YUI Library Open RoundTable, 19 Sep 2013

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: YUI Open Roundtable September 19, 2013

Description

YUI Open Roundtable September 19, 2013

A

Everybody to this week's yy open round table it's thursday september 19th 2013 and I'm so glad you're here whether you're watching this later or on irc. I'm glad to see I should post this on there.

A

um So things off, we've got a number of things, sort of light on uh on content in terms of just having people doing talks, we got a lot of information, though, to dispense um and we had a potential demo from derek yeah yeah. I constructed some istanbul stuff cool, um so we've got. uh Let me take a look at the page here.

A

I can just go through some of the things that, if you, if you've been following our blog lately, you can see that we've got uh y conf coming up now. We've got registration open for that now.

A

And that will be since the page loads november, 6th and 7th and it'll be the highest place in downtown san jose.

A

What you may not know is that there is the day before an all-day class that you can sign up for and take that can help you get up to speed on yy there uh there's two tracks: um one is like an intro class. I think the other one is more for experts. We didn't do uh workshops last year. Do we?

A

uh Yes, I think so I think it happened the day after I was the day after it was like here.

B

A

Something or somewhere else, not at the venue right, yeah, yeah, so yahoo has this internal uh teaching uh group called l d and they're going to do either a yy introduction or a wi-fi building blocks.

A

So if you're coming out for yui conf, it's a great way to sort of get warmed up for that conference and if you are sort of new to yui, it's a great way to get up to speed so that when you are at the y by conf, you can ask intelligent questions and all that stuff is available on the blog and um tila was working on an awesome version of wow landing page that will be on landing soon.

C

A

Or something no we're going to use our own our own stuff. Oh, so, there's that okay.

A

So go to y white comp sign up now um take advantage of the early bird specials.

A

um There's a discount for the registration, as well as the hotel rate, has a discount rate right now until that uh queue gets filled up. What.

B

Are tickets? What's the price for tickets this.

A

Year, it's usually pretty cheap, so the registration for wyomington. I think it's like 15 75 yeah, so up until october 25th, it's 50 bucks and then after that 75. but wait. There is something more about that: a hidden secret.

B

If you thought.

A

If you um come to wildcomp as a speaker and well you're going to need lots of speakers this year by the way um your fee will be waived, so you can come for free and there's all kinds of cool secret, uh special things that are going to go on that.

A

I can't even tell you about yet, but they're going to be awesome, so I would definitely encourage you to sign up soon and on the same vein, um if you are even remotely interested in and speaking at yy conf, I highly encourage it, because not only can it help, you really learn what you think you know about why, but um you can have um there's a lot of benefits of that you.

A

Basically we videotape every talk, so it becomes a permanent record that you can point to on the net to point to potential employers or anything saying. Yes, I got it together enough to do the talk at a conference, so I encourage you to.

B

Do that yeah, it's uh quite common for larger conferences, that you provide them with videos of your previous talks um so before they accept you. So if you wanted to speak it out something like jsconf or something uh having previous talks is very very helpful to get accepted for this.

A

So what is the best airport for use at yycamp? um I've been through both san francisco and san jose yeah, since it's in san jose I'd, say, san jose is probably your best yeah.

B

San jose san jose airport's, literally probably five minutes from the venue, um oakland and san francisco- are both within 45 minutes um and you can kind of maybe get public transportation down here, but definitely san jose uh quicker.

A

Yes, and probably cheapest too yeah, I think I've done a lot of flights in and out both. I think uh san francisco is sort of a better airport. It's kind of big, though you just have to know it, but san jose is very cozy yeah, so it's not very intimidating at all.

B

Yeah, if you're doing that international, then you might just kind of because prices can really vary between the three airports, so just kind of go with whichever one is makes the most economic sense for you, but you can get there from any three um san jose is the easiest, though yeah. If.

A

You go to san francisco. All you have to do. There's a shuttle to the train. You can hop on the train I'll, take all the way down to practically the hotel itself. So.

B

You can cabinet from caltrain.

A

To the to the hotel, yeah so really same with uh the oakland airport too yeah and I wouldn't go home.

