fluid-work IRC Logs-2011-10-27
[08:47:55 CDT(-0500)] <jhung> mlam, I'm finding a bug in the Uploader...
[08:48:10 CDT(-0500)] * jhung oops. Habit.
[13:18:00 CDT(-0500)] <Bosmon4> Hi cindyli - are you there?
[13:18:42 CDT(-0500)] <cindyli> yes, Bosmon4
[13:19:15 CDT(-0500)] <Bosmon4> How are you getting on?
[13:19:31 CDT(-0500)] <Bosmon4> Thanks for your mail of yesterday - I think this suggests that it is time to return to our original plan for "new UIO"
[13:19:43 CDT(-0500)] <cindyli> still stuck at the status in my last email
[13:19:53 CDT(-0500)] <Bosmon4> That is, I think the slider issue can only be properly resolved if you return jQuery (and also jQueryUI) into code included within the iframe
[13:20:09 CDT(-0500)] <cindyli> oh wo
[13:20:15 CDT(-0500)] <cindyli> ok
[13:20:21 CDT(-0500)] <Bosmon4> Then you will need to inject the new jQuery into the component, and build a bridge so that the renderer can accept a new option "jQuery"
[13:20:42 CDT(-0500)] <Bosmon4> This will be the jQuery object which the renderer uses when it comes to honour a "jQuery decorator"
[13:21:31 CDT(-0500)] <cindyli> ic. does framework need to provide a new option for jQuery injection
[13:21:35 CDT(-0500)] <cindyli> or it's been there?
[13:21:41 CDT(-0500)] <Bosmon4> It will need a new option
[13:21:56 CDT(-0500)] <Bosmon4> Please feel free to make a JIRA for this and include it in your branch
[13:22:08 CDT(-0500)] <cindyli> sure
[13:22:18 CDT(-0500)] <cindyli> include it in my branch?
[13:22:24 CDT(-0500)] <Bosmon4> Yes
[13:22:28 CDT(-0500)] <Bosmon4> That is, to do it
[13:22:49 CDT(-0500)] <cindyli> to do it? resolve the new jira?
[13:22:55 CDT(-0500)] <cindyli> adding the new option into framework?
[13:22:59 CDT(-0500)] <cindyli> in my branch?
[13:23:06 CDT(-0500)] <cindyli> i'm frightened
[13:23:08 CDT(-0500)] <Bosmon4>
[13:23:10 CDT(-0500)] <cindyli> :-P
[13:23:10 CDT(-0500)] <Bosmon4> It will be fine
[13:23:14 CDT(-0500)] <Bosmon4> Your code will get reviewed
[13:23:29 CDT(-0500)] <cindyli> oh, my. ok, i will try
[13:24:07 CDT(-0500)] <Bosmon4>
[13:59:56 CDT(-0500)] <michelled> jameswy: the ILDH seems to have an old name and a new one
[14:00:07 CDT(-0500)] <michelled> should the official name have the word 'Design' in it?
[14:00:33 CDT(-0500)] <michelled> jameswy: anastasiac is doing some clean up and consolidation and we want to get the name right
[14:15:52 CDT(-0500)] <jameswy> michelled, anastasiac: use ILDH, not ILH.
[14:20:55 CDT(-0500)] <colinclark> michelled: Definitely Inclusive Learning Design Handbook, as jameswy says
[14:21:07 CDT(-0500)] <colinclark> jhung: What was the Uploader bug you discovered this morning?
[14:21:34 CDT(-0500)] <michelled> colinclark: what about the capitalization of floe? all caps? first cap? no caps?
[14:21:34 CDT(-0500)] <colinclark> cindyli: It's exciting news to hear that you're going to add a new feature to the framework!
[14:21:58 CDT(-0500)] <colinclark> We are united against acronyms and excess capitalization
[14:22:00 CDT(-0500)] <cindyli> yes, exciting and frightening, colinclark
[14:22:13 CDT(-0500)] <colinclark> michelled: Most definitely "Floe"
[14:22:19 CDT(-0500)] <colinclark> cindyli: You've got tons of support, don't worry
[14:22:26 CDT(-0500)] <michelled> thx
[14:22:38 CDT(-0500)] <colinclark> Give it a try, see how it goes, and you'll inevitably get lots of help
[14:22:39 CDT(-0500)] <cindyli> cool. thanks, colinclark
[14:23:24 CDT(-0500)] <michelled> colinclark: anastasiac and I were looking at the workflow for the ILDH. There is currently some stuff that has been pushed to the development branch that isn't quite working. I was thinking of the development branch as a staging area before we get to master - is that not the plan?
