text stringlengths 0 2.46k |
|---|
[1288.54 --> 1294.64] but not a lot of like guaranteed support so it's like how much of this is support and name only |
[1294.64 --> 1300.70] like yeah we're behind you put our name on the website but are we actually gonna like because |
[1300.70 --> 1304.96] it takes a lot not just up front you get the energy and the excitement and everybody slapped |
[1304.96 --> 1310.38] their logo on in the beginning but like over the course of years to support a project like that's a |
[1310.38 --> 1315.52] that's an ongoing initiative that requires dedication and how many of these companies are |
[1315.52 --> 1322.58] actually dedicated obviously time will tell but he didn't seem too worried about it so yeah i listened to |
[1322.58 --> 1327.02] that episode i heard the question i was waiting for a concrete answer as well we didn't get like |
[1327.02 --> 1332.18] a super concrete answer which is fine yeah you know they're still early but i agree i i time will tell |
[1332.18 --> 1337.10] on that i hope they can maintain it because it's it's a beast of a of a tool to maintain the people |
[1337.10 --> 1341.32] that work on terraformer have been there for quite a few number of years built up the context around it |
[1341.32 --> 1345.96] right it's it's a pretty decent large code base and it's a complex problem domain right like the whole |
[1345.96 --> 1350.38] idea of terraformer is just you're graphing your infrastructure and you're making api requests so if you |
[1350.38 --> 1354.88] don't understand that whole idea of graphing and whatnot and dependency resolution it's gonna be |
[1354.88 --> 1358.90] a little bit of a difficult thing for you to contribute to the question is will you be a |
[1358.90 --> 1364.50] contributor uh i think so i've already contributed to some of the terraform providers so i probably |
[1364.50 --> 1368.68] keep contributing in that respect i haven't contributed i have i think i have contributed to |
[1368.68 --> 1373.58] core maybe like small little contributions but nothing major to the core code base does cockroach |
[1373.58 --> 1377.56] choose terraform yeah they have a terraform provider but they don't use it for their production |
[1377.56 --> 1383.32] infrastructure gotcha yeah they're on palumi i believe from what last i heard we'll see i guess |
[1383.32 --> 1389.24] right yeah yeah we'll see there's at the end of the day infrastructure is if you think about it it's |
[1389.24 --> 1392.80] a solved problem we know what we need to do with it we need to spin it up we need to manage it |
[1392.80 --> 1397.80] the tool that you use use the best one for your team right the one that's going to provide you the |
[1397.80 --> 1402.96] best benefit that's the one you should be using curious if you have takes on some of the |
[1402.96 --> 1409.60] more recent releases in the infra world system initiative the stuff that the dagger folks are |
[1409.60 --> 1415.42] doing so what's interesting to you yeah the dagger stuff's interesting i heard about it in the podcast |
[1415.42 --> 1421.68] saw like looked at the website whatnot haven't used it yet have not used it system initiative however |
[1421.68 --> 1428.64] i have used i've contributed to and i interviewed them okay yeah so excited about that yeah i'm very |
[1428.64 --> 1432.74] excited about the system initiative stuff yeah adam jacob great person i know i think you had him on |
[1432.74 --> 1438.44] the show right yeah we call him a friend there you go perfect yeah perfect yeah great great people |
[1438.44 --> 1443.84] there um a couple former hash core people that are there talked to a few of them they have a wonderful |
[1443.84 --> 1448.30] discord that if you are really interested in system initiative go join they're wonderful people |
[1448.30 --> 1452.32] they do everything out in the open as much as possible and that's how i got involved so i interviewed |
[1452.32 --> 1456.38] with them uh didn't get that that role they went with somebody else because you know startups |
[1456.38 --> 1460.18] there were only like 14 people yeah they're small right they're small so they got to be very very |
[1460.18 --> 1465.44] picky which is great but then i liked the product did the beta like went through the beta testing or |
[1465.44 --> 1470.02] whatnot gave them feedback and all that and then i contributed their podman driver to run system |
[1470.02 --> 1476.80] initiative on podman is it the future is it the future is a good question i don't i don't know i i like |
[1476.80 --> 1482.14] the ergonomics of it a lot honestly it's very fun because if some like when you're thinking about |
[1482.14 --> 1487.16] infrastructure the one thing that really like left a bad taste in my mouth with terraform is when |
[1487.16 --> 1491.16] you're trying to find out what risk like what other resources you can use with this resource it's |
[1491.16 --> 1494.86] very difficult you need to know the name of the resource that you want to use right like finding |
[1494.86 --> 1499.12] the dependencies and the connections between them is tough you have to look in their docs and the |
[1499.12 --> 1505.40] docs are there's a lot but with system initiative you drag an asset onto the pane and you know the |
[1505.40 --> 1508.42] dependencies that you can use with that resource you know what can plug into it you know what it can |
[1508.42 --> 1513.52] output to and that's great like i thought that was cool so from the visibility of how you can like |
[1513.52 --> 1519.26] build your graph of infrastructure i think system initiative is great in that regard uh outside of |
[1519.26 --> 1524.12] that like there's obviously they're still in very very early release phase so they have like a few ui |
[1524.12 --> 1530.32] things to smoothen out but i don't know is it the future again the future will tell what the future |
[1530.32 --> 1535.52] is right do you think it could be just a ui that others like could it be a ui on top of terraform |
[1535.52 --> 1541.86] for example like could it become the the interface that we begin to use to orchestrate services and |
