ng-conf 2014

La comunidad de AngularJS

Igor Minar  · 

Presentación

HTML (pincha para descargar)

Vídeo

Transcripción

Extracto de la transcripción automática del vídeo realizada por YouTube.

hey first an errand as well as you saw the first thing they did they electrocuted me and the next thing they did they asked me for permission to organize this conference it was actually quite funny because the way they approached me it was similar to like

when a young man is asking for permission to marry once daughter you know they promised that they would make this a really good conference you know take good care of you and actually not organize a conference but make this a experience and I think they delivered

on the promise and I would like to thank them thanks Dave Joe Mary Sami Andrew Leon other people that I haven't even met that made this conference possible thank you so today I would like to share a story with you story about angular story about passion

a story about community when I joined Google back in 2010 that's how dorky I looked with the help properly properly prepare ahead I joined specifically to work with Mishka on angular actually back then there was no angular there was Google feedback but

I joined because I saw how passionate he was about the project and I believed back then that together with a lot of work we could make this into something extraordinary something that would change the way developers feel about the web and also change indirectly

how the users use the web and perceive it those were the crazy times when the mailing list was silent because only few crazy Souls wandered in there and the poor requests queue was empty and angular had literally no bugs chuckling when I joined um Angela had

no versioning scheme no releases no Twitter Google+ none of these things that you know as angular today but things were simple now we could iterate very quickly we could gather feedback very quickly because there were only a few people who would give it to

us um we could make breaking changes easily because nobody will depending on our api's and because there were only a few document documents or documentation about angular to be had there were only a few typos and only a few very poorly written Doc's

actually all five of them but I actually wondering how many of you remember that back then we used to have weekly releases something that we do now again and I'm super happy about that anybody here anybody here goes that far in the past I don't see

anybody yeah so we used to have like weekly releases and it was awesome and I really loved how somebody from the community maintains the angularjs Wikipedia page after each release a new new field in the stable is added with the release date the version number

and the code name about these code names and we get a lot of praise for them and we actually have a ton of fun coming up with him back in 2010 we felt like we needed a theme for angular and I believe it was Brad who decided or decided who suggested that it

should be superheroes so naturally we set out to come up with a new superpower superpower sometimes and add this new superpower to this hero that we were building and it's actually funny because we get suggestions and lists of like dozens of really funny

code names from people all around the world and it's just great and the Wikipedia page it makes it easy for us to pick a unique name every time we need to pick one by the way there is a typo you need unicorn zapper is there twice so if anybody wants to

go me pick Vika pedia and fix it let's go and do it so we got better and things got a little more serious we got better the releasing and versioning sometimes a little too serious if you remember our c12 yeah but as SP as we were making angular better

more and more of you were becoming interested in angular and we crawl we worked very closely with you you know we reply to hundreds of emails and watch Twitter replied there but as the community grew it became more and more difficult for us to keep track of

all the emails all the IRC Twitter Google+ and scariest of all you don't get help I don't know how many of you have three thousand five hundred and nine hundred github notifications I do so as Brad mentioned and mishko yesterday angler was not built

by was not created by a decision of some VP of engineering it was born out of love and passion and it slowly build up its reputation at Google and adoption unfortunately the time was just right and when the community totally exploded angular was already well

established in many key projects inside of Google this allowed us to hire more people and I would like to thank Google for this and also Brad he's the most awesome manager you can have so one by one we had Brian naomi james jirayu geoff to be us join us

in the office and we had also folks joining us remotely like Julie Pete Powell who's not here on this slide but should be and Matias and recently Katelyn we even built Mary Poppins as guys mentioned yesterday but one thing that you know we notice is that

even though we grew as a team the community grew even faster late in 2012 I had the honor of be of being at da Jes in Paris friends vojta gave a very inspirational talk about following your passion because that's what vojta does and I don't mean given

inspirational talks but what always follows his passion he was later followed by fat or Jacob one of the creators of Twitter bootstrap who gave a talk about how open source is killing passion and creativity no fat totally burned out was just talking about

how he loves to create things and how he loves to be passionate about them but the problem he faces is that as they get more popular they grow from this small cute little puppy into a big dog or a big monster that they have to maintain and it's not fun

anymore it's it's just this being a slave to project that you created well back then I wasn't brave enough to admit it but I had the same problem you know I helped to create this monster called angular it was a cute little puppy when it started

but as the the community grew it became harder and harder for us to just do anything and the maintenance overhead was just becoming too big he closed his talk by challenging the audience to solve the problem of open source so that we have more people come

in start new projects be passionate about them but not get burned out by the maintenance overhead later that day we had a dinner with fat and other speakers and we talked about this and we couldn't come up with a solution no I was I was pretty depressed

because I felt like what can I do about this and after a while I was like you know I really want this cute little puppy back so I need to come up with something and I started working on a plan but it wasn't without complications in like late 2012 early

2013 we were approached by a team at Google who really liked some of the features in angular they they wanted angular but they also wanted server-side pre-rendering so we thought about it and we decided to give it a shot so we implemented the version of angular

with the server-side pre-rendering this means that you can use the same templates on the server as well as on the client and when the syrup renders the HTML is sent to the client and the client that when the JavaScript is loaded it will just bootstrap and

start from that point as if you somehow hibernated the application that stayed it worked really well it worked better than anything else out there at the time but we didn't love it it wasn't it wasn't great it wasn't beautiful and when the

project team decided not to use a solution for non-technical reasons we decided to abandon this effort primarily because it would take another three or four iterations to completely rethink this and make it something that we would love to use ourselves and

unfortunately would most likely also lock us into some back-end and we didn't want angular to be tied to back-end at a time another reason why we abandoned was that we realized that the two main reasons I people want to recite paren during SEO and the

performance improvements would not be an issue for much longer especially if we focused on the client side and made many improvements on the client side unfortunately like this learning experience it was great for us we learned a ton and I don't regret

it but it put it all on on angular it it delayed the 1.2 release significantly and we spent the rest of 2013 just trying to catch up and shape one or two as we were working angular 1.2 we were constantly constantly being amazed by you know applications that

you guys were launching and it really gave us boost and helped us made angular 1.2 awesome just take animations that's a good example of that I think they're fabulous but to get to those results you have to iterate a lot and for example with these

animations you know we rewrote them three times completely from scratch and only then we were satisfied like yes this is pretty good we should ship it but it's hard to iterate when the infrastructure on the project is crumbling the CI server that I built

a couple of years ago started to show signs of inability to scale as especially as the team grew you know the build queues were getting longer he was taking longer to test anything and a velocity was going down but the biggest issue we faced was that our our

core our team only solution completely excluded the community what this meant was that when somebody from the community sent a pull request it was most likely broken because it was not possible to automatically test test this for us so we faced issues with

[ ... ]

Nota: se han omitido las otras 4.768 palabras de la transcripción completa para cumplir con las normas de «uso razonable» de YouTube.