fluid-tech IRC Logs-2013-05-07

[13:09:56 CDT(-0500)] <colinclark> Bosmon and I were just discussing this thread on the Mozilla lists about abolishing native MathML in Firefox: https://groups.google.com/forum/?fromgroups=#!topic/mozilla.dev.platform/96dZw1jXTvM

[13:10:03 CDT(-0500)] <colinclark> He was just saying that MathML isn't very good

[13:10:18 CDT(-0500)] <colinclark> And I was going to ask, "In what way is it not very good, in your mind?"

[13:14:26 CDT(-0500)] <colinclark> "* In particular, the entire digital textbook market and thus the entire educational sector comes out of XML/MathML workflows right now. "

[13:14:28 CDT(-0500)] <colinclark> quite interesting

[13:29:39 CDT(-0500)] <Bosmon> Hi cindyli1 - those are some quite sophisticated bugs you have posted

[13:30:00 CDT(-0500)] <cindyli1> yay!

[13:31:49 CDT(-0500)] <Bosmon> They are actually instances of a separate bug that I haven't finished writing up yet, but which is related to FLUID-4925 and FLUID-4982

[13:32:09 CDT(-0500)] <Bosmon> A description could be "Cannot influence the set of subcomponents of a component once instantiation has started"

[13:32:30 CDT(-0500)] <Bosmon> In general I think you should leave the template loading issues aside for this first work of yours

[13:32:49 CDT(-0500)] <Bosmon> And just concentrate on getting the uiOptions.mapOptions system replaced by Skywalker options

[13:33:28 CDT(-0500)] <Bosmon> cindyli1 - if you can get to the point where UIOptions is free of all calls to initView and manual options mapping, that will be good enough for a push

[13:34:54 CDT(-0500)] <cindyli1> i see, Bosmon, however, 5012 is a stopper at this point since all our default settings are passed in thru grades such as "fluid.uiOptions.defaultTemplateLoader". they won't get merged into the the target templateLoader cuz of 5012

[13:35:30 CDT(-0500)] <cindyli1> we can live with 5013 at this point

[13:35:36 CDT(-0500)] <Bosmon> Ok, I see

[13:35:53 CDT(-0500)] <Bosmon> Yes, it may be possible to deal with 5012 without 5013

[13:35:56 CDT(-0500)] <Bosmon> Thanks for the test cases!

[13:36:05 CDT(-0500)] <cindyli1> my pleasure

[13:40:34 CDT(-0500)] <yzen> Bosmon: !

[15:17:08 CDT(-0500)] <Bosmon> yzen E'EVICH!

[15:17:27 CDT(-0500)] <yzen> Bosmon: hi

[15:17:41 CDT(-0500)] <yzen> lets take about page enhancer

[15:17:45 CDT(-0500)] <yzen> talk

[15:18:03 CDT(-0500)] <Bosmon> yzen - say on

[15:18:09 CDT(-0500)] <yzen> or rather page enhancer and relation to a settings store

[15:18:43 CDT(-0500)] <yzen> is that correct to say that page enhancer requires a settings store to operate ?

[15:18:59 CDT(-0500)] <yzen> Bosmon: ^

[15:20:51 CDT(-0500)] <Bosmon> yzen - yes, it is

[15:21:03 CDT(-0500)] <Bosmon> We have to imagine everything requires a settings store to operate

[15:21:07 CDT(-0500)] <Bosmon> Otherwise.... how would it get its settings!

[15:21:09 CDT(-0500)] <yzen> GREAT

[15:21:41 CDT(-0500)] <yzen> this leads to an earlier conversation michelled, anastasiac and I had today about a global settings store possibility , similar to the page enhancer . Your thoughts ?

[15:26:32 CDT(-0500)] <Bosmon> yzen - Can you be more specific!

[15:26:50 CDT(-0500)] <Bosmon> In what sense would a store be similar to the page enhancer!

[15:27:11 CDT(-0500)] <Bosmon> Would it be somewhat in the way that a pint of milk is similar to a cow? : P

[15:27:45 CDT(-0500)] <yzen> Bosmon: sure so we faced the same issue that pretty much everything requires the settings store. we were thinking that it might make sense to have a settings store in a static environment..

