text
stringlengths
0
1.17k
[962.40 --> 968.06] only changing or rebuilding the pieces that need to be rebuilt which is a you know an np hard problem
[968.06 --> 972.20] yeah it's probably a huge saving too for the the disc you're on and just in general this is speed
[972.20 --> 976.76] exactly exactly it would be a huge win um and there are a lot of people who've said this would be
[976.76 --> 980.70] amazing this would be amazing and of course i mean you know we've just both agreed it would be an
[980.70 --> 986.32] amazing feature um the as i was going through the you know several hundred issues i think there were
[986.32 --> 993.80] like 623 or something uh open issues when i took when i came on to the project um as i was going through
[993.80 --> 1000.20] them depending upon the number of comments and the sort of logic of each argument um i sort of weighed
[1000.20 --> 1007.38] them in a certain way um if there was a and and sort of got to know how people were using it based on
[1007.38 --> 1013.42] their comments and the issues and the you know occasional site that i would come across um on
[1013.42 --> 1020.02] some repository on github um and as that as i sort of got to know a little bit more about how people
[1020.02 --> 1026.12] were using the product um i said you know this we should support this or or this is not really how we
[1026.12 --> 1030.28] envision this product to be used but maybe we can make a compromise and just make it easily
[1030.28 --> 1034.56] extensible so they can build it on their server without having to do all this crazy monkey patching
[1034.56 --> 1041.40] etc um and so sort of weighing what are jekyll's primary objectives based on what was in the readme
[1041.40 --> 1048.64] versus how are people using it versus how do people want to be using it what is the best sort of
[1048.64 --> 1053.70] middle ground between those three elements um and i've i've that that took a lot of thought
[1053.70 --> 1060.14] and when you first take over a project or first enter into a project like jekyll that has been
[1060.14 --> 1068.08] around for five years um and is is relatively successful um it it took me probably six months
[1068.08 --> 1075.68] to figure out exactly what what the trajectory for this product should be i'd imagine even your
[1075.68 --> 1081.70] your methodology had to be pretty methodical too to kind of go through comments and i mean maybe
[1081.70 --> 1085.50] you're weighing them based on is there a code sample you know how passionate is this person is
[1085.50 --> 1092.38] it you know is this person commenting on several other issues as well or kicking up issues um i gotta
[1092.38 --> 1097.28] imagine that was a pretty tough job to triage and like you said six months to even get a heartbeat
[1097.28 --> 1102.88] that's that's a lot yeah um and it's it part of it is because there's so many people using it and
[1102.88 --> 1108.04] and as well it was still being used on github pages so i sort of had to weigh in well how would
[1108.04 --> 1114.04] this change github pages is this still secure for github pages um and one of the things that tom said
[1114.04 --> 1120.82] to me during our initial chat was um basically like instill the fear in me of change which is very
[1120.82 --> 1129.24] interesting um that said jekyll is is good as it is it is good at present it's good um it can be great
[1129.24 --> 1135.82] but it shouldn't be but the the the way to get to greatness is not through completely rewriting
[1135.82 --> 1140.58] everything that you have um basically to say add on to what you have change the things that
[1140.58 --> 1147.20] absolutely must be changed um but don't don't go too crazy basically um so in terms of accepting
[1147.20 --> 1151.76] pull requests that made me very skeptical um originally i was like oh you want this feature
[1151.76 --> 1158.38] let's you know let's merge it in it'll be great but as long as the ci passed um but as after time
[1158.38 --> 1164.54] after some amount of time even the the general idea not even just the code i would scrutinize
[1164.54 --> 1168.88] um significantly is this something that is useful to the majority of jekyll users for example
[1168.88 --> 1177.52] is this something that is safe to run on on github pages is this something etc etc so um it it has
[1177.52 --> 1184.30] taken a long time but but yeah it's it's been i think that's a really key part to to taking over a
[1184.30 --> 1192.58] project into to making something cool so what um i guess maybe playing off of if i'm tracking with
[1192.58 --> 1199.58] you uh early on tom said you know hey you know kind of toe the line so to speak you know when you
[1199.58 --> 1206.00] first took over how is that how is that contrast against how you are now with the project and what
[1206.00 --> 1213.72] changed so when we when i initially took over tom was still very much present um
[1213.72 --> 1220.46] you know quote unquote so i could i had a tag on on the issues that was at my jumbo and i would email
[1220.46 --> 1225.20] him if that got too high maybe 35 issues or something like that and it just needed a decision
[1225.20 --> 1230.72] by him it just needed you know hey what do you think about this um is this a good idea bad idea
[1230.72 --> 1236.98] so i sort of at the beginning was was really chatting with him a lot um as much as possible
[1236.98 --> 1243.52] and getting his idea about what the product and where it should go and as a contrast to now
[1243.52 --> 1252.38] i have complete control um i can't imagine so i have complete control in the literal sense in that
[1252.38 --> 1258.96] i can change anything but uh there's still some philosophical constraints of course in that i want
[1258.96 --> 1263.42] it to be something that people like to use etc um and something that that continues on with the
[1263.42 --> 1268.70] tradition of what jekyll has been um if you take a product and you completely modify what it's like
