She hat a style like ***athena7 then there's #bookmarksPortlet .hidden
[15:39:59 EDT(-0400)] <anastasiac> another option could be this:
[15:40:19 EDT(-0400)] <anastasiac> uportal could override the default avatar creator function to replace the id of the avatar, rather than erase it
[15:40:26 EDT(-0400)] <anastasiac> and then the stylesheet could include a style for the avatar
[15:40:30 EDT(-0400)] <jacobfarber> why not, like you said, just add an additional declaration within the orderable-avatar *
[15:40:54 EDT(-0400)] <Bosmon> athena7: is there currently no suitable enclosing class in the markup for this portlet?
[15:41:00 EDT(-0400)] <jacobfarber> there is
[15:41:01 EDT(-0400)] <anastasiac> jacobfarber, yes, you may be right
[15:41:12 EDT(-0400)] <anastasiac> we apply orderable-avatar to the avatar right?
[15:41:18 EDT(-0400)] <jacobfarber> this shouldnt be an elaborate fix!
[15:41:35 EDT(-0400)] <jacobfarber> just take the auto-applied class name
[15:41:37 EDT(-0400)] <athena7> Bosmon: i'm not completely familiar with the portlet, but looking at the build server, it looks like "bookmarksPortlet" is meant to be the enclosing div for the portlet
[15:41:42 EDT(-0400)] <anastasiac> then if they add ".orderable-avatar .hidden" as a selector for the hidden style...
[15:41:55 EDT(-0400)] <jacobfarber> and add the original hidden declarations so its withing THAT scope as well
[15:42:04 EDT(-0400)] <Bosmon> Yeah, we apply .orderable-avatar to the avatar
[15:42:16 EDT(-0400)] <Bosmon> But this doesn't seem like a completely scalable solution...
[15:42:24 EDT(-0400)] <athena7> i don't think we can really ask people to add fluid-specific styles or workarounds to their portlets
[15:42:33 EDT(-0400)] <jacobfarber> but you shouldnt!
[15:42:41 EDT(-0400)] <jacobfarber> thats the reason you can re-define it
[15:42:42 EDT(-0400)] <Bosmon> It seems to imply taking all styles scoped anywhere in the CSS to nested markup, and applying them to .orderable-avatar
[15:42:45 EDT(-0400)] <jacobfarber> so in other words
[15:43:02 EDT(-0400)] <jacobfarber> overwrite or add to .orderable-avatar with your existing class
[15:43:20 EDT(-0400)] <Bosmon> Yeah, but the issue is that every portlet is different
[15:43:43 EDT(-0400)] <athena7> right
[15:43:44 EDT(-0400)] <jacobfarber> can you not take the common declarations and dump them into a common class?
[15:43:48 EDT(-0400)] <anastasiac> jacobfarber, I think that what athena7 is saying, though, is that they don't want to have to ask any portlet creators (who are independent) to have to consider fluid class names in their stylesheets
[15:43:53 EDT(-0400)] <athena7> right
[15:44:02 EDT(-0400)] <athena7> and many portlets may not even be written w/ uportal in mind
[15:44:13 EDT(-0400)] <Bosmon> Well
[15:44:14 EDT(-0400)] <athena7> the goal is to be able to deploy reasonable JSR-168 compliant portlets
[15:44:27 EDT(-0400)] <Bosmon> as a "general recommendation" I think it is reasonable to ask that they not use id-based style rules
[15:44:35 EDT(-0400)] <Bosmon> These are typically quite fragile
[15:44:47 EDT(-0400)] <Bosmon> And, for example, prohibit more than one instance of something ever appearing on a page
[15:44:58 EDT(-0400)] <Bosmon> But of course we can recommend all we like
[15:45:08 EDT(-0400)] <athena7> actually id-based styles in portlets aren't good anyway
[15:45:09 EDT(-0400)] <Bosmon> We still have this problem...
[15:45:13 EDT(-0400)] <jacobfarber> I think thats all you can do at this point
[15:45:14 EDT(-0400)] <ecochran> yes, a good rule...