[14:23:50 CDT(-0500)] <michelled> colinclark: we are just about ready to push the UIO integration but I'm wondering what to do about the non working functionality in the dev branch
[14:24:18 CDT(-0500)] <michelled> my inclination is to remove it, push the UIO work to dev and then push the dev branch to master
[14:25:00 CDT(-0500)] <colinclark> Ah, good question, michelled
[14:25:08 CDT(-0500)] <colinclark> I think that our approach to dev sites is going to evolve a bit
[14:25:12 CDT(-0500)] <colinclark> so now's a good time to think through this
[14:25:25 CDT(-0500)] <colinclark> I don't have all the background on the plans for the work within the dev version of the ILDH...
[14:25:43 CDT(-0500)] <colinclark> but my sense was that it was an ad-hoc experimentation ground for Heidi to start looking at new design ideas from jameswy
[14:25:58 CDT(-0500)] <colinclark> Unfortunately, even Heidi wasn't quite sure what the state of that branch was, or what its relationship to the main site was
[14:26:09 CDT(-0500)] <colinclark> So, I think it's up to us to go ahead and make a sensible workflow
[14:26:22 CDT(-0500)] <colinclark> What's the non-working functionality?
[14:26:28 CDT(-0500)] <michelled> anastasiac and I did a diff - it's just the pdf generation stuff that differs from the master as far as we can tell
[14:26:49 CDT(-0500)] <michelled> there are 3 plugins related to pdf generation and a non-working button on the UI
[14:28:39 CDT(-0500)] <jhung> colinclark: actually wasn't an uploader bug. I was just missing a classname on a container. Everything is good now.
[14:28:51 CDT(-0500)] <colinclark>
[14:31:09 CDT(-0500)] <colinclark> michelled: Do you have any idea what it would take to get PDF generation working?
[14:31:13 CDT(-0500)] <colinclark> And does ePub generation work okay?
[14:32:12 CDT(-0500)] <michelled> colinclark: I have no idea
[14:32:21 CDT(-0500)] <colinclark> So was your idea to just take it out?
[14:32:27 CDT(-0500)] <michelled> I haven't looked into it at all - only looked at what was different in the branch
[14:32:35 CDT(-0500)] <colinclark> ok
[14:32:38 CDT(-0500)] <michelled> yes, my thoughts were to pull it out - maybe into another branch?
[14:32:45 CDT(-0500)] <michelled> and push up the UIO integration
[14:32:50 CDT(-0500)] <colinclark> That makes sense to me
[14:32:53 CDT(-0500)] <michelled> get that in and on the live site
[14:33:06 CDT(-0500)] <colinclark> Just to be clear, I guess we can distinguish between two notions of "development space"
[14:33:22 CDT(-0500)] <colinclark> There's an actual running instance on our server, pointing to the development branch on Github
[14:33:25 CDT(-0500)] <colinclark> and then there's the branch itself
[14:33:39 CDT(-0500)] <colinclark> I'm still waiting on huslage to move directories around
[14:34:02 CDT(-0500)] <colinclark> but, in short, I think we should treat the dev branch as the place where we have working code that we want to test and share with others
[14:34:14 CDT(-0500)] <colinclark> but that is not yet necessarily ready for production deployment
[14:34:23 CDT(-0500)] <colinclark> So the development workflow would look something like this
[14:34:56 CDT(-0500)] <michelled> colinclark: I thought the 'dev' site had no actual relation to anything in github
[14:35:11 CDT(-0500)] <colinclark> Work on a feature, in a feature branch in the repository -> When it's baked, merge the feature branch into the development branch, and pull those changes to the dev site -> Once it's solid, merge to master and pull the changes to the production site
[14:35:15 CDT(-0500)] <colinclark> michelled: That's not quite true
[14:35:26 CDT(-0500)] <colinclark> Or rather, briefly, that's true
[14:35:42 CDT(-0500)] <colinclark> We just need huslage to move the production and development sites over to the git-backed directories
[14:36:01 CDT(-0500)] <michelled> ah, good
[14:36:29 CDT(-0500)] <colinclark> Does this all make sense?
