Fork vs Continue development

Jul 2, 2012 at 8:34 AM


After checking several different projects ( the usual NServicebus, Masstransit etc) i have started using Simple Service Bus over a 2 years ago and have it running in production for over a year ( about 550-600 endpoints running it). I love it's simplicity (duh) and the fact that the mechanics are actually understandable without having to read half the code base which is the case with some of the other projects..


However, ever since the beginning, there have been things bugging me: the API could use some freshening up and the whole codebase could do with some reorganisation and cleanup / updating (this is offcourse partially personal taste)

Next to this, some very cool features are missing where for me, saga's for example, are one of the most important together with rabbitMQ.

I have been toying around with the idea to write it myself (looking at masstransit code could give a start as to where to begin, i like their fluent api a lot) but have been in doubt:

Do I start a clean project , Fork simple service bus OR is there enough interesst here to further develop Simple Service Bus ?

I see that allthough the community seems to be small, many would not want to switch to any of the other solutions , and neither would i. 

I actually started reconsidering MT now that they have RabbitMQ support and since they have working saga's but looking at the code and samples again, i started shivering thinking of how "simple" SSB does all this (minus the saga) and uses the MQ itself to do the heavy lifting ( which is it purpose anyway) , it makes some of the other solutions just look like "a hack"


As I see it, If choosing the SSB way, i would love to first of all cleanup libraries, upgrade where needed so there is a new cleaned up version.

After that, the code could be cleaned up and the API tweaked and reorganised. After that, new features could be added. (where for me the saga engine would be primary.

Is this something others would want/support ?

Jul 9, 2012 at 4:33 AM


thanks for posting. I'm also still using SSB, processing almost a million messages a day. I'm excited about your interest in SSB, and I'd be happy to upgrade to your cleaned-up code, but unfortunately my main project that's been using SSB is in a state where we probably won't spend a lot of time on the infrastructure (where SSB comes in).

Please let me know what way you're leaning and let me know what I can do to assist.