Open cjdelisle opened 7 years ago
I've just come across this project, and I was able to follow the instructions in a matter of minutes using redis & mongo via Docker...
I would however, recommend moving the docker instructions from the last section of the README to the top somewhere- so that people looking to do a "trial run" can do so in a simple command. I read the README top to bottom... and then realized darn, they provided a docker-compose image already.
Most users who come across the project would like to try it out first, and then if it works or if it almost works, then they'll scroll down and see the rest of the README when they're looking to make contributions.
If the Docker solution isn't suitable, then perhaps OmniBus? this is how GitLab provides an easy to install package, which manages it's own dependencies (such as redis) embedded in /opt/gitlab/embedded/
@moriarty you are right, this is 2017 and I totally agree that we can quickly setup working environment by Docker. We will improve our instruction but you can also contribute and that will be super cool.
I was told by a colleague that he was interested in possibly deploying hublin but he found the install instructions too difficult to warrant a "trial run", specifically he cited the installation of redis and mongo as sticking points because they need to be installed globally on the computer and therefore can have negative impacts on the server, whereas embedded databases such as leveldb or sqlite can be integrated in the npm install process and will remain local to the hubl.in installation itself.