[14:36:33 CDT(-0500)] <michelled> colinclark: so I guess the question right now is what to do when there are two baking features. does it make sense to pull one out to let the other through the whole pipe?
[14:36:43 CDT(-0500)] <colinclark> yes
[14:36:54 CDT(-0500)] <colinclark> I think we should develop features in feature branches, and when they're reasonably baked, merge them into development
[14:37:24 CDT(-0500)] <colinclark> So, as you say, the "development" branch is really the "staging" phase of the worflow
[14:37:37 CDT(-0500)] <michelled> so in an ideal case we wouldn't need to pull a feature out - it would be basically ready to go in by the time it reached the dev branch
[14:40:04 CDT(-0500)] <colinclark> yes
[14:40:11 CDT(-0500)] <colinclark> This was something we couldn't do in the past
[14:40:17 CDT(-0500)] <colinclark> since we did all of our active development on the server
[14:40:27 CDT(-0500)] <colinclark> which means inevitably some things would be broken at any given time
[14:41:05 CDT(-0500)] <colinclark> Whereas using Git and developing locally takes a bit more work to get an environment set up, but then allows us much more flexibility to roll out features when they're ready
[14:41:53 CDT(-0500)] <colinclark> I can even imagine that once we hire our systems developer, we'll be able to do things like create reusable virtual machines that are all set up for local development
[14:42:02 CDT(-0500)] <colinclark> even further lowering the barrier to development entry
[14:44:06 CDT(-0500)] <michelled> sounds great
[14:44:39 CDT(-0500)] <michelled> anastasiac has put together some starter docs on what she did to get her system ready for working on the ILDH so we've got a start anyway
[14:47:57 CDT(-0500)] <colinclark> oh, that's great
[14:49:13 CDT(-0500)] <colinclark> Hey, I've been hacking on something very sketchy while at the conference that people might be interested in
[14:49:27 CDT(-0500)] <colinclark> A prototype of some developer tools for working with Infusion
[14:49:43 CDT(-0500)] <colinclark> Helping with the process of generating the boilerplate configuration as well as injecting components into a page
[14:49:44 CDT(-0500)] <colinclark> https://github.com/colinbdclark/infusion-authoring
[14:49:59 CDT(-0500)] <colinclark> Right now I'm using a really nice code editor widget called CodeMirror
[14:50:11 CDT(-0500)] <colinclark> eventually we might imagine integrating it into Cloud9 or TextMate or other things
[14:51:28 CDT(-0500)] <colinclark> right now all it does is inject a standard file closure into the editor
[14:52:44 CDT(-0500)] <colinclark> but I'm close to the ability to generate component configuration
[14:52:47 CDT(-0500)] <colinclark> eventually demands blocks, too
[14:56:34 CDT(-0500)] <michelled> ah, it will be so nice to have demands blocks generated - don't you think yura?
[14:58:18 CDT(-0500)] <yura> yes that would be interesting to see
[15:58:01 CDT(-0500)] <yura> colinclark: minute for a question ?
[15:58:09 CDT(-0500)] <colinclark> it's worth a shot
[15:58:16 CDT(-0500)] <colinclark> I dunno if I'll have an answer
[15:58:18 CDT(-0500)] <yura> single file uploader
[15:58:26 CDT(-0500)] <colinclark> ok
[15:58:28 CDT(-0500)] <yura> template for it is outside of the container
[15:58:38 CDT(-0500)] <yura> how does the uploader pick it up when it degrades?
[16:03:18 CDT(-0500)] <colinclark> Unfortuantely in a somewhat presumptuous way
[16:03:23 CDT(-0500)] <colinclark> lemme check the code
[16:03:35 CDT(-0500)] <colinclark> but I think it assumes where it will be located, and that there will be only one
[16:06:37 CDT(-0500)] <colinclark> yura: https://github.com/fluid-project/infusion/blob/master/src/webapp/components/uploader/js/Uploader.js#L734-739
[16:06:59 CDT(-0500)] <michelled> colinclark: I just pushed anastasiac's branch which puts UIO into ILDH to the development branch. I feel it's ready for prime time. I'm not sure what the process should now be. Perhaps you need to push it up to master and rebuild?
