Riak Adoption - What can we do better?
demeshchuk at gmail.com
Thu Apr 19 14:00:52 EDT 2012
The problem of any book is Riak rapid evolution.
Sometimes very cool and important features are added within a minor
release. That's generally okay, but I wouldn't read even a half-year
old book, it will contain very outdated information.
I'd add some cross-referencing between code (or features in general)
and the docs related to them (especially wiki pages of course).
Currently, wiki cannot keep up with the code, and probably releasing
docs updates along with pull requests could help that.
On Thu, Apr 19, 2012 at 9:53 PM, Evan Kinney <Evan.Kinney at sas.com> wrote:
> Upvote for a book like this.
> I've spent a good deal of my time when evaluating Riak searching the Googles (and perusing the Basho git repo - thanks for that, by the way) for information like this. This list is a great resource (and so is the wiki, for that matter) but having a set of defined and sanctioned "best practices," as it were, in such an easily-avaialble and accessible form would be killer. I'm sure Packt (or maybe even O'Reilly!) would love to pick up something like this.
> Heck, I'd even offer to help write it; my love affair for Riak is already at a ridiculous level as it is.
> A good example of this is the MySQL Cluster project. I'm intimately familiar with its internals, as I've written many patches for the engine... but it's hard to explain to other people (especially management). Alex Davies published a book through Packt that explains everything and even gives advanced performance tips and scenarios. I keep this on my bookshelf in my office and hand it to anyone who comes to me asking about NDB stuff.
> Evan Kinney
> Systems Architect, Research and Development
> SAS Institute Inc.
> P.S.: This is what part of the alphabet would look like
> if "Q" and "R" were eliminated.
> On Apr 19, 2012, at 1:37 PM, Zabrane Mickael wrote:
>> +1 for a solid book on advanced Riak usage.
>> On Apr 19, 2012, at 7:31 PM, Mark Phillips wrote:
>>> At this point Riak has been deployed thousands of times. This is excellent, but as you can imagine we here at Basho feel this is just the tip of the iceberg. We want everyone to be running Riak. The software obviously needs to solve enough problems for people to want to deploy it. However, we need to get the software into developers' hands before this happens. So the question is:
>>> What can we be doing better?
>>> I'm not talking so much about features in Riak itself. I'm more interested in the other resources, tools, docs, processes, etc., that you think we need to put in place or refine to take the Riak community from 1000s of users to 100,000s of users. By virtue of being on this list, you've all spent some time using or investigating Riak. As such, you're opinions and experiences would be very valuable.
>>> Any comments, ideas, or thoughts you had would be great. On or off-list will work.
>>> Thanks for your time. Long live the vnode!
>>> riak-users mailing list
>>> riak-users at lists.basho.com
>> riak-users mailing list
>> riak-users at lists.basho.com
> riak-users mailing list
> riak-users at lists.basho.com
More information about the riak-users