A

B

Just in general, or just in general,.

A

I don't know you should go, you should have.

A

You might wander into areas that you don't want to be.

A

um Oh and he's looking at uh is it dooby.

C

B

Is that his name.

A

Or adobe dobie, though it doesn't say dude, he says it's 2 000 uh aud 28 hours flying each way from where he is to get out here. Well, um that's why, like you have companies that help you pay for that kind of stuff?

A

It looks like uh ryan is saying that if he's coming from australia, you should definitely do sfo yeah.

B

That's what I said if it's international, I mean you're, probably looking at some 500 to a thousand dollar difference between each airport. Depending on I don't know what arrangements those airlines have so yeah, for in that case it makes more sense to uh just bite the bullet and take a cap if you get. uh If you go into sfo.

A

Yeah plus, um oh, it's adobe, dopey doe, oh, I was saying it wrong completely. I'm sorry!

A

Well, there's a song that goes doobie dvd, um so definitely look into like getting sponsorship or something you might be able to get um somebody like a company or something to help, help them and I've known people that aren't necessarily working at a company but want a presence there and.

C

A

Had people in the community uh come so yes and there's yes, ezekiel. We see you come on in.

B

Well, 33 hours to sfo isn't too bad that's about what it takes with uh 101 traffic around five o'clock here so yeah.

A

That's about right! You just sort of check your brain.

A

Yeah so um so you definitely look into getting sponsorships or uh the other thing are.

B

We offering sponsorships we'll.

A

See if you can get this, you can get someone to help. You.

B

uh But in terms of why ui sponsorship packages are.

A

B

A

I don't think so.

B

Okay, well maybe I don't know uh if, if for whatever reason, there might be some interest out, there, certainly.

A

B

Email we'll see if we can work anything out.

A

Yeah yeah definitely and also consider if you're gonna come out. If you're gonna fly that car uh do the whole thing. Do the three-day thing instead of just the two days, yeah there's, I know some people who are coming out the whole week starting on monday and doing stuff yeah so and there's also cool things around san jose and san francisco to do as well. Yes, bayer.

B

Is a very purpose if you've never been here before.

A

So that checks off, why why calm, except we are we're an open call for speakers right now? So definitely, if you even have a faintest inkling about wanting to do some speaking, you should definitely.

C

Get a deadline.

A

For those um there is- but I don't know at the moment probably some time before the conference- guess it probably within the next few weeks- we're gonna start uh yeah. So there's uh one thing about that. um This may influence your decision, we're thinking about doing shorter sessions like 30 minute sessions. So um that's good for everybody, because you can sort of distill your talk down. It's going to be basically like 20 minute.

C

A

With 10 minutes questions yeah, so if you were daunted by having to talk for 45 minutes, think seriously about you know, anyone can talk for 20 minutes about just about anything. So I.

B

Really like conferences that basically offer options on uh the the speaking slots and a little more flexible, instead of forcing everybody to go 45 minutes or something like that. In my opinion, you should actually just tailor the schedule around the talks that you, uh whether they're 15 minutes or 45, minutes kind of whatever um there does need to be a little bit.

B

Some structure there, of course, but offering flexibility is really good, because otherwise you get people that are that have enough content for 15 minutes and then the rest of the talk is just really really dry. So.

A

Yeah and that helps I think, people get get their knowledge together and if you, if you can't come up with something you know you need to be able to like think about, if you're passionate about something you could probably talk about it so yeah. The other thing is uh so they'll probably be like keynote speakers that will be at the beginning and ending of the day for about an hour. So that'll.

C

Sort of anchor the day.

A

And then, like I said, we'll have these 30-minute talks and supplementally. If um some of these are popular, we may actually schedule them twice, so that if there is a talk that you missed, because there are two in one spot, because that's that's nothing there'll be two tracks that they'll that we'll have there'll be one one track. So there'll be two sessions going on simultaneously for both days, interesting yeah. So if you miss a talk, there's the potential of being able to catch up with it again.

A

I think it helps make it a little bit more intimate too, because you can, you can either catch it the first time or you can come up ideas right.

B

Right right and the speaker.

