Updates On Zenbo And Zenbo-hosted
September 23, 2012
After a week of doing nearly nothing, thanks to a really bad cold, I managed to get some testing and coding done. I thought I give you some updates how your favorite project is doing and how soon you will see the release of the last blogging platform you will ever need.
This is one of the names and slogans that were suggested for Zenbo-hosted. While I liked this one, even if it maybe not 100% accurate, here are some more.
- yet another blogging platform
- static blogging
- scaled blogging
- blogging with fun again
I also thought about just using Zenbo. Name the service after the core software that is used. I had far worser ideas, believe me. The final decisions will be made in some weeks, so I am not really stressed making my mind up.
Today Zenbo learned what tags are and how they work. The documentation is not really finished but I did not find any bugs in the implementation so I just pushed it to the repository on GitHub.
One of the main things I have to fix before the first release will be the way Zenbo reads and uses the configuration. A plugin should ship with sane defaults. Currently you have to add all configuration options to
config.yaml which is not really comfortable, fast or clean.
Also it will be a lot easier to integrate Zenbo in zenbo-hosted if the configuration is internally represented as an object.
I think we will see the last beta version next week. After adding some more tests and finished the crawler everything should be ready for 1.0.0, except the documentation which is a different topic.
I have a basic implementation of the backend. Currently only working if the backend and web server are running on the same server. Separating the backend from the fronted needs some time. I am not sure if I will just use Fabric or integrate SSH to run commands on a remote system. Fabric seems to be the better way.
Adding multiple fronted servers on the other hand is already integrated. To be honest I am just to lazy to change the database scheme, even if it is quiet easy thanks to South.
I am also not happy with the way I create git repositories. I subprocess too much for my taste but finding a python git library that does not suck is not as easy as it sounds.
Pricing is also nearly set. Free for non-commercial use. I am not sure how to price companies but I thought about something like 10k requests / $x. Not sure what x needs to be to make this service sustainable.
I have some work to do the next two weeks but I think I will be able to release an early beta in October.