Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

HashiCorp’s multi-product strategy

foundationinc.co

Post reply on HN
Print view View original

41–50 of 91 posts

Re: HashiCorp’s multi-product strategy

#41

Post by mixedCase »

datatrashfire wrote: post #29
jen20 wrote: post #17

Earlier quoted context omitted.

I'm assuming this is Terragrunt, which itself perpetuates all kinds of horrific practices which are absolutely unnecessary if you apply basic software engineering principles to infrastructure as code/config.

Terragrunt is an antipattern. Made the mistake once with a very poorly structured project I inherited, never again.

Terragrunt is not an anti-pattern, but it is a tool that is easier to get wrong than Terraform and so people create great messes with it.

Used right it's the only way to remain sane on any non-trivial set of infra where you don't end up hand-rolling a half-baked version of Terragrunt.

Re: HashiCorp’s multi-product strategy

#42

Post by swyx »

why is this absolute dreck on HN?

nothing in this piece gives any indication that the authors have any idea what Hashi even does. the extent of analysis is "hurr durr cloud computing is big" and then a long nondescript "Open Source is Key to Product-Led Growth" buzzword bingo that reads like a noob discovering how bottom up open source works for the first time

i need a downvote button at the post level, i want a refund on my time spent reading this shit

Re: HashiCorp’s multi-product strategy

#43

Post by LtWorf »

14u2c wrote: post #13
throwaway787544 wrote: post #9

> There’s company-wide alignment on delivering the best user experience possible. That means sales, marketing, product engineering, product design, and customer success all channel their efforts into creating a product that provides an unparalleled UX. Product development puts all departments on the same wavelength in terms of the end goal. Who the fuck are they talking about? Clearly not HashiCorp. The most popular…

I'm assuming you are talking about terraform? The UX of all the other hashi products I've used have been absolutely sublime compared to the industry equivalents. Nomad compared to kubernetes is a particularly stark contrast.

Yeah vault is great… unless you need to store 4096bit keys on a postgres backend… can't do that.

Re: HashiCorp’s multi-product strategy

#44

Post by girvo »

kfrzcode wrote: post #37
moralestapia wrote: post #31

Earlier quoted context omitted.

>HashiCorp’s tech is a mess of over engineering, poor UX, and insufficient documentation. Second that. Very steep learning curve for some use cases that could be accomplished in a much easier way. Another kubernetes in disguise.

k8s in disguise? Vault? Have you used either of these tools?

They're talking about Nomad here, with the "tech" in the GP comment widening the discussion, I assume.

Re: HashiCorp’s multi-product strategy

#45

Post by hericium »

Vault is great. But if I were to start learning Terraform now, I'd probably ditch it in favor of Pulumi.

Terraform was supposed to make writing DRY easy. It's often more reasonable to copy-paste parts of configuration than to try make it DRY. HashiCorp seems to have the approach of "if something can be hacked around, even in a very convoluted way, we can ignore any feature requests for that thing to be doable in a normal fashion".

Instead of fixing Terraform, Hashi is creating more and more products that likely will be increasingly unpleasant to use and won't get reasonable fixes, just as Terraform.

The way HashiCorp approaches creating an ecosystem, makes me want to stay away.

Re: HashiCorp’s multi-product strategy

#46

Post by jillesvangurp »

Open source rules the world. A lot of people use HashiCorp's OSS products; not a lot pay for them. I certainly haven't and I don't think I've ever met a paying Hashicorp customer even though I know lots of people that use their products.

They are a consulting business ultimately. And like many such companies they thrive on complexity. They are doomed to become part of the problem they are trying to solve. If they succeed in their mission, the result is OSS software that is so simple to use that nobody needs their consulting services.

Like many OSS companies, they confuse control over the software with controlling the wider market. Amazon, Google, MS, etc. don't provide just software but their 630B dollar industry that is enabled by their massive data centers and infrastructure. That's what people pay for.

Hashicorp is a small fish in that pond as they don't really own or control any such infrastructure. I'd say they might end up being gobbled up by a bigger consulting company. Oracle and IBM come to mind.

Re: HashiCorp’s multi-product strategy

#47

Post by raffraffraff »

hericium wrote: post #45

Vault is great. But if I were to start learning Terraform now, I'd probably ditch it in favor of Pulumi. Terraform was supposed to make writing DRY easy. It's often more reasonable to copy-paste parts of configuration than to try make it DRY. HashiCorp seems to have the approach of "if something can be hacked around, even in a very convoluted way, we can ignore any feature requests for that thing to be doable in a no…

I found a way to make Terraform code completely dry by using Hiera as a data source and writing every stack and module to be entirely data-driven (ie: only ever use "for_each" to iterate over a data map, never hard code anything, use strict naming conventions). However there are some major obstacles to using Terraform this way, that emerge from it being a clunky DSL with hidden "special rules", instead of behaving in a consistent way like a proper language. Like these:

https://github.com/hashicorp/terraform/issues/31072

https://github.com/hashicorp/terraform/issues/24476

Re: HashiCorp’s multi-product strategy

#48

Post by sverhagen »

jillesvangurp wrote: post #46

Open source rules the world. A lot of people use HashiCorp's OSS products; not a lot pay for them. I certainly haven't and I don't think I've ever met a paying Hashicorp customer even though I know lots of people that use their products. They are a consulting business ultimately. And like many such companies they thrive on complexity. They are doomed to become part of the problem they are trying to solve. If they suc…

Honest question/suggestion: don't you think many big companies are more than happy to pay for the consulting to give them a feeling of comfort and safety, someone to turn to if there's an outage in the middle of the night?

Not that I'd need much Hashicorp folks on my "bridge call" when my Websphere cluster runs out of memory, but since IBM and Microsoft are on speed dial (in their million(s) dollar support contracts), we may want the same for Hashicorp. This "we" is not "me", but I did work for this sort of company in the past. Open source made them uncomfortable, consulting may help with that.

And sure, it lasts only as long as it does.

Re: HashiCorp’s multi-product strategy

#50

Post by wernerb »

datatrashfire wrote: post #29
jen20 wrote: post #17

Earlier quoted context omitted.

I'm assuming this is Terragrunt, which itself perpetuates all kinds of horrific practices which are absolutely unnecessary if you apply basic software engineering principles to infrastructure as code/config.

Terragrunt is an antipattern. Made the mistake once with a very poorly structured project I inherited, never again.

"Code generation" is pure evil.

I second that Terraform needs to work at least on dynamically specifying providerblocks. This is where people usually resort to terragrunt.

Workspaces and state layering are in my experience hard on the novices and unfortunately they turn to terragrunt.

Post reply on HN