A

Is great because that helps them kind of get their thoughts together. They've already talked about it once and they can sort of maybe answer the questions that were raised at the first time.

B

Yeah, it's a good idea. I haven't seen that used at other conferences, but certainly something worth exploring.

A

So with that, um we can go right into derek's.

B

uh Yeah, so one of the projects I've been working on lately is just uh adding in uh coverage information line coverage information to our testing, so um I guess yeah, just kind of general uh speak a little bit more generally about uh what we use and some of the cool features with it. So uh you guys are all doing unit testing right. Yes, you should be nodding. Your ads.

A

uh Collective internet.

B

Yeah unit testing is really good. There's.

A

Probably a few people going what yeah what's unit testing.

B

uh They look at yui test, so uh that's what we use for all of our unit tests, but uh if you're doing unit testing, that's also a really good idea to do uh to measure the code coverage too of your unit test.

B

So you describe what code coverage means yeah, so coverage is basically um a report that you'll see that will show you what lines inside of your code are actually being executed uh for whatever tests you're running against it, so um yeah very useful to know which functions you might actually be skipping over or not covering. So in an ideal world, you have 100 code coverage um now, sometimes getting it all of those little branches and testing. All of the uh conditions can can get a little excessive but anyways it's a good rule of thumb.

A

B

A

But you mentioned branches yeah.

B

A

Big thing that you may be mentioning uh people have code branches for different browsers. How do they have coverage for that.

B

Yeah so when you do code coverage for different browsers um well, so uh let's take a step back within your code coverage. The metrics that you can get out of it is lines covered um branches covered, I'm going to cheat here and look functions, covered and statements covered, so statements branches functions lines um now those aren't always uh there's not certainly not a one-to-one relation to all of those so um yeah when you have.

B

uh If I e do this, otherwise do that or more preferably you're actually testing the feature you're using feature detection to to switch between browser code paths. So um the problem that you can run into is when you're doing code coverage- and you have those branches.

B

uh If it's how you do this, otherwise do that it's with with most tools out there, you can't you it's impossible to ever get 100, because if you execute it in chrome, you're going to go one way. If you go in ie you're going to go another way, so you can get 99.9 but you're not going to get 100. So um one of the cool things that I've uh discovered in the tool that we use, which is sustainable uh here. Let me share my screen and they can dive into this a little bit.

B

So let's take a look at here: we go, let's just search uh istanbul, uh uh which is in fact not constantine.

B

Yes, so istanbul is a code coverage tool that is an open source project from an engineer here at yahoo um and we've been using it within yui, for I don't know about the last nine months or so very, very cool, because it offers a little bit more detail than what we previously had with our code coverage. So um anyways yeah, if you there's a url here,.

A

By the way, this is not limited to why. Why does this yeah be used for anything that uses node? I guess.

B

Yep uh yeah, so it uses node. uh Where is the url to.

A

Where are you trying to find me.

B

No, the url to um it's. I want to scroll no.

C

No, the url to the to.

B

The documentation, the api documentation- oh it's in the bottom right. What um no usually okay, well, maybe we'll get that out of it. uh So it is where's the home page for it uh here. If I go back, maybe it's um here we go okay, so here's the home page for what I wanted to pull up was the documentation.

B

So if you are already using istanbul um here is a cool feature that you can take advantage of to uh uh aggregate code coverage across different browsers. So in the collector um there is a add method, and so you basically do collector.add and then take your coverage. Information dump it in there and then at the end, you feed your collector into the h into the reporter.

C

B

You get your report with all of it aggregated, so the reason why I bring it up is because that's really really useful and in a lot of conversations I have with just other developers outside of yahoo, typically everybody's just doing code coverage in one browser, at least in ci and they're, not actually aggregating, all of it. So um if we look at, uh let's see so, here's just a code coverage report that I did using yogi serve in chrome on one of scrollview's modules.

B

So here's kind of what I'm talking about, which is uh force hardware transforms. So if it's a webkit browser.

B

Then return true, otherwise returned false. So since I tested it in chrome, it went through this code path to true, but the false is highlighted, meaning it did not actually go through that branch.