[15:45:26 EDT(-0400)] <athena7> because that means if you add a portlet twice you could have multiple ids on a page
[15:45:33 EDT(-0400)] <Bosmon> Yes
[15:45:41 EDT(-0400)] <Bosmon> We recently saw the kinds of mayhem that can cause
[15:45:45 EDT(-0400)] <athena7> so that may not be a totally unreasonable recommendation
[15:46:07 EDT(-0400)] <athena7> but there is a "safe" way to namespace ids, so it's not totally inconceivable that someone would go that route
[15:46:15 EDT(-0400)] <Bosmon> Well, you can namespace them
[15:46:21 EDT(-0400)] <Bosmon> But you still can't prevent them from having to be unique
[15:46:22 EDT(-0400)] <athena7> though it's sort of inconvenient to namespace a css file, so it will probably be less common
[15:47:08 EDT(-0400)] <Bosmon> I mean, you can't use this to deal with the "I can't create 2 of this kind of thing" issue
[15:47:14 EDT(-0400)] <athena7> so is the recommendation for now to ask Eric to change the CSS, and recommend that people not use id-based styles?
[15:47:17 EDT(-0400)] <Bosmon> You can only use it to deal with the "some sod has stolen my id" issue
[15:47:42 EDT(-0400)] <Bosmon> Well, that seems best as a "long-term" solution, yes...
[15:47:44 EDT(-0400)] <athena7> no there's a portlet:namespace tag that will generate different namespaces even for the same portlet
[15:48:07 EDT(-0400)] <athena7> well we haven't released up3 yet, so it could be a now-solution too, i guess
[15:48:11 EDT(-0400)] <Bosmon> athena7: But in that case, how can you generate a CSS style which hits this namespaced id?
[15:48:22 EDT(-0400)] <Bosmon> Does it "munge" the CSS file at the same time?
[15:48:59 EDT(-0400)] <athena7> no, that's why i was saying people wouldn't generally try to write css against namespaced ids
[15:49:02 EDT(-0400)] <athena7> because it would be a pain
[15:49:06 EDT(-0400)] <Bosmon> Right
[15:49:09 EDT(-0400)] <Bosmon> It spells Global Doom
[15:49:13 EDT(-0400)] <athena7> yes
[15:49:19 EDT(-0400)] <athena7> particularly if you have an external css file
[15:49:25 EDT(-0400)] <athena7> pretty easy to do if the css is inline
[15:49:58 EDT(-0400)] <athena7> so i don't think it's probably crazy unreasonable to recommend people not do all that
[15:50:06 EDT(-0400)] <Bosmon> aha!
[15:50:16 EDT(-0400)] <Bosmon> I think I now see why you didn't have this problem in the previous release
[15:50:30 EDT(-0400)] <Bosmon> No, wait, I don't...
[15:50:37 EDT(-0400)] <athena7> the previous fluid release?
[15:50:40 EDT(-0400)] <Bosmon> Well anyway, I saw this now commented-out code
[15:50:43 EDT(-0400)] <jacobfarber> athena7, do you have a link I could see?
[15:50:47 EDT(-0400)] <Bosmon> // $(":hidden", avatar).remove();
[15:50:52 EDT(-0400)] <athena7> we did have the same problem with the post-0.4 release we first used, actually
[15:50:53 EDT(-0400)] <Bosmon> Yes, the previous fluid release
[15:50:55 EDT(-0400)] <athena7> this isn't new as of 0.5
[15:50:56 EDT(-0400)] <Bosmon> Oh, ok
[15:50:59 EDT(-0400)] <Bosmon> That's fine then
[15:51:04 EDT(-0400)] <anastasiac> jacobfarber: http://build.fluidproject.org/uPortal/render.userLayoutRootNode.uP
[15:51:04 EDT(-0400)] <Bosmon> At least, for some values of fine
[15:51:09 EDT(-0400)] <athena7> jacobfarber: http://build.fluidproject.org/uPortal/render.userLayoutRootNode.uP
[15:51:15 EDT(-0400)] <athena7> it's on the front page onece you log in [15:51:23 EDT(-0400)] <anastasiac> jacobfarber, try dragging the bookmarks portlet [15:51:29 EDT(-0400)] <Bosmon> I was thinking this was a rule to remove physically from the cloned material anything which was "instantaneously hidden" [15:51:30 EDT(-0400)] <athena7> i think we'd have time to change the bookmarks portlet before the uportal 3.1 release [15:51:37 EDT(-0400)] <Bosmon> But anyway, this is fantastically expensive to run [15:51:42 EDT(-0400)] <jacobfarber> thanks anastasiac [15:52:13 EDT(-0400)] <jacobfarber> dragging => completely dead? [15:52:52 EDT(-0400)] <anastasiac> no, dragging causes hidden fields to show up in the avatar [15:53:37 EDT(-0400)] <jacobfarber> well [15:53:43 EDT(-0400)] <jacobfarber> i see what you mean [15:54:01 EDT(-0400)] <anastasiac> [15:54:07 EDT(-0400)] <jacobfarber> and all I could think of, like everyone said here already, is to slightly adjust the scope [15:54:14 EDT(-0400)] <jacobfarber> so its not disenganged when dragging [15:54:36 EDT(-0400)] <anastasiac> right - the question is how [15:55:21 EDT(-0400)] <jacobfarber> how to scope styles to be more flexible? [15:55:50 EDT(-0400)] <Bosmon> Tell people to use class names [15:56:12 EDT(-0400)] <jacobfarber> something I dont understand though [15:56:19 EDT(-0400)] <jacobfarber> is the #bookmarklets id [15:56:24 EDT(-0400)] <jacobfarber> should still work [15:56:33 EDT(-0400)] <jacobfarber> even when disengaged [15:56:39 EDT(-0400)] <Bosmon> wot? [15:57:54 EDT(-0400)] <jacobfarber> i dont understand why "#bookmarksPortlet" dissappears in the avatar....is it being stripped? [15:58:07 EDT(-0400)] <Bosmon> Of course it is bloody being stripped! [15:58:15 EDT(-0400)] <Bosmon> Because we don't want two nodes in the same document with the same id... [15:58:28 EDT(-0400)] <jacobfarber> but [15:58:40 EDT(-0400)] <jacobfarber> you dont need to to strip EVERY id do you? [15:58:43 EDT(-0400)] <Bosmon> Yes [15:58:45 EDT(-0400)] * athena7 awards Bosmon 10 points for enthusiasm [15:58:49 EDT(-0400)] <jacobfarber> just the ones your acting upon [15:58:56 EDT(-0400)] <Bosmon> Every id, by definition, is already the id of something already in the document! [15:59:01 EDT(-0400)] <Bosmon> Since it is part of something we are cloning [15:59:04 EDT(-0400)] <jacobfarber> yes, for JS [15:59:10 EDT(-0400)] <Bosmon> Therefore, it is to be stripped... [15:59:12 EDT(-0400)] <jacobfarber> but thats not the be-all and end-all for ids [15:59:27 EDT(-0400)] <Bosmon> I fear you may be verging close onto blasphemy [15:59:35 EDT(-0400)] <jacobfarber> hear me out [15:59:38 EDT(-0400)] <anastasiac> jacobfarber, the avatar is a clone of the portlet, so the id on the avatar is stripped so that it's not the same as the portlet [15:59:48 EDT(-0400)] <Bosmon> Are you suggesting there are some circumstances in which it might be appropriate to have two DOM nodes in the same document with the same id? [15:59:48 EDT(-0400)] <jacobfarber> i understand [15:59:53 EDT(-0400)] <jacobfarber> but heres the catch! [15:59:58 EDT(-0400)] <jacobfarber> YES [16:00:01 EDT(-0400)] <Bosmon> Hmm [16:00:03 EDT(-0400)] <jacobfarber> thats exactly what im saying [16:00:07 EDT(-0400)] <colinclark> Bosmon: Let him type. [16:00:07 EDT(-0400)] <jacobfarber> heres how [16:00:09 EDT(-0400)] <Bosmon> We need to discover a lightning bolt from somewhere [16:00:14 EDT(-0400)] <jacobfarber> ! [16:00:34 EDT(-0400)] <jacobfarber> ok, for DOM traversal, is is IMPOSSIBLE to work with clashing ID's [16:00:41 EDT(-0400)] <jacobfarber> you run amok of everything [16:00:47 EDT(-0400)] <Bosmon> Most things that were once farbed, like Greek Architecture, eventually weathered away under the influence of the Elements... [16:01:35 EDT(-0400)] <jacobfarber> Farbing aside for a moment clashing ID's have no place in DOM operations, but they work just fine in the workd of CSS and in fact [16:01:43 EDT(-0400)] <jacobfarber> they're important for cascadence [16:01:51 EDT(-0400)] <jacobfarber> so [16:01:52 EDT(-0400)] <Bosmon> What on earth [16:02:00 EDT(-0400)] <Bosmon> We can't possibly advocate creating of broken documents [16:02:04 EDT(-0400)] <jacobfarber> no no [16:02:19 EDT(-0400)] <Bosmon> Just because things "generally seem to work right" when the victims of a CSS styling engine [16:02:20 EDT(-0400)] <jacobfarber> but this doesnt need to become OUR problem (which it is now) [16:02:35 EDT(-0400)] <jacobfarber> Bosmon: hold on [16:02:39 