Could I ask, why does this ever need to be part of the project? it
seems perfect for a separate Github repo, like a 'contrib' repo. I
would have said the same about Spark stuff. I would also vote against
stuff like this in the project if I were voting, and the Spark bits,
just on the grounds that it sounds best as a downstream extension, if
indeed the plan is not to actually rebase the entire project on a new
tech.

On Fri, Apr 4, 2014 at 10:13 PM, Ted Dunning <[email protected]> wrote:
> To add to Sri's comments:
>
> A prototype implementation of the Mahout Matrix and Vector interfaces with
> h2o back-end is in progress outside of Mahout with the intention of using
> it as a test-bed for developing implementations of several Mahout
> algorithms such as SSVD.  This code is intended for contribution if the
> objections of one committer are over-come by the concrete results of the
> prototype.

Reply via email to