B

However, when I have when I run code coverage through yeti by having chrome and ie hooked up to it, this is actually the report you get out of it. So let me find that same scroll view, module scroll view, base scrollybase.js and then go down and boom. Now we actually have both branches covered and the percentages will be a little bit different up here, too so uh 79.78, and then it should be a little bit lower over here. So out of all the different browser code paths, I gained an additional 5.

B

So theoretically now you can't actually get 100 code coverage. um So what's in your reports, why ui's um standards for code coverage? Ninety.

A

B

Ninety percent- uh ninety percent statements- I think um I don't think we've actually uh gone beyond that- to introduce any additional thresholds, but yeah 90, state or 90 lines.

C

B

Yeah, I don't know, but that that uh that metric comes out of kind of the the older tools that we have. That would actually only measure lines, but now we have all this so so one thing you kind of jumped a little bit quick through was uh you don't need to get this information yeah, you sort of describe how you do that yeah absolutely so. I can uh show you here um so I'm in the scrovia module, if I do yogi serve, that is going to pop up you'll, be served.

B

uh Your yogi's yeah just serve page four to help me out with scrollview, and so, if you want to actually run unit tests with the coverage, you'll see this column here and and when you click on scroll v base, it's actually going to go run through the unit tests and then redirect to the uh to the html report um that has all the code coverage information.

B

So so the question iris is: how do you get the multiple browsers to work? So that is a feature that I'm currently adding to yeti. um We don't have an eta and when that's going to ship, but hopefully soon um in the meantime, if you wanted to, I don't know if you're familiar with yeti and wanted to kind of hack. It yourself.

C

I do have a branch out there on github.

B

uh Let's see that would be github.com.

B

And then I have a branch called coverage.

B

And yeah my last commit was a prototype for introducing code coverage to yeti. So we take a look at that commit um and you'll see. Basically what's going on here is at the very er when it handles all the results coming back from the agents or from the browsers.

B

It's um grabbing that coverage information out, that istanbul delivered or that istanbul set as a global uh within the within the test page and then I pull up and return it back to yeti with the with the unit test results, and then I just basically write that file out to the file system and then, when everything's done, um this is probably more ideally going to be broken out into its own separate reporter um a reporting script that'll be like within the yui 3 repository, but for right now it's built into yeti.

B

The general idea here is that when, when all the testing is done, you go through grab all of those raw json files that um uh represent coverage per test per browser, and you just loop through them all uh add them all into the collector.

B

And then you pass that collector object into istanbul's reporter and then write it up to ditch. Istanbul knows how to sync those together and resolve things out. Yeah um yeah istanbul is pretty smart about um not duplicating any lines um yeah, it's meant to as the documentation as it's meant to handle thousands of uh objects and it's uh to collect thousands of objects in the data set um and then do intelligent uh merging of all of those without double counting any lines or anything.

B

So um do you think it's far enough along to have like a why my contrib, probably not yet, um because yeah once I think, once we get it into getting um and we start working with it then yeah. I think at that point. It's certainly worth kind of throwing it out there, but right now it's really jus.

B

It would just really kind of be talking about what what we um have in the works as opposed to um and not really having anything that people can get their hands on to play with so um yeah, hopefully, it'll be landed in yeti soon, and at that point, um then anybody can kind of use this information. What.

A

Really brings it really makes the the coverage like a reality right, because what you said before is you could only see half the picture or part of it, but you still have to run like yeti with a lot of different browsers right, yeah.

B

Yeah, we've always had this problem of uh I mean there's some with so like within scroll view. I mean it's just a couple percent difference, but there's other browsers that or there's other modules scrollview, for example, even has so. Let's take a look at that: yogi output for scroll view actually yeah. So uh so there's a scroll view base um ie file that uh enhances scrovybase to work in uh in legacy ie and get rid of that uh so, but that module doesn't appear in this code coverage report because we test it in chrome.

