[zeromq-dev] what is the purpose of the Go binding for czmq?

Peter Kleiweg pkleiweg at xs4all.nl
Sat Jan 17 23:47:15 CET 2015

Brian Knox schreef op de 17e dag van de louwmaand van het jaar 2015:

> Peter - really, it was more about wanting to see how czmq worked as the
> basis for a language binding (since I contributed some code to czmq), than
> anything particularly lacking in zmq4.

Ah, OK.

I'm a bit insecure. Unsure of what to do with zmq4. I started 
implementing zmq3 after reading about ZeroMQ and the current Go 
binding (which was mainly for ZeroMQ 2, and hard to adjust for 
version 3), and I thought I might have much use for a technology 
like ZeroMQ myself.

So I did zmq3, then zmq2. Then came version 4, and I did zmq4. I 
maintain all three versions. And when I add something to one 
version, I add it to the other versions as well, if possible.

And I implemented all the examples from the tutorial. 

The thing is, I never actually got to using it myself. I have 
only one application running that has a PUB socket, and two apps 
with a SUB socket to the first app. Possibly the simplest of 
uses of ZeroMQ.

Now, I have forgotten most of what I learned by doing the 
tutorial. Still, I want to keep zmq2/3/4 up-to-date. But I don't 
actually know about using these techniques in real world, 
complex situations. What are recurrent patterns? Are there 
patterns that could be useful enough and general enough to 
include in the package?

After some discussion, I opened a wiki page for this stuff, 
demonstrations for how to do real world things with zmq4:


I was hoping for contributions by people using zmq4, but until 
now, nothing.

Peter Kleiweg

More information about the zeromq-dev mailing list