[the_real_bto]

There are a lot of problems to solve, and docker by itself only gets you part of the way. Orchestrating docker containers is the interesting problem, not so much Docker in itself (for me anyway). The handwringing that docker should do less (as a company), because each company should only solve a small problem is just stupid.

If the argument is that the tools should be small and composable, then I agree 100%. Maybe image distribution should be a separate tool from the tool used to run a container. That is actually a pretty good point.

[23david]

I think the "handwringing that docker should do less" has a lot to do with the fact that just getting Docker containers right is a huge effort in itself, and they have a long way to go even there. It's not a solved problem by a longshot.

[the_real_bto]

I agree, there are problems all around. I was responsible for getting docker adopted on a fairly small (but a number of moving parts) software project. It has been somewhat painful.

[angelbob]

It's less "handwringing that Docker should do less" and more "get one thing right first."

[the_real_bto]

Agreed. I was referring to the rocket/systemd remarks. There were some pretty big docker hate going on when rocket's announcement was posted here on HN. Most of that hate was because docker was getting into the orchestration business.

[skj]

I don't think it's the fact that docker-the-company is working on these problems, but rather the fact that they're tying everything together very closely, making it hard to use one thing they're working on without the other for no fundamental reason.