EDT(-0400)] <anastasiac> gotta go to a session - back soon [16:03:08 EDT(-0400)] <jacobfarber> so here's the deal - we need to care less about the markup [16:03:19 EDT(-0400)] <jacobfarber> instead of the heavy handed approach of re-writing it [16:03:26 EDT(-0400)] <Bosmon> .... [16:03:33 EDT(-0400)] <jacobfarber> instead of stripping ALL ids we're concerned with [16:03:45 EDT(-0400)] <jacobfarber> we need to just remove the ones used in our selector systems [16:03:50 EDT(-0400)] <Bosmon> No, that is not correct [16:04:07 EDT(-0400)] <Bosmon> Precipitating a document with duplicate IDs violates our contract with the browser [16:04:09 EDT(-0400)] <jacobfarber> why would an id we never touch break our JS? [16:04:16 EDT(-0400)] <Bosmon> It doesn't care whether we are "concerned with them" or not [16:04:30 EDT(-0400)] <Bosmon> As soon as we hand the browser a broken document, it can do what the hell it likes with us [16:04:43 EDT(-0400)] <jacobfarber> ? [16:04:51 EDT(-0400)] <Bosmon> That is, even more than it currently does [16:05:13 EDT(-0400)] <jacobfarber> time for a little testing, methinks! [16:05:17 EDT(-0400)] <Bosmon> ...... [16:05:52 EDT(-0400)] <Bosmon> I am not convinced that is a good use of your time [16:06:00 EDT(-0400)] <jacobfarber> spare time [16:06:15 EDT(-0400)] <jacobfarber> gotta make a theme right now - pretty graphics and all [16:06:15 EDT(-0400)] <Bosmon> There is simply no way were are going to have code in the framework that encourages or abets the creation of duplicate ids in any way [16:06:41 EDT(-0400)] <Bosmon> It is one of those things which are a Global Crime [16:06:54 EDT(-0400)] <Bosmon> What makes you think you are the only person traversing the document looking for ids? [16:07:07 EDT(-0400)] <jacobfarber> i know what you mean [16:07:08 EDT(-0400)] <jacobfarber> but [16:07:11 EDT(-0400)] <Bosmon> It is simply irresponsible to think that just because you "think" you aren't using the id, that it is ok for it to be a duplicate [16:07:13 EDT(-0400)] <jacobfarber> thats not the point [16:07:18 EDT(-0400)] <Bosmon> ........ [16:07:29 EDT(-0400)] <Bosmon> I'm not sure whether it is the point [16:07:37 EDT(-0400)] <Bosmon> But I feel that it is a somewhat unanswerable point [16:07:46 EDT(-0400)] <jacobfarber> ok, [16:07:48 EDT(-0400)] <jacobfarber> in a way [16:07:49 EDT(-0400)] <jacobfarber> yes [16:08:06 EDT(-0400)] <jacobfarber> i had to slow you down with a crazy comment [16:08:15 EDT(-0400)] <jacobfarber> you type very quickly [16:09:58 EDT(-0400)] <jacobfarber> I see what you mean, but without cloning thevisual properties of the IDs from a css point of view, you are forcing the implementor to re-write their code, no? [16:10:02 EDT(-0400)] <jacobfarber> Thats a problem too [16:10:31 EDT(-0400)] <jacobfarber> i think working with classes is FAR better, which I agree with you [16:10:51 EDT(-0400)] <jacobfarber> (the skinning stuff works this way) but the other approach is very heavy handed too [16:10:55 EDT(-0400)] <Bosmon> Well... there is no way we can spare the implementor from the consequences of their decisions here [16:11:06 EDT(-0400)] <jacobfarber> [16:11:39 EDT(-0400)] <Bosmon> I know browsers maximally encourage the "hack and hope" mentality [16:11:46 EDT(-0400)] <Bosmon> But I feel sure that we have to draw the line somewhere [16:12:01 EDT(-0400)] <Bosmon> If they want duplicate ids, they had better write their own avatar creation strategy [16:12:08 EDT(-0400)] <jacobfarber> lol [16:12:45 EDT(-0400)] <Bosmon> They could quite easily break the document for everyone [16:12:51 EDT(-0400)] <Bosmon> Not just the searches that might be in error... [16:13:08 EDT(-0400)] <Bosmon> I'm sure you've seen some of the many ways that whole documents can go "bad milk" in IE [16:13:15 EDT(-0400)] * anastasiac (n=stasia@72.33.23.226) has joined #fluid-work [16:13:49 EDT(-0400)] <jacobfarber> then I think this important but muted