Also: container orchestration if _far_ from solved problem. Kubernetes is a nice try, but it screams for a better layer above it that would allow devs to deploy services _easily_.

There is plenty of ways Docker Inc. can create and capture value in this space. Their unique value is that whatever product they make, the world will check it out... They "just" need to make it good and figure out the business behind it.

Dockerhub doesn’t even support MFA, not to mention their IAM is all or nothing (there’s only 2 roles). I don’t think they’re focusing on the things that they should be.

Well, if you're focused on MFA and elaborate IAM for Dockerhub as "big problems", then I don’t think you're focusing on the things that you should be.

Those are more like enterprise checklist items than real needs...

Yes. As a response to a comment talking about enterprise support contracts as a main source of revenue for Docker, I think enterprise checklist items are entirely valid items to focus on.

I guess if the focus is "enterprise support contracts", that's legit.

Well this is the inherent push-pull of the industry, isn't it? The ones with money generally have insane compliance requirements (because they have money, they have more risk); the places without money care more about getting things done (since they're trying to make money...).

If it weren't so, we'd all just sell our hot new WhateverAsAService to banks straight away, right?

That's not a billion dollars company though.

Seems like it's more valuable and ubiquitous than mongo db, and their valuations exceed 1.5B

Ubiquitous does not equal value. Docker doesn’t do anything you can’t do with lxc tooling and an object store (registry). That’s the benefit: containers are free, the orchestration is free, and the registry tooling is free and easily replicated.

Mongo equally so (JSON in PostgreSQL). These valuations are fantasy.

Building a scalable service on top of that is not free, nor easy :)

You're spending those scaling dollars regardless (developers with devops knowledge, infrastructure/SRE folks). Better to spend it on competent staff than a company lighting VC money on fire, no?

> Better to spend it on competent staff than a company lighting VC money on fire, no?

False dichotomy though. You either acknowledge that getting a vendor to do that for you also implies competent staff or you acknowledge that hiring "competent staff" is not like flipping a switch and requires training, ramp-up time, few wrong hires, getting managers to guide the teams, etc

I'm not sure which way is better, but I don't think the industry has decided either :)

i think it's a bit more complicated than that. very skilled developers that know this tech won't jump ship to many borderline broke startups which could easily just get started with docker with less than 'rockstar' devs

It's already a billion dollar company.

https://techcrunch.com/2018/10/15/docker-has-raised-92-milli...

That's the point of parent and the article: It has a billion dollar valuation, but they doubt it can back that up with actual revenue and profits.

Of course not, a product company like docker is built for acquisition.

I can think of a few companies that might want to buy Docker. I can think of no company that would buy it at a billion dollar valuation.

Microsoft's containerization strategy heavily depends on Docker (they've got a licensing deal so that Windows server can use Docker EE for free) they might well pay good money to by Docker

Didn't IBM just buy Red Hat for $34 Billion? Can you imagine a future where Amazon acquires Docker to optimize it to run on AWS and perhaps perform a bit worse on GCP. Any small advantage you could get in the space would equate to large changes with as ubiquitous Docker is and as large as AWS is.

Perhaps but given AWS is a heavy Xen and KVM shop something like Kata Containers might suit them better.

I can think of a few companies that may pay an absurd vaulation for Docker; IBM and Microsoft.

Fun fact: if you use Red Hat, you have to use Docker EE. You don't even have the option of getting CE and taking responsibility for Docker support yourself.

Could you explain that a bit more?

> if you use Red Hat...

"Red Hat" as in RHEL specifically, or are you meaning any of the family (eg CentOS)?

Asking because I've not hit any problems using Docker CE with CentOS 7. Well, aside from general bugs (etc). But nothing seems to force the use of EE instead of CE.

Docker CE is officially supported (well, it's not supported because its CE, but it's packaged, repos, install guide) etc. but only EE is packaged for RHEL. RedHat also has their own docker package like many other distros, so you can still get it free easily, but you have to get it through them and wait on them for updates.

If you're using RHEL why wouldn't you default to using Red Hat's build of Docker? I definitely prefer it to the official builds. Red Hat in general seems very annoyed at Docker and carries quite a few QoL patches like being able to add registries to search.

That isn't true. You can also use podman, and cri-o in the case of Kubernetes.