B

So for all this report knows there's it I scroll view base ie doesn't have any tests uh written for it and is isn't touched by any unit tests, uh but when you actually run uh run it through um yeah run it through yeti with ieee and chrome. That's when you see a scroll view based on e-pop-up is a tested component, so yeah. So basically, if you want really comprehensive unit test uh coverage, this is pretty much the only way to go about it. um Yeah.

A

Well, another way you could use this, perhaps as a way to let's say you're trying to get rid of code, forks and things you could use this to find out what you know is it still trying to test for ie?

A

You know you can find out how it's affecting your coverage to move it.

B

Yeah, uh so that's basically it yeah check out uh istanbul, uh github.com, yahoo istanbul and it's a very, very cool, uh very well written node.js application to handle code coverage for you. Do you get that for free when you install yogi? uh No, I don't think so, yeah! No it does. It does yeah. uh So yogi includes istanbul, so you would run yoki. What to get coverage.

B

Yeah yogi serve is the only way to click on those links.

A

B

Don't know if you do yogi test, I don't know that there's any coverage output that you can get from yogi test. That would actually be really useful, though, like to skip all the ui and just go straight to the yeah. If I just want to know the general code coverage for phantomjs, a webkit webkit-based browser, um which is what yogi test uses, that would be really um useful info to have on there. So so um for all, I know, dave, probably actually everything that I come up with.

B

Oh it'd, be really nice to have dave, already wrote it. You just have to actually find the uh find the switch to actually turn it on. So it.

A

Was an interesting topic, we're going to bring up at some point about how um we've got this great till yogi, but how it's not being maintained right now and what? What.

C

A

uh After yogi, what's supposed to yogi world, do we continue to keep yogi running or do we? uh You know break it up and you don't use the grunt. You know that's a conversation that we're having yeah.

B

Probably a little bit more uh than what we should probably bite off in this round people, but.

A

B

That's a future thing, but.

A

It's something that um if you do find uh issues with yogi, please definitely file them as github issues right now, so that we can keep.

B

Yeah- and we are- um I mean so uh to that point, while yogi isn't getting a lot of new features, we are still doing um any essential maintenance and stuff on it too. Oh.

A

B

And kind of figuring out.

A

What the road map for it might actually be yeah. There may still be something called yogi it may, just under the hood, be doing different things.

B

A

Plus it's extensible via plugins as well. Hey, then you turned up here: do it nothing else to share yeah? Last week you hear you, you said you showed the screen on, but it didn't show anything because it wasn't. It.

C

Was right, main screen, yeah.

A

So everybody got to watch me, checking, isn't yogi really aimed at white wine gallery rather than our own projects. Can you use yogi for.

B

Yeah, you can use uh yogi for your own projects too, uh but.

C

B

Mean it's certainly it um like, I think, a lot of things that come out of the thing projects that core team members are working on. We, it works best, of course, doing library. Development then also gallery development too, uh and then beyond that um I mean it's all kind of designed to work outside of the context of yui. um But if there's, I think yoga is very specific about well, it stands for yui or gallery interface.

A

C

B

Within there that you can use within um yogi test, for example, um yogi serve. I mean those work outside of the context of uh core modules or gallery modules, so the test is really doing yeti underneath behind the scenes right serve is doing, selling test is using grover, so test doesn't use uh yeti at all. um So yeti uh is just kind of in my mind.

B

It's just kind of the the brain behind directing web browsers where to go um and does it have to exclusively be used in the context of unit testing, so I've modified it to use it for performance testing. You can use it to really direct to any type of page. You want to. It doesn't really matter much, but so yogi test is something that's written exclusively to work on the command line that doesn't need the use for traditional ui browsers. So.

C

It doesn't need to use.

B

Yetio, that's cool yeah.

C

Yeah- and uh I guess we're actually using shifter under the hood for yogi, build right and a bunch of yahoo properties are using that now to build their application modules.

B

Yeah yeah, when you run with the performance testing stuff, when you run yogi perf, um it uses yui benchmark uh so really yogi's. Just a wrapper around all of these um unifunctional tools, a single piece, even when we do a release.

A

We just do a yogi release and that packages up everything.

B

Yeah, it's basically that, like one kitchen gadget that blends and mixes and tosses and makes your food for you, yeah.

A

That's a food replicator.

