text
stringlengths
0
1.62k
[3847.86 --> 3854.92] a consumer would see um is obviously phones um and arm 64 the big arm 64 platform is the iphone
[3854.92 --> 3866.42] so everything from the 5s up is an arm 64 processor okay and um i think apple won't actually let you run
[3866.42 --> 3875.42] 32-bit processes on some of the newer ios versions on the arm 64 processes so for this go mobile stuff to
[3875.42 --> 3881.60] work you know we need to we need to target arm 64 so that that's what arm 64 is about um there are
[3881.60 --> 3886.66] probably some people doing um processes in this in the server environment but i'm not really aware of
[3886.66 --> 3892.66] them power pc 64 is something i don't have a lot of visibility into but i know that you know there
[3892.66 --> 3898.50] have been contributors from um companies like canonical who are who are pushing that support as well as
[3898.50 --> 3904.16] other people in the community and people from google you know power pc is is is an ibm
[3904.16 --> 3914.56] processor architecture and um so i think the power pc 64 machines are very very high spec multi-core
[3914.56 --> 3922.78] server machines and so the kind of kind of processes that you you wouldn't have much to do with as like
[3922.78 --> 3928.42] an average developer um but i think it's it's possible that we'll see them making inroads into the
[3928.42 --> 3932.80] server market i'm not really i don't really have a lot of visibility into it but i think it'll make some
[3932.80 --> 3941.04] particular sort of enterprise and customers happy and if you're trying to supplant java then you've
[3941.04 --> 3948.20] got to do that right yeah i mean you know the go tool chain has been a cross-platform thing since
[3948.20 --> 3955.04] the very beginning um and part of the rewrite actually has made it much much easier to support
[3955.04 --> 3961.98] other platforms like the actual uh architecture dependent part of the tool chain has gotten smaller and
[3961.98 --> 3968.66] smaller over time and now it's actually quite small so we just we just want to support go everywhere
[3968.66 --> 3974.36] that we can um and if there are people willing to help contribute that support then we'll happily
[3974.36 --> 3980.08] help them do that let's talk about the builder infrastructure real quick i know we don't have
[3980.08 --> 3987.28] much time but i was really impressed by go mode and google compute engine so talk a bit about the
[3987.28 --> 3992.06] the uh google compute engine and go mode what you were doing there that was that was pretty amazing
[3992.06 --> 3998.76] right well so the basic backstory is you know since go is so cross-platform we need to test it on all the
[3998.76 --> 4005.76] platforms that we support and so you know we have this sort of homespun builder infrastructure um that
[4005.76 --> 4011.40] you know we have a build dashboard and we have these these machines like this heterogeneous array of
[4011.40 --> 4018.44] machines that are all around the world like mac os machines linux and free bsd net bsd open bsd plan 9
[4018.44 --> 4026.94] windows and all these machines run this go binary that like fetches our latest revision of go builds
[4026.94 --> 4032.30] it runs all the tests reports the results back to the dashboard but you know maintaining this array of
[4032.30 --> 4040.08] machines is total pain right it's it's they're all different they all work differently they're all owned by
[4040.08 --> 4046.02] different people um you have to like email someone when something goes wrong and then if you have a
[4046.02 --> 4051.02] problem on a particular architecture you know you need to um get access to one of the builder machines
[4051.02 --> 4057.90] to be able to actually test your code on whatever that processor and architecture is but so brad fitzpatrick
[4057.90 --> 4064.32] mostly and a bit me have been working on some new builder infrastructure that uses compute engines so
[4064.32 --> 4071.26] for all of the all of the operating systems that we can we're running those on virtual machines on
[4071.26 --> 4076.32] compute engine and so we have this kind of like deterministic build environment that we can spin up
[4076.32 --> 4082.88] at will and so we can do many many more builds in parallel than we could before and also now we can do
[4082.88 --> 4089.62] we can do speculative builds so if someone sends a change we can run uh the tri bot on it as we say so
[4089.62 --> 4094.96] you get to see whether the change builds before you actually merge it into the tree and so that's
[4094.96 --> 4101.06] been really nice um currently works for like windows linux uh freebsd and a couple of others
[4101.06 --> 4108.18] um but uh we look forward to doing something similar with mac os it'll be a little bit different since we
[4108.18 --> 4113.80] can't run that on compute engine but we have plans for how that might work um but a nice side effect
[4113.80 --> 4117.94] from that is we have this tool called go mode where if you're a go developer and you see that your
[4117.94 --> 4125.60] change breaks on like open bsd um you can spin up an open bsd instance on compute engine with just one
[4125.60 --> 4132.40] command line invocation and then you know push your local changes to that machine run them see what
[4132.40 --> 4137.56] happens and you know you get to actually sort of develop on the architecture that you're trying to
[4137.56 --> 4143.64] support and then you don't have to actually have a bunch of vm sitting around or actual machines if
[4143.64 --> 4147.48] that's the route you choose to go if you're a developer for go then this makes it a little
[4147.48 --> 4152.46] easier to yeah sort of build you know write code for many but uh not have to actually own those
[4152.46 --> 4157.36] machines yeah the one of the really nice side effects is uh we'll be able to make this available
[4157.36 --> 4164.58] to go programmers in general um so this will be free to to anyone who's developing for go yeah yeah
[4164.58 --> 4173.92] so you can just like um plug in your own google uh cloud project uh credentials and then it will spin
