Akka core developers, congratulations for the 2.3.0 release. My team are 
using it to develop our product right now. Thank you.

One thing we are discussing (and still haven't come to an conclusion) is 
how to make sure all persistent events (commands if we use 
command-sourcing) will be backward compatible.  We can certainly use 
protobuf WDL to defined all events/commands (and I did try), but it turns 
out working with Java POJO classes with Akka is so unnature as we are all 
spoiled by pattern matching. I also tried to use Twitter's scrooge to 
compile thrift files into scala code (so all thrift structs are case 
classes), but I didn't manage to come up with a general serializer (by 
'general' I mean there is no need to serialized each event/command type 
case-by-case as the protobuf serializer). I'm going to try Kryo serializer 
soon. I wish I can have some luck.

Processor's in-memory state snapshots are also need to be backward 
compatible, otherwise each upgrade of events/commands will reply all 
history events/commands, given events/commands persistence are backward 
copyable. The idea on my mind is to 1) convert the scala-collection based 
memory snapshot into a huge thrift/protobuf strutter, then 2) serialize the 
thrift/protobuf structure to byte arrays. To make the conversion async (not 
to block processor event loop), I may have to write a customer serializer.

It will be great if someone from the Akka dev team can comment and share 
their thoughts. 


-- 
>>>>>>>>>>      Read the docs: http://akka.io/docs/
>>>>>>>>>>      Check the FAQ: 
>>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html
>>>>>>>>>>      Search the archives: https://groups.google.com/group/akka-user
--- 
You received this message because you are subscribed to the Google Groups "Akka 
User List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.

Reply via email to