# No docker

It looks like I succesfully built Hived 'by hand'.
I did not use anybody's image of the program and made no image of the program.
I simply downloaded the source and compiled the program myself.
I still don't _really_ understand why I would use docker.

![image.png](https://files.peakd.com/file/peakd-hive/felixxx/23tS2thTx9dzKRCaHvCQ9EQYYvMQLkyc1mDcmoT354zbzD4qNwkvuLFeAXxf3keh3mAvt.png)

The node is up and running.
I just need to set up my configuration now.
That is also poorly documented, if at all.

# Improvements

Node replay went quick. 
When I last tried this, it took **days**.
Looks like, they removed a lot of ballast.
What I understand is, I could disable some things on my witness node, that before I could not easily disable; Things that only API nodes or other types of nodes need.
I appreciate these improvements. They lowered the barrier of entry to some degree.

# Regress

I understand that with docker you have a lot of advantages, when managing and setting up your nodes. At the same time, you do not have to understand much.
The long list of dependencies I mentioned yesterday, are only needed to _compile_ the software.
Once that is done, there is only the config left.
What annoys me is, that there is little documentation about it, and it's scattered and unprecise.

https://gitlab.syncad.com/hive/hive/-/blob/master/doc/building.md

``If at any time you find this documentation not up to date or unprecise, please take a look at CI/CD scripts.``

I did and guess what: can't figure it out.

https://gitlab.syncad.com/hive/hive/-/tree/master/ciscripts

In none of these files I can find the different options to compile with and not a single piece of description. It looks like what I went with worked, but I am not exactly sure, what I did and why it worked to be honest.

# Final Configuration

What's left to do now is editing the config file.

https://developers.hive.io/nodeop/node-config.html

That is all I can find and it is not really helpful at this point.
I will have to look into the docker images of other people, instead of a documentation.

# Final Thoughts

Can't say I learned much from this adventure.
I am still struggling with some basic Linux and web protocol questions and problems, which doesn't help.
Generally, it feels like these informations are unprecise by design.
'They' seem to create this barrier of entry, yet at the same time 'they' promote dockerized nodes... I don't get it.

Then again, gtg is a wizzard; Great at code, terrible at explanations. It might not actually be intentional, just a typical programmer's problem.

This is kind of the reason why I think I should run for witness again: Maybe I am able to be an intermediate; Trying, explaining things to other intermediates, while the top ranks can not be bothered with such things.

#### Expect a witness proposal later today