[4173.92 --> 4180.66] up instances on that you know that you pay for um but you only have to pay for like you pay by the
[4180.66 --> 4187.86] minute um so it's pretty much something maybe yeah yeah probably less even for for an hour or something
[4187.86 --> 4193.26] but um so so it means that if you're a go programmer and you know you want to test your stuff in other
[4193.26 --> 4198.70] places um you should be able to do that pretty easily and also i think it's just a nice kind of
[4198.70 --> 4204.06] demo of cloud orchestration stuff although you know there are a lot of tools in that sphere
[4204.06 --> 4210.14] um i don't necessarily recommend writing tools like this for us it was kind of an experiment to see
[4210.14 --> 4215.94] how our cloud libraries were in ways in which we can make that better um because obviously you know
[4215.94 --> 4220.14] the go team here at google we have a close relationship with the cloud team at google and we're very focused
[4220.14 --> 4226.62] on trying to make um you know go the best language to use on google's cloud platform and also other cloud
[4226.62 --> 4233.08] platforms and so you know we we've we've found a lot of weak points things that we want to improve
[4233.08 --> 4240.50] um and you know part of building this project was uh was just seeing how good our offerings are you
[4240.50 --> 4243.90] know we're pretty happy with it but obviously you know we find things that we want to improve and
[4243.90 --> 4249.62] we're working on those too all right last question before we go into a couple of closing questions
[4249.62 --> 4254.72] which are really short but i can't let it go without asking this question which is you've been here
[4254.72 --> 4261.06] twice now was it has it been three times you've been no it's been twice rob was on way back when
[4261.06 --> 4266.26] then you were back with rob on episode 100 and then now you're back here for episode 148 and it's been
[4266.26 --> 4273.40] a year and a half since you've been on the show so what is the future for go like you got this last
[4273.40 --> 4279.92] you know 1.4 to 1.5 we're shipping later this year what what beyond that where are you seeing past like
[4279.92 --> 4289.58] 1.6 or or beyond i really think you know as far as like the go core is concerned you know it's going
[4289.58 --> 4294.30] to get faster it's going to run in more places it'll be more efficient you know better optimization in the
[4294.30 --> 4302.44] compiler optimization in the libraries you know improvements to the tool chain maybe you know
[4302.44 --> 4309.80] some new sort of developer tools that kind of thing but i i think that you know the most exciting
[4309.80 --> 4316.60] developments around go are really in the greater go community and actually one thing that i'm involved
[4316.60 --> 4323.94] in is a project called go kit which a guy called peter borgon from soundcloud
[4323.94 --> 4331.32] has initiated and it's an open source project that's basically trying to build a standard
[4331.32 --> 4339.32] library for building distributed systems so it's kind of described as like a toolkit um the the purpose
[4339.32 --> 4346.70] is you know if you want to build distributed systems in go so a lot of a lot of the this the cloud
[4346.70 --> 4353.88] services that people build are basically distributed systems and there are or people often call them
[4353.88 --> 4359.58] like micro services you know you have um many services to talk to each other via rpc systems
[4359.58 --> 4363.06] if you know what i'm talking about then you know what i'm talking about but basically
[4363.06 --> 4372.02] basically um go kit is is trying to define like a set of things that people need to build these kinds
[4372.02 --> 4380.22] of systems and then provide like a canonical sort of set of recommendations or even like libraries
[4380.22 --> 4387.66] and interfaces that you know various tools and and libraries can satisfy so that you know you can
[4387.66 --> 4392.74] have a well integrated developer experience for building distributed systems in go so that's that's
[4392.74 --> 4397.68] kind of like the goal you should check out peter's talk um go in the modern enterprise which he gave at
[4397.68 --> 4403.76] fosdem uh for for we're linking that up in the show notes so yeah we'll definitely have that talk in
[4403.76 --> 4408.22] there yeah you know speaking of go kit we were actually wanting to have peter on the show but i wanted to
[4408.22 --> 4413.56] have him on after we talked to you to kind of get an update on where go has been at for the last
[4413.56 --> 4417.48] year and a half and what you've been doing so i wanted to have you on there and i didn't know you
[4417.48 --> 4421.36] were working with peter on this so it might make sense to have you back on with peter if that's the
[4421.36 --> 4426.20] case yeah well when i you know when i saw his talk it really i was like oh my god you know he's just
[4426.20 --> 4431.00] really clearly articulated a lot of the same things that i'd been feeling about go you know we really
[4431.00 --> 4438.12] need to focus on making this work um he called it like the modern enterprise which is sort of these
[4438.12 --> 4446.38] you know medium-sized companies that are building you know these distributed systems um and i i feel
[4446.38 --> 4451.38] like go is a great language for those companies but they we we really need to focus on making it
[4451.38 --> 4455.86] clearer and easier for people to make the right choices when building those kind of systems
[4455.86 --> 4460.96] and also i think there are a lot of people in the go community working on these problems and we've
[4460.96 --> 4465.82] actually seen it with gokit you know when it was announced a lot of people joined the gokit project
[4465.82 --> 4471.22] and they were like oh i've been working on x and the number of people that had kind of all been
[4471.22 --> 4476.26] working on the same x in parallel it's like oh we should just focus those energies together