Does zetta have a future?

[Scott Erholm]

The future of zetta seems to be jeopardy.  It seemed like such a great project with endless possibilities and it fit our needs very well.  However, there hasn't been much activity since the apigee acquisition, and not much traction even with the broader open-source community.  I'm wondering if it's time to move on.

I don't expect Google employees to divulge specific company planning, but can anyone offer any broad long-term forecasts?  

[Kevin Swiber]

Hey Scott,

I'm one of the original Zetta developers.  I was working with Apigee on a commercial product built on top of Zetta, named Apigee Link.  I am no longer affiliated with Apigee/Google, so I can't speak to their intentions.

Independently, after my departure from Apigee, I was briefly working on moving the core of Zetta into Rust so it can be used across programming language barriers with a tie-in back to Node using Neon.  (Basically, you can think of being able to build drivers in whatever language makes sense to your organization.)  I was also working on a solution that used lightweight container technology to run apps on single-board computers.  However, that work was barely started when I accepted a position at a new company.  (Gotta pay the bills, y'know?)

My new job is fairly demanding (early stage startup), but I'm happy to participate in discussions about moving the project forward.  Zetta has always been open source.  Prior to my departure, I was assured it will _always_ be open source.  I still have contributor access for Zetta on GitHub and can merge PRs.  Based on the low amount of recent activity, I believe it's safest to move forward assuming most feature enhancements will have to come from our Zetta community.  I know Adam Magaluk has been taking care of bug fixes here and there, but I can't say whether or not this has been an act of love carried out on his own time.

Feel free to start any feature enhancement discussion here or on GitHub.  I wouldn't bet on corporate-funded developers contributing in the near-term.  Maybe one day in the future.  (Again, I don't speak for Google.  Just a gut feeling based on the recent lull in activity.)

Also, if you're interested in my wacky idea about moving to Rust, I'm happy to talk more about that.  If I'm ever able to revive it, I'll post something in the community.

Cheers,

Kevin Swiber

On Mon, Nov 27, 2017 at 12:03 PM Scott Erholm <sc...@agilatech.com> wrote:

The future of zetta seems to be jeopardy.  It seemed like such a great project with endless possibilities and it fit our needs very well.  However, there hasn't been much activity since the apigee acquisition, and not much traction even with the broader open-source community.  I'm wondering if it's time to move on.

I don't expect Google employees to divulge specific company planning, but can anyone offer any broad long-term forecasts?  

--
You received this message because you are subscribed to the Google Groups "zetta-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to zetta-discus...@googlegroups.com.
To post to this group, send email to zetta-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/zetta-discuss/be96ac02-459e-42ed-b36d-14998985b335%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Scott Erholm]

Thanks for the info Kevin. I'm familiar with some of your substantial contributions to the zetta platform. Your side projects sound interesting as well.

I didn't mean to be too pushy or demanding.  It just has such great potential, but has languished a bit. I'm certainly not one to throw glass stones since I have not contributed, and I also can't demand that Google or any other company pay their engineers to work on open source projects.  It appears that Adam may be the only one left performing any updates.  I do see a guy from France is also interested and talented enough to contribute.

There has been some discussion on github and elsewhere about modernizing the codebase and updating the dependencies.  Overall though, I was just curious if zetta will be left to the community and will eventually fade away or if there is enough interest and need to keep it up-to-date.   I've been trying to add some security layers and have run into a few problems which are tied up in some of the older dependencies. If I can get my code to a high enough standard and working with an updated codebase I would gladly release it publicly.

[Adam Magaluk]

Hi Scott,

In my personal opinion not anything official from Google I think it could have a future if it solves a real problem which I think it does. As far as maintaining the software I don't expect much in the form of a corporate sponsorship anytime soon so it's up to the community at this point. I am actively working to get Zetta working with the latest versions of Node and upgrade our spdy/ws dependencies please follow along here: https://github.com/zettajs/zetta/pull/360 once that is done i have a few things I want to experiment with regarding MQTT and Kubernetes playing into Zetta.

Besides just cleaning the cruft off of Zetta I think the main thing that may help get others get interested is real world use cases and examples/extensions. Scott as someone who is actually using it for real world things it would be immensely helpful if you could share anything you've done. Not sure exactly what that may entail, lessons learned, useful patterns, and the auth extension you were building etc...

Hope that helps and thank you for your interest.

Adam

--

You received this message because you are subscribed to the Google Groups "zetta-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to zetta-discus...@googlegroups.com.
To post to this group, send email to zetta-...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/zetta-discuss/dccc051f-ee7c-4a30-843e-b9765ed9385a%40googlegroups.com.

[Scott Erholm]

Adam, thanks a lot for the work you've done and are doing.  I did see that you had been working to achieve compatibility with the latest spdy.  I do think getting all the dependencies up to date will help improve the attractiveness for new development.

Fair enough to ask for real-world examples and experiences--we can share some things.  I'm not really sure if this group is the best place to share such things?  Yes, I do have a couple auth extensions for the API and peers, but one requires modification of a couple core zetta files.  I'll create an issue and see if that is something you'd accept.  It is a bit hypocritical or selfish for me to ask for improvements without contributing to the cause.  I see what I can do.

Thanks to all who have contributed to the development of Zetta.  I certainly do hope it remains a valuable IoT asset.

Scott