Who would maintain the IBM code? Having dead code is never a good idea, imho.
From: paho-dev-bounces@xxxxxxxxxxx [mailto:paho-dev-bounces@xxxxxxxxxxx] On Behalf Of Nicholas O'Leary
Sent: October-10-14 6:03 AM
To: General development discussions for paho project
Subject: Re: [paho-dev] IBM .Net MQTT Client - to contribute or not?
I would add, we've seen first hand on both this and the MQTT mailing list, the issues over people finding 'dead' code and trying to get it working...
On 10 October 2014 02:57, Dominik Obermaier <dominik.obermaier@xxxxxxxxx> wrote:
I believe this would only confuse people if there are two .NET code bases available. I also don't think an incubator sub-project is an ideal place to park "dead" (= no one wants to maintain it) code if there is another actively maintained code base in the same project.
However, having a second .NET code could be a great opportunity for anyone learning how to write a MQTT library (and see how different codebases tried different approaches for achieving the same thing).
I'm not sure if this is doable by your legal department, but it would certainly be interesting to see one of the code bases in Paho and the other library could be uploaded to e.g. Github or Bitbucket so other people can also learn from it.
Ian Craggs wrote:
I have in my possession the code for a .Net MQTT client written by several people in IBM. I was going to contribute it to Paho before
Benjamin asked Paolo about M2Mqtt.
I much prefer to have M2Mqtt in Paho, because Paolo is an enthusiastic and committed maintainer of M2Mqtt, and there is no one like that for the IBM code.
So this is a simple question: should I put the IBM code into Paho simply as a collection which can be learned or stolen from, or
not contribute it at all, as that could lead to confusion? (If contributed, it could go into the permanent incubator sub-project which I've asked to be set up).
I asked Paolo, and he left it up to me. :-) Any other opinions please?