C

A

That's cool so, in terms of other things going on, there are a number of folks on the team doing lots of really cool sort of uh experimental type stuff. I know that clarence is working on. uh He was originally starting. He started looking at like data binding and it started off with um what was it dated it wasn't. Data table was it that was simple. Wasn't it like, or one of the other things that we were going to talk about with him at some point was yeah.

A

He originally was really working on a way to do like lists right yeah. I think it was like simple model list, yeah model that stuff and that led to um data binding stuff and now it's leading to uh eric- and he are working on this great uh project within yy that will deal with data binding it'll deal with um web.

C

Components: yeah web.

A

Component spec, one of the ideas behind uh future work and yy will be conforming closer to the web components back being spearheaded.

C

A

Chrome- and I guess firefox.

B

Yeah yeah there's some really cool stuff with. What's that is actually a really good idea for a roundtable.

B

But yeah we can uh at a future round table in the near future. Yeah I'd love to kind of do it. Do it deep dive into web components and kind of show off some of the stuff? That's in there, because it's really.

A

Really neat stuff, there's web components. It has to deal with shadow dom where you can create your own tags that have your own.

A

It just create tags that has their own markup underneath.

B

Of it so yeah um and it it really lends itself well to or it can lend itself well to yui development if we kind of take these steps to embrace web components a little bit more because so much of it is kind of based on ideas that have been um that have been in built into yui, as well as other ui libraries for a while. So it's kind of a there's attribute like stuff and then there's widget-like stuff in there. um So yeah yeah, cool.

A

um I know that other members of team are working on things like finding ways to make um when you, when you create like combo handlers. You know there's like 20 or 30 of those, depending on how many uh components you unload and finding ways to shorten those urls. So.

B

A

The number of requests that you make are less so your overall performance goes up. There's cool things like that. I know that um trying to think of what other people are working on, there's uh an effort to get all the swifts out of the repo yeah coming up soon. We just killed some yui.

A

It keeps dying, we keep killing, it um kill the forums as well. Oh, yes, definitely yeah! That's the ladies.

B

And gentlemen,.

A

We, uh our new nickname for andrew, is the executioner. Yes, I've killed so many things this past week, including the forums. So if you have any yy2 questions, please go to the yy deprecated forums. There's not a lot of activity there yet, but I suspect that people are going to, like other places, to find they're. Quite like what's another. What's that popular, uh you know the place where you go when you ask for.

B

A stack overflow stack overflow.

A

Yeah, so many people are like just wandering off there because there's not a lot of activity on the dedicated forums.

B

We've talked in the past about yui two's deprecation. Is there anything going on there or is it still just kind of in a work in progress to get to the point of deprecating? Oh yy2 has been dedicated, for it has been africa. Okay,.

A

Oh yeah yeah yeah, okay, but um if you go to um why why not get into io slash yy2, we have finally archived all the old yy2 documentation. The user guides the examples the ui spec.

A

I mean the api, all that stuff now is available on there to peruse, and I feel sorry for you if you have to go there, but if you do it's there, yeah um so yeah, definitely, and if you are finding that a question that you have in the new forums is not getting answered, feel free to ping me and I can help make make the answer come about and you can do that either via irc or you can message me on twitter or anything like that.

A

um That takes care of looking throughout the list of cool things that we're going to talk about. um There was one thing about: maybe coming up with topics for my white conference talks. I mean we came up with one already, which was um someone was wanting to do uh even a workshop, potentially on unit testing and code coverage.

A

So that was one good idea.

C

A

um If you have ideas for talks- and you don't necessarily want to do it yourself, but you think hey this make a great white comp talk. Please send us a message somewhere somehow like send me an email and if there's enough people wanting that kind of talk, we can just find a person to do that. I really love.

B

The idea of having the 15 to 20 minute talks and then we'll just have a whole ton of videos like there could be the istanbul talk. There could be the yui test talk and there could be all this stuff yeah and that we can quickly direct people to about how you use all of these different tools.

A

B

um And they're, not while the longer more extensive talks, they kind of go through how you would do uh unit testing in all of the little caveats to it and best practices and that stuff longer form talks are good, but it would actually be nice to have kind of these little short form. Topics too.

