Greetings! I wanted to chime in here as well.  I am an SDAP PPMC member and 
user.  I was the original developer of several of the SDAP analytics algorithms 
built on Apache Spark.  I am still an active member, but these days it is more 
as an SDAP user and evangelist, bringing SDAP to various projects at JPL/NASA.  
I appreciate Trevor’s focused enumerated list in his email below, as a guide to 
graduation.  It seems all of those steps are easily achievable as long as we 
don’t let the pursuit of perfection be an obstacle to getting a release that is 
“good enough”.  I personally have seen no slowdown in progress or in the team’s 
engagement with SDAP.  Quite the contrary as the local SDAP developers and 
stakeholders have been meeting regularly to coordinate and share information 
across projects that are using SDAP, and we’ve seen new developers make 
significant contributions.

My latest activity, ongoing right now, is an experiment in using Planetary data 
in SDAP (first time for that as far as I know) and maybe the initial 
experiences with that can be a part of the next SDAP report.  The long term 
vision for that effort is to include SDAP among a menu of infrastructure 
elements and applications that NASA missions in Earth, Planetary, or Space 
Science domains can select from in automated deployment of their science data 
systems to the AWS (or other) cloud.  For another project I am managing two 
distributed deployments of SDAP across cloud platforms In the USA and in 
Europe, to support oceanographic applications.  For that one, we are making 
heavy use of SDAP’s automated forward ingest capability to continuously (on a 
daily basis) update the dataset time series for active satellite-based missions.

I expect to return as a more active developer/committer in the next couple 
months with a plan to add an entirely new algorithm/endpoint to SDAP related to 
the calculation of spatial gradients, required for the oceanographic 
application I mentioned above.  I think efforts like these will be important to 
build the community of SDAP users and stakeholders beyond the projects for 
which SDAP has already been used in operations for years.

Welcome Julian, and thanks to Trevor and Justin for your contributions!

Joe


From: Nga Chung <[email protected]>
Date: Friday, October 14, 2022 at 12:30 PM
To: [email protected] <[email protected]>
Subject: [EXTERNAL] Re: [DISCUSS] Steps to graduation
Hi all,

I am an active member as well and am currently invovled with several NASA
funded projects that leverage and contribute to SDAP.

As both Trevor and Frank had already mentioned, getting an official release
out has been the biggest hurdle.

We had an attempt last year to set up GitHub Actions to automate the build
process but ran into a hurdle with pushing images to Docker Hub, so that
was left unfinished.

Besides a formal release, I've heard from Stepheny that we also need to
identify a PMC chair, though I'm not clear what the process for that is.

We could really use a clear checklist of items we must complete in order to
graduate.

Thanks,
Nga

On Fri, Oct 14, 2022 at 10:57 AM Frank Greguska <[email protected]> wrote:

> Hi Julian,
>
> I am an active member but mostly just admin stuff at this point; not able
> to provide much development support. I agree with Trevor's assessment that
> the hardest part has been trying to get a release together for review by
> Apache. There's a lot of history there I won't get into but I know there is
> a current effort underway to try again.
>
> - Frank
>
> On Fri, Oct 14, 2022 at 9:52 AM Trevor Grant <[email protected]>
> wrote:
>
> > Based on what I've seen from the project for the last couple of years,
> > here's my short synopsis of the status wrt graduating.
> >
> > To graduate, a project must cut an official release. This is the main
> point
> > holding SDAP back from graduation. Apache as an organization was designed
> > for creating and releasing Java projects, SDAP is a Python project.
> (There
> > are Python projects elsewhere in the ASF, but I will stand by my original
> > statement, that Java is first class, everything else should adapt to
> Java).
> > SDAP as a community has focused the majority of its cycles to supporting
> > existing deployments of their software as opposed to refactoring to
> conform
> > with ASF things. I don't think there exists undue stress from the ASF to
> > conform- just a lack of documentation for non-Java paths.
> >
> > My recommendation in terms of cutting a release (and thus removing the
> > largest hurdle to graduation) would be to
> > 1. call a code freeze
> > 2. tag / create a zip file of the current code base.
> > 3. have various PPMC member test the code (what ever that means in
> relation
> > to SDAP)
> > 4. vote
> > 5.manually push the zip file to whatever repository that is needed.
> >
> > I know there are lots of maven plugins that automate #5, that will be
> > effectively unusable since this isn't a Java project. Just get ahold of
> > someone at Infra and ask them what you need to do to manually release a
> > source tar ball. Then document it, maybe put it on the website- and make
> a
> > plan to do it again once a year to keep the board off your back. I know
> > your real life users don't care about official releases anyway, so it's
> > just a formality that has to happen every so often.
> >
> > My .02
> >
> > tg
> >
> >
> > On Thu, Oct 13, 2022 at 1:17 PM Justin Mclean <[email protected]>
> wrote:
> >
> > > Hi Julian,
> > >
> > > Thanks for reaching out to the SDAP project, I also think it time this
> > > project needs to consider graduating or the other option is retirement
> > from
> > > the ASF.
> > >
> > > Kind Regards,
> > > Justin
> > >
> >
>

Reply via email to