policy is critical to communicate [16:14:43 EDT(-0400)] <Bosmon> Yes [16:14:48 EDT(-0400)] <jacobfarber> so people dont have to re-write all their code! [16:15:13 EDT(-0400)] <Bosmon> An important role of Fluid is to encourage good practice [16:15:30 EDT(-0400)] <jacobfarber> but back to athena7 though..... [16:15:45 EDT(-0400)] <jacobfarber> oh wait [16:15:47 EDT(-0400)] <Bosmon> Yes [16:16:01 EDT(-0400)] <Bosmon> For this use case, I can't see there is any option but to dig up a class name from somewhere [16:16:02 EDT(-0400)] <jacobfarber> i think I answered my own question [16:16:05 EDT(-0400)] <Bosmon> At least, for this release [16:16:10 EDT(-0400)] <jacobfarber> yeah [16:16:27 EDT(-0400)] <Bosmon> If there isn't one "roughly" in the right place, it shouldn't be too hard to invent one [16:16:39 EDT(-0400)] <Bosmon> Via styles.avatar or otherwise [16:17:39 EDT(-0400)] <jacobfarber> thats what I was wondering - why they wouldnt have set up a class name hook for themesleves already [16:17:48 EDT(-0400)] <jacobfarber> to tap into to avoid all this [16:33:36 EDT(-0400)] <michelled> anastasiac: how'd the lightning talk go? [16:33:54 EDT(-0400)] * theclown (n=theclown@guiseppi.atrc.utoronto.ca) has joined #fluid-work [16:34:11 EDT(-0400)] <anastasiac> didn't get to a fluid lightening talk this morning, but there was lots of interest in fluid, so allison and I led a one-hour session on fluid [16:34:27 EDT(-0400)] <jessm> anastasiac: super cool! [16:49:17 EDT(-0400)] * EricDalquist (n=EricDalq@static-210-59.vpn.wisc.edu) has joined #fluid-work [16:51:14 EDT(-0400)] <Bosmon> jacobfarber: Sorry, didn't mean to come across as too antagonistic in that exchange.. [16:51:24 EDT(-0400)] <Bosmon> Some issues can strike me as particularly obvious examples of "morality"... [16:51:36 EDT(-0400)] <jacobfarber> no, I agree with you [16:52:01 EDT(-0400)] <jacobfarber> i would rather put an ID through my lungs before making 2 of em [16:58:46 EDT(-0400)] * athena7 tries to decide how extreme of a statement that is [16:58:52 EDT(-0400)] <athena7> i'm just not sure how much an ID hurts . . . [17:01:29 EDT(-0400)] <jacobfarber> they're pretty pointy [17:01:54 EDT(-0400)] <jacobfarber> IE6 broke and it splintered into my eye [17:02:07 EDT(-0400)] <athena7> ow now that does sound painful [17:03:45 EDT(-0400)] <jacobfarber> IE6 is now serving time for assault, so its ok [17:08:30 EDT(-0400)] <michelled> if only that meant that people would stop using it [17:09:46 EDT(-0400)] <jacobfarber> sadly, IE6's parents have a lot of money and keep bailing him out of trouble. I think his older siblings want to kill him though next time he gets out. [17:10:40 EDT(-0400)] <michelled> [17:12:38 EDT(-0400)] <jacobfarber> next, on True Hollywood Stories: The Browser That Refuses To Die [17:13:26 EDT(-0400)] <athena7> lol [17:13:55 EDT(-0400)] <jacobfarber> nerd humor at its worst [17:33:00 EDT(-0400)] * apetro (n=apetro@72.33.23.236) has joined #fluid-work [17:34:35 EDT(-0400)] * theclown (n=theclown@guiseppi.atrc.utoronto.ca) has left #fluid-work [17:36:03 EDT(-0400)] * phiggins (n=dante@c-68-34-199-67.hsd1.tn.comcast.net) has joined #fluid-work [17:39:59 EDT(-0400)] * michelled (n=michelle@CPE001310472ade-CM0011aefd3ca8.cpe.net.cable.rogers.com) has left #fluid-work [17:51:59 EDT(-0400)] * phiggins (n=dante@c-68-34-199-67.hsd1.tn.comcast.net) has joined #fluid-work [18:15:40 EDT(-0400)] * EricDalquist (n=EricDalq@static-210-59.vpn.wisc.edu) has joined #fluid-work [18:52:12 EDT(-0400)] * ecochran (n=ecochran@adsl-70-137-130-162.dsl.snfc21.sbcglobal.net) has joined #fluid-work [19:09:43 EDT(-0400)] * EricDalquist (n=EricDalq@h69-128-111-235.mdsnwi.dedicated.static.tds.net) has joined #fluid-work [22:23:42 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined #fluid-work [22:25:08 EDT(-0400)] * jessm_ (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined #fluid-work