[Road map][R SDK]

[AndrasNagy <st...@gmail.com>]

Hello,

After a bit of googling i decided to ask the following:

Is there an "official" roadmap for 2017?
If yes, please send me a link. (i found one for 2016, but not so sure if
that was retrospective)

Is there demand for a R SDK?
Is there already in construction something?

If yes I would like to ask for a git repo

Thanks in advance!

NagyAndras

Re: [Road map][R SDK]

[Sourabh Bajaj <so...@google.com.INVALID>]

Hi,

1. I don't think there is a R SDK in development within the main repository
not sure if someone is building it in a fork or not.

2. There might be some demand for it in the data science community but
currently people have been using rpy2 and doing the processing via the
Python SDK based on the few emails we've seen on the mailing list.

3. The biggest blocker in building a R SDK might be good support for GRPC
as the Fn API uses that to communicate between the runner and the worker.

Others might have more thoughts on this.

Best,
Sourabh

On Tue, May 23, 2017 at 8:19 AM AndrasNagy <st...@gmail.com> wrote:

> Hello,
>
> After a bit of googling i decided to ask the following:
>
> Is there an "official" roadmap for 2017?
> If yes, please send me a link. (i found one for 2016, but not so sure if
> that was retrospective)
>
> Is there demand for a R SDK?
> Is there already in construction something?
>
> If yes I would like to ask for a git repo
>
> Thanks in advance!
>
> NagyAndras
>

Re: [Road map][R SDK]

[Davor Bonaci <da...@apache.org>]

>
> Is there an "official" roadmap for 2017?
>

We don't have an official time-based roadmap. So far, the project went
through the "standard" cycle: getting accepted into incubation, code
donation, first release, graduation, and the first stable release last week.

Going forward, I'll be looking into discussing and formalizing some of the
project's goals, and publishing them on the website. This will help set
expectations with current and prospective users, but it won't, however, go
as far as the "official roadmap" -- this is not something we can provide, I
think.