A

Yeah for each of these tools, so um eugene uh just put you on a spot someone's asking about the new combo urls. I know that you're saying it's in the future, but can you talk a little bit about what what that feature might look like? Is it uh oh.

C

Yeah so let's see, um if we're just looking at the core yui modules, we were able to reduce the length of combo urls by about 75. um It's it's nothing like super complicated.

C

It's basically just removing all the redundancy in the urls um that you know everybody has been talking about so the hard part in getting there was actually more about uh working with other teams within the company, because there's a there's a lot of most of the hard work or most of the the changes are in the infrastructure and so and also my had integrated that into like their qa environment and like we were all ready to go, live with it today.

C

But then we found out that you know they misconfigured something for their their https uh proxy, so they couldn't launch with it, um but I think they're gonna. They should be able to get that resolved within the next few days and then we'll probably see it uh live in production.

C

um Once we confirm that nothing blows up in production, then I think we're going to try to get it pulled back into decor.

A

Oh cool so it'll show up in the configurator when you go to build your own uh combo urls.

C

um Yeah, I guess we'll have to update configurator too, um but yeah yeah.

A

Yeah, I I know what you're saying that it'll it'll there's you're talking about the the ones that are generated when you just do a use statement, but you're right. The configurator also sort of. Does that for you, when you don't want to take advantage of the the loader. I guess.

B

Yeah exciting stuff, though, when this final lens, because it's something that we've wanted for years and there's been a number of attempts at projects, I'm trying to resolve it and it seems like we're heading down the right path with this one. So nice work yeah this.

A

Is huge so we'll we'll feature this one when it goes, live and we'll maybe do a demo or something maybe next week, yeah, uh let's see so in terms of items that we usually cover.

A

um I went through all the sort of pending github requests issues, open issues for, and the criteria for, the ones that are sort of stale are ones that have been they're older than like a week and they are owned by someone who's, not on the y team and they're, not a work in progress and a lot of those are still sort of in the state they were last week.

A

A number of those have gotten resolved, no one to call out ezekiel for helping to find some of those to get them resolved, and a number of them are much less than a week now, which is great. So there isn't anything new on that on that side of things, it's pretty much the state. It was before, there's a lot of ones that are owned by yyt members that just sort of need to push them forward, we're kind of in the middle of the cycle.

A

So there are things that are still in mid production, so there's that and then again we've talked about before the um up for grabs bugs. Do you know if there's any new programs.

B

uh I don't know um I will say there is one that should actually probably be classified as upper grabs. um I don't have many spur cycles to work on it now, but uh there's one involving scroll view scroll bars. So, um let's see, if I can find this one and then we'll add it to the upper grabs.

A

I'm looking at for grabs, there's not been anything new for a couple months. Now it's like scroll bars.

B

A

Well, there was a thread while we're talking about that on um on the uh white yy channel with mayumi some other folks about a data table. I know that tony can't make it today, but um there's still a lot of people here. Talking about how oh yy2 the table is so great yy3d isn't there yet and just to reiterate that this is basically tony's sort of main project right now. So if you run across anything that anything in data table 2 that you want to see in data table 3 for yy3, please file issues.

A

There's an open, trello board that talks about the ongoing issues, ongoing features that are going.

C

A

There, I don't have the link on me, but you could search for it and tony is available at apipkin and um basically, if you're running into something where let's say that you're working on a you're working on a website, that's using datatable for yy2x and you need to transfer it over.

A

There are just. I guess the main question is find out what it is. That's missing the features that you need and we can work with you guys to help make that happen, so ping uh pipkin on irc.

A

So that's my public service announcement, but I know that I know this is an ongoing issue. I know there's a lot of pain involved there. A lot of people are like they wish that they had a table. Three was up at the park.

B

But it's getting there so so the uh the issue that I was talking about scroll view, scroll indicator, rendering bug on ipad issue number one one, six: four, so that that would be a for grabs bug. If anybody wants to do some investigation so.

A

Do you enter the label.