[1541.86 --> 1547.50] infrastructure and stuff like that rather than just being its own silo it's possible they could |
[1547.50 --> 1552.70] open that up because they have like the capability technically under the hood all the assets are just |
[1552.70 --> 1557.16] typescript under the hood right so it's like a function or whatever you run as long as you write |
[1557.16 --> 1561.20] it in that interface way you're good i think so would they want to do that i'm not sure |
[1561.20 --> 1567.72] right that seems to be the most innovative thing really yeah and what it offers right is the |
[1567.72 --> 1574.48] visual interface to connect the nodes and see the dependencies rather than scouring through yaml or |
[1574.48 --> 1579.62] whatever else you might have for configuration exactly that's challenging right it's it is and |
[1579.62 --> 1584.70] the example they they run you through in the beta is basically spin up ec2 instance security group |
[1584.70 --> 1590.42] sh key right you just put all together and you see a graph in your aws region and whatnot and it's all |
[1590.42 --> 1594.96] graph really nice for you and you get to apply it and they you know similar to terraform they have |
[1594.96 --> 1599.92] their graph based way of applying so dependencies get created first and blah blah blah which is great |
[1599.92 --> 1606.40] um i like their extensibility though so for terraform if you want to extend terraform you need to |
[1606.40 --> 1610.74] contribute to the provider if there is one if there's not a provider you need to create a provider |
[1610.74 --> 1618.44] build that binary ship it in system initiative side you can just edit the typescript or you can go in |
[1618.44 --> 1621.94] their job typescript functions in and now you have a new asset to manage so like from the |
[1621.94 --> 1627.10] extensibility side i think they have a like a more extensible platform you know for the average |
[1627.10 --> 1632.50] developer right right right and you're an ops guy who likes typescript i don't actually use a lot of |
[1632.50 --> 1636.78] typescript i it's i use a lot of go i use a lot of go you don't seem to have a problem with that |
[1636.78 --> 1642.06] i don't it's it's typescript very readable right it's like it's not my favorite language but coming from |
[1642.06 --> 1647.60] another you know strongly typed language typescript works for me happy to hear that i just remember |
[1647.60 --> 1652.76] that one of the i think kelsey was on saying that one of the concerns is that it needs to be |
[1652.76 --> 1657.94] multilingual specifically you know back-end folk infrastructure folk aren't going to want to use |
[1657.94 --> 1664.60] typescript and so counterpoint yeah i think if you're like if you're a good engineer the tools |
[1664.60 --> 1670.24] matter less than knowing how to use them correctly amen right that that's what it is so if you if they're |
[1670.24 --> 1675.38] using typescript and i know how to use typescript and to do what i need to do why do i care so much |
[1675.38 --> 1681.28] right i'm using the thing it's okay right yeah i'm using the thing so for me it works out like |
[1681.28 --> 1687.26] do i maybe wish it was something like go or rust maybe but everybody knows typescript right like at |
[1687.26 --> 1691.98] this point it's it's it's a pretty ubiquitous language i think it's a good first choice for them |
[1691.98 --> 1697.92] if they want to expand later it's got a wide footprint of users it does it really does and then so not |
[1697.92 --> 1702.74] regard it smart yeah when you contrast it with something like palumi who like supports many |
[1702.74 --> 1708.38] languages i don't know if that's the right choice i think when you give people too many choices you |
[1708.38 --> 1714.44] fall into that like analysis paralysis situation right where you're like what choice what language |
[1714.44 --> 1719.54] do i use if team if this team is using python and that team's using go can they like contribute to |
[1719.54 --> 1724.26] each other's stuff or am i creating silos right so i don't know i don't know the right answer but |
[1724.26 --> 1728.24] yeah we'll see well it's a different thing so one thing that solomon said on the show about |
[1728.24 --> 1733.28] dagger is like they were qlang you know yes which is basically yaml on steroids if you don't know |
[1733.28 --> 1739.54] about it it's type it's strongly typed configuration language and that was a real hang-up for people |
[1739.54 --> 1745.32] because they wanted more power and so now they went the other direction right go sdk elixir sdk |
[1745.32 --> 1753.60] typescript sdk etc and i wonder if if like that distinction is significant from a declarative |
[1753.60 --> 1759.76] yaml-esque thing to a programming language but once you get to that point the language itself |
[1759.76 --> 1766.54] is less significant right yeah i think the win there is getting off of like the dsl yeah for them |
[1766.54 --> 1770.48] and then giving the opportunity to really just plug in whatever you need i don't know if it's |
[1770.48 --> 1775.66] right code right it's a right code proprietary versus whatever's out there exactly yeah exactly |
[1775.66 --> 1779.82] because there are there have been a lot of people that i've i work with so many customers over my time |
[1779.82 --> 1785.96] hashicorp that they asked for loops like proper like loops in programming languages and they had |
[1785.96 --> 1793.56] good use cases for them and you know the hashicorp hcl dsl wouldn't really enable that in that regard |
[1793.56 --> 1797.72] so yeah it's interesting to see what's out there though i'm i'm excited for all these new tools and |
[1797.72 --> 1802.92] i wish when i was doing more ops work earlier in my career a lot of these tools existed because it |
[1802.92 --> 1807.68] would have gave more choice i was kind of stuck with bash and ansible and in a sense right so |
[1807.68 --> 1812.68] well man appreciate you stopping by and telling the story thanks for having me it's a fun deep dive |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.