[16:07:08 CDT(-0500)] <colinclark> It looks for the first element with a .fl-progEnhance-basic, yura
[16:07:31 CDT(-0500)] <colinclark> michelled: When you say prime time, do you mean the actual production site?
[16:07:33 CDT(-0500)] <yura> yes but within the container ?
[16:07:43 CDT(-0500)] <colinclark> no
[16:07:45 CDT(-0500)] <colinclark> anywhere in the document
[16:07:54 CDT(-0500)] <colinclark> This is the key line: https://github.com/fluid-project/infusion/blob/master/src/webapp/components/uploader/js/Uploader.js#L737
[16:08:02 CDT(-0500)] <yura> right
[16:08:04 CDT(-0500)] <colinclark> Where you can see I rather lamely just use jQ instead of the DOM Binder
[16:08:04 CDT(-0500)] <michelled> colinclark: I think so - is there anything else you know that would hold us back? I suppose a review from one of the designers might make sense …
[16:08:24 CDT(-0500)] <colinclark> yeah it would be great for jameswy or tina or joanna to take a look
[16:08:26 CDT(-0500)] <michelled> colinclark: I'll ping jameswy tomorrow
[16:08:31 CDT(-0500)] <colinclark> And then the process is this:
[16:08:46 CDT(-0500)] <colinclark> 1. Get huslage to update the server to the git-backed versions
[16:08:48 CDT(-0500)] <yura> thanks, colinclark
[16:09:04 CDT(-0500)] <colinclark> 2. ssh to the server and pull changes down from master into the production repo
[16:09:05 CDT(-0500)] <colinclark> that's it
[16:09:49 CDT(-0500)] <colinclark> "there is no step three"
[16:10:52 CDT(-0500)] <colinclark> so, michelled, can you explain one thing?
[16:10:58 CDT(-0500)] <michelled> yes
[16:11:09 CDT(-0500)] <colinclark> If I remember, the dev branch was quite different from the master
[16:11:13 CDT(-0500)] <colinclark> with a different theme, etc.
[16:11:26 CDT(-0500)] <colinclark> so UIO is integrated into that new theme?
[16:11:30 CDT(-0500)] <anastasiac> michelled, colinclark: jameswy did have a look at the visuals, and he thought it was ok as in interim step.
[16:11:46 CDT(-0500)] <colinclark> great!
[16:11:53 CDT(-0500)] <michelled> colinclark: yes, it's integrated into that theme
[16:12:11 CDT(-0500)] <colinclark> that's really cool
[16:12:13 CDT(-0500)] <colinclark> okay
[16:12:19 CDT(-0500)] <colinclark> so once huslage gets things into place
[16:12:27 CDT(-0500)] <colinclark> we can figure out who should be on our ops team while heidi is away
[16:12:39 CDT(-0500)] <colinclark> and then whoever that is can just pull it down onto the dev site
[16:12:44 CDT(-0500)] <colinclark> we'll test it out a bit
[16:12:50 CDT(-0500)] <colinclark> if it looks good, we'll pull it to the production site
[16:12:55 CDT(-0500)] <colinclark> does that sound right?
[16:14:53 CDT(-0500)] <michelled> yes, that makes sense
[16:15:51 CDT(-0500)] <colinclark> michelled: are you interested in serving on the ops team for a bit until we hire someone to do that role?
[16:16:01 CDT(-0500)] <colinclark> I think mostly it would involve having an ssh key registered with the servers
[16:16:08 CDT(-0500)] <colinclark> and occasionally do some pushing and pulling with git
[16:16:18 CDT(-0500)] <colinclark> no system administration or anything like that
[16:16:51 CDT(-0500)] <michelled> sure - I can take that on colinclark
[16:16:59 CDT(-0500)] <colinclark> thanks
[16:17:12 CDT(-0500)] <colinclark> when you get a chance, ping huslage and get him to register your ssh key
[16:17:18 CDT(-0500)] <colinclark> i'll add you to the group in github
[16:17:25 CDT(-0500)] <michelled> thx
[16:19:21 CDT(-0500)] <colinclark> michelled: https://github.com/organizations/fluid-project/teams/100501
[16:22:23 CDT(-0500)] <michelled> thx