B

uh No, I have not had it. Yeah just drag that tag.

A

Cool so again, the ground bugs are issues that developers are not um they. They know that they're nice to have, but they just don't have time to work on so yeah. uh It's just it's the closest thing we have right now to you know I want to contribute, and you don't really know where to start. If you go through these up through grab bugs, you could definitely find ones that would uh uh offer up a great help.

A

It could be things that we could definitely use, and I think that the ones that are marked up progress will probably get fast tracked for committing yeah and.

B

While we might not necessarily have uh enough time to actually fully uh fix the problem and write all the tests and everything we're very happy to work with you guys um for anybody who actually does want to help contribute to them. So, okay,.

A

And someone asked: is yahoo open source yeah the whole company uh yahoo is not open source. I don't think it's written by source. Anything yeah who is constructed by a source code right now, who's a corporation but yui is open source and it's open source in both ways. It's a two-way open source and someone asked before what is two-way, open source two-way open source means.

A

Well, there are some companies that, like nasa, that has one-way open source where you can get the code and you can read it and you can use it, but you can't contribute back and two-way open source. Is you can submit for requests and get code back in some people call the first.

B

One filter source, f-e-a-u-x.

A

But I mean some companies, like that's the first step right right and there's also um I've seen companies that have two licensing: dual licensing. Where one track is it's not it's like you know, it's open source that you can read, but you can't write, but then they have another license that you can write to, but they don't like have any. uh They don't like uh support that at all.

A

B

One way we prefer the simple method on the team, which is just everything, goes out there open source and it's all under bsd, which is just about as liberal.

C

B

You can get on.

A

Contributions so yeah and you don't even know- and you can write stuff on top of yui and it doesn't have to be open source. You can go into the gallery and have your own closed source components or open source. If you choose yeah, which makes it that much more flexible.

B

Another roundtable idea- licensing, oh man, we don't get guild involved.

A

But um so one thing back to the y account for a minute: if you, if you come up with an idea for a conference topic, um please let me know somehow: either you can email me my username at github, gmail.com and I'll be happy to convey that onto the team. So even if you don't feel like you can speak or that you're not going to attend. If you do have a great idea about a meeting topic, we can use that.

A

C

Outside of that, I don't have.

A

Anything else new for this week do we have anything going up for wi-weekly.

A

Yeah so right now we're just.

B

A

B

Development cycle.

A

Everyone sort of heads down working and.

B

A

We have folks at different conferences.

B

Yeah tc39, uh which is the uh ecmascript standards body, so they're often venting.

A

B

Javascript right now in boston, yeah.

A

So we should be able to hear back from them when they come back. We might be able to get some information back from there.

B

When they return uh next milestone, looks like we have one coming up next week september. 25Th is the code freeze, pull request deadline which is next wednesday? So if there's anything you want to go into what I believe will probably be your 3.14 13. 13., I don't even know yeah. We remember we.

C

Raised the beta.

B

All right, okay, 3.13, yes, so the next release will definitely um plus we'll run into any blockers in the release window will definitely be 3.13, so anything you want to get in there uh next wednesday is your last chance.

A

Oh and someone asked: will the workshops be recorded and available on the theater? I do not think the workshops will be. I believe the all our sessions will be, um but I will ask about that and find out. That was a good question.

A

A

Yeah I'll find out one way or the other, and it could be that uh they may. They may have already like videotaped examples that might be available too. So.

B

Good mayor two weeks from no next week, not this coming weekend but the weekend after is open hack at uh yahoo's sunnyvale headquarters. um So yahoo developers.tumblr.com uh there's a post on there and there's some links in the y way weekly um posts as well, so yeah yeah.

A

That's a great one if you're anywhere in the bay area, and you want to meet marissa. This is definitely the way we meet all a lot of iui engineers will be there. A lot of other yahoo engineers will be there there'll be other companies like. Is it pinterest yeah? So it's a great way to network and a great way to um meet your favorite person.

B

A

Cool all right: well, thanks everybody for coming. We have we're nine minutes early, but you can take that nine minutes and go look under the up for grabs pool requests. Thanks, bye, see you later.

B