[15:28:30 CDT(-0500)] <Bosmon> yzen - the page enhancer's settings store would certainly have to have the same scope as the page enhancer

[15:28:46 CDT(-0500)] <Bosmon> You don't have a choice about that

[15:29:06 CDT(-0500)] <Bosmon> But when you say "pretty much everything requires the settings store" I think you mean "pretty much everything requires A settings store"

[15:29:17 CDT(-0500)] <yzen> Bosmon: sure

[15:30:29 CDT(-0500)] <yzen> Bosmon: and i think that it's safe to say they all need the same settings store. if that's so, you think it is reasonable to have it in static env?

[15:30:50 CDT(-0500)] <Bosmon> yzen - do they indeed all need the same settings store?

[15:30:58 CDT(-0500)] <Bosmon> I feel that the UIO UI itself has its own one

[15:31:15 CDT(-0500)] <Bosmon> Otherwise it would be unable to operate the reset/apply semantic that it has

[15:31:21 CDT(-0500)] <Bosmon> So there are at least 2 settings stores

[15:32:02 CDT(-0500)] <yzen> Bosmon: could you elaborate why a second settings store is needed for reset/apply ?

[15:33:29 CDT(-0500)] <Bosmon> yzen - under some UIO integration models, the settings only affect a "preview" pane and do not affect the ultimate pageEnhancer itself, wherever it is

[15:34:12 CDT(-0500)] <Bosmon> And I think there are even some "main" integration scenarios where the UIO UI settings do not change the main page until they are explicity committed with "Apply"

[15:34:34 CDT(-0500)] <Bosmon> So its own settings store may or may not be the same one held by the pageEnhancer

[15:40:01 CDT(-0500)] <yzen> Bosmon: i was thinking you are talking about the DOM scope of the enhancer here where if there are 2 enhancers they will have different models, however settings store is somewhat separate from the model , isnt that the intention ?

[15:40:48 CDT(-0500)] <Bosmon> yzen - I'm not sure it is the intention for the settings store to necessarily be separate from the model

[15:41:12 CDT(-0500)] <Bosmon> Especially since as we have it now, it is the settings store itself which operates the transfers of state which are requested by the operations RESET/APPLY/CANCEL

[15:41:20 CDT(-0500)] <Bosmon> Of course we could change that through a refactoring

[15:41:31 CDT(-0500)] <yzen> Bosmon: i am changing it

[15:41:40 CDT(-0500)] <yzen> that's the whole purpose of this conversation

[15:41:55 CDT(-0500)] <Bosmon> good good

[15:42:15 CDT(-0500)] <Bosmon> Please publish your draft API and planned semantic to a mailing list posting

[15:42:28 CDT(-0500)] <yzen> Bosmon: lol

[15:43:10 CDT(-0500)] <Bosmon> I should say..... "kindly" do so : P

[15:45:46 CDT(-0500)] <Bosmon> yzen - given that the "Settings Store" was always just a "DataSource" there's no reason we can't honour reset/apply by use of a "local store" configured as a cache

[15:47:33 CDT(-0500)] <yzen> Bosmon: fyi this is what i was looking at, dont be too harsh https://github.com/yzen/infusion/compare/fluid-project:master...FLUID-4686

[15:49:55 CDT(-0500)] <yzen> Bosmon: ok ill ask/propose settings store stuff on the list, another question

[15:50:03 CDT(-0500)] <yzen> default site settings

[15:50:34 CDT(-0500)] <yzen> we were talking about it previously and were inclined to keep them in the uiOptions defaults rather than the settings store defaults

[15:50:44 CDT(-0500)] <yzen> it works for all use cases but 1

[15:52:57 CDT(-0500)] <yzen> i see lots of tests where page enhancer draws from the defaults action ants grade

[15:53:15 CDT(-0500)] <yzen> this grade grabs default values for its enachtors from the default model

[15:53:24 CDT(-0500)] <yzen> which i currently have in uioptions

[15:53:46 CDT(-0500)] <yzen> which obviously is not correct when page enhancer is present without the uioptions

[15:54:53 CDT(-0500)] <yzen> Bosmon: I was wondering where do the default site settings (until we deal with schema) really belong. Unless of course page enhancer in real life never uses them for its enactors