[1268.70 --> 1273.54] completely change everything then it's no longer the same thing so existing jekyll sites for example
[1273.54 --> 1279.80] i don't want someone to write a site and then for it to immediately break um we've with the 2.0 release
[1279.80 --> 1286.46] we did our best to to maximize the number of backwards compatible changes um i think there was like
[1286.46 --> 1293.14] maybe one backwards incompatible change and it was we still had a way to to work around it so
[1293.14 --> 1302.00] not to mention too you also had github as as uh i guess a uh a customer so to speak um right because
[1302.00 --> 1306.20] they're using it for pages and they obviously have a trajectory where they're taking uh pages and what
[1306.20 --> 1312.46] they're doing with it not it's obviously a large part of the open source ecosystem where people host
[1312.46 --> 1318.54] their docs on there or they host their you know single page kind of here's my repo kind of thing or
[1318.54 --> 1324.80] even just simple sites it was a part of what github was doing so how did uh i guess maybe to break the
[1324.80 --> 1330.60] seal on this so right now you're an intern also at github so that that kind of had to blossom into even
[1330.60 --> 1337.82] new opportunities for you can you talk a little about that yeah so so um my my exact title is a
[1337.82 --> 1346.88] as a github pages contractor and so i as a contractor and um i'm working on github pages and trying to make it
[1346.88 --> 1351.78] an even better platform um some of the changes that we've released like the site.github namespace
[1351.78 --> 1357.28] um there's been there's been a complete rewrite behind the scenes in the back since i've i've
[1357.28 --> 1364.46] joined the team um and basically what that's done is allowed me to gain new insights into how
[1364.46 --> 1369.72] jackal's being used um in particular i'm working with ben balter who's an amazing guy um really
[1369.72 --> 1376.80] brilliant um he recently he graduated from law school um was a wordpress core contributor um
[1376.80 --> 1384.40] was a white house presidential um innovation fellow is just a crazy cool guy um and he's sort
[1384.40 --> 1388.94] of been my mentor on that project um the guy watching over me making sure i don't mess up too
[1388.94 --> 1397.06] many things and uh and because he's focused on government i've gotten a huge i've i've gleaned a
[1397.06 --> 1406.42] new or gained a huge appreciation in how um how jackal is being used on on the massive scale or larger scale
[1406.42 --> 1412.64] um so if you're if you're a government institution that wants to publish data how can how are they
[1412.64 --> 1418.72] using jackal to publish data for example um they're using wordpress and jackal in many many occasions
[1418.72 --> 1429.18] to publish open data to um publish process um project open data is a jackal site um and they're
[1429.18 --> 1435.36] using that to write policy around open data if you're you know the city of chicago how should your how
[1435.36 --> 1439.74] should your data be released and what are the what are the guidelines surrounding that that entire
[1439.74 --> 1446.74] project that um every everything within that all the content is written um written as a jackal site
[1446.74 --> 1453.18] that's funny that you mention uh ben because you know he also just being a core contributor to
[1453.18 --> 1459.06] wordpress uh you know for those who know about jack or maybe this is you're a listener this is the
[1459.06 --> 1464.18] first time you're hearing about it i don't know where you've been but um wordpress and jekyll tend
[1464.18 --> 1470.58] to fall into the same conversation because it tends to be a fork a choice of left or right jekyll or
[1470.58 --> 1475.60] wordpress and you know a lot of the reasons why developers like is one because it's just
[1475.60 --> 1479.68] developer centric i think far more than maybe wordpress is but not in a bad way
[1479.68 --> 1484.40] wordpress is kind of designed and delivered as a product for different types of people and different
[1484.40 --> 1489.32] types of developers but you know the separation of the database and stuff like that so does does
[1489.32 --> 1493.44] ben get involved with the product is he involved with jekyll now or is he just kind of advisor to you
[1493.44 --> 1499.44] he's he's definitely involved in the product um not as much as as i am you know he's not he's not
[1499.44 --> 1504.44] around day to day um but when i have sweeping questions or large questions i would have sweeping
[1504.44 --> 1511.28] effects on the product i tend to tend to run them by him um he actually was kind enough to
[1511.28 --> 1516.52] invite me out to san francisco um or out to the github headquarters i was interning for visual
[1516.52 --> 1521.72] supply company at the time out to the github headquarters one saturday to sort of host a
[1521.72 --> 1526.44] jekyll nano summit uh there's a github issue with all of the on jekyll jekyll with all of the details
[1526.44 --> 1534.88] of that summit um and we sat in the situation room in the github office um and it was ben and me and
[1534.88 --> 1541.80] mislove um who's a great guy and tom came for about an hour um and or an hour and a half and then
[1541.80 --> 1550.06] um garen um as well gj tarikian i think um i'm not sure about your name sorry garen uh your last name
[1550.06 --> 1556.52] rather um and so we sat around and matt matt uh rogers of course my co-maintener um he was in he
[1556.52 --> 1563.98] wasn't able to come physically but but he was he was beamed in via blue jeans um and so we we sort of
[1563.98 --> 1569.14] chatted about the the future of jekyll what we wanted for 2.0 what we wanted for 3.0 sort of what
[1569.14 --> 1578.32] the future would be um so ben has been a significant advisor um and has has sort of opened up or suggested
[1578.32 --> 1583.90] things like a nano summit where we would all meet and chat about it you know in in meet space um that