Are there plans for a commercial fork?

[Jamie B.]

I'm in the process of researching Salt to replace a home-grown
configuration management system built onto of Fabric. Thus far, it
seems like an interesting project with a lot of potential. However,
I'd like to know what the road ahead looks like. Since "Salt Stack
LLC" has existed about as long as Salt itself, it would seem that it
was created with a commercial endeavor in mind. I'm fine with that; I
run a small business myself. But what I hate seeing is project
sponsors who treat contributors as an unpaid R&D staff and early
adopters as beta testers. Resources eventually get pulled from the
open source project and all of the really cool stuff is only released
in a commercial "Enterprise" version. I don't want to see my code end
up in a "loss leader" product for a commercial software company.

I really don't mean to sound accusational, but past experiences have
made me a bit cynical. I apologize if my tone is overly harsh. But
could we get some clarification on what the commercial plans are? Are
you going to go the way of Sphinx Search and make money purely on
consulting/training with an open source product, or are you planning
on going the commercial fork route?

[Matt Ryanczak]

Thanks for bringing this up Jamie. I'd love to hear about any plans to
commercialize Salt Stack and what that would mean to its open source
nature and status.

~Matt

[Thomas S Hatch]

I caught up on the emails, so I can get to the long response that this one needs :)

I started writing Salt last year in February, my original intention was to create a generic remote execution system that was fast enough to use as an API for controlling large groups of servers, primarily because I wanted to use it as the backbone to a simplified private cloud controller (That was butter - https://github.com/thatch45/butter - don't use it, future plans are in another direction). I thought the project was mostly just for fun up until October 2011 when the first company came to me and asked for help setting up Salt States because they wanted to migrate away from Puppet. If you go back and look at the impact graph in github or on the Ohloh stats that there was far less code and the community was much smaller, we had about 15 people in IRC and 17 code contributors.

Shortly after this engagement started with my first production States deployment someone in the community (I think if was SEJeff or Whiteinge, but I don't remember for sure) suggested that I go on FLOSS Weekly. I decided that it might be fun and would not hurt anything so I emailed Randal Schwartz about it. After the interview (found here: http://twit.tv/show/floss-weekly/191) salt traction picked up a great deal, and I decided to form Salt Stack, LLC and start to explore how to make money with Salt. If  you look at the developer traction that Salt gained in November and December you will see what I mean!

At the end of December my employer (Beyond Oblivion) went out of business and I was left without a job, while a stream of offers came in, I decided to try and turn Salt Stack into a viable company. Quickly I was contacted by a number of investors who wanted me to stop making Salt open source. I tried to explain to these people that this would kill Salt overnight, they did not understand, so I told them to take a hike. I spent the next few months working to get the state system to a more reliable place and filling out features that I felt would be critical for enterprise deployments.

In the mean time I had a few consulting jobs but have avoided being too pro active in seeking work, primarily because I wanted to grow the code and community to a healthier place before getting too bogged down in work. But about 2 months ago I started looking more aggressively into raising capital and securing customers. Right now we have a number of support customers and I have been absent from the development this last week because I decided to update the training for some new customers (it is very cool now, all the labs use salt-cloud and it walks the students through building a real deployment out on EC2).

So that's the story, what about the plan then you are asking? As for an enterprise version, I am planning on one, but it is not what you have suggested! The main Salt code it getting rapid releases and massive contributions, it has proven itself as a great platform for rapid development, I don't want to hinder that at all! So the plan is to create a long term supported version of Salt that would still be open source and have all the same policies about contributions as Salt. This version will pass through more aggressive testing and QA, and will only  be released every 6-9 months. The idea being, that it will be a better platform for Salt Stack to support without slowing down the innovation and excitement around the rapid release branch. This I feel also adds a great deal of credibility to companies seeking to use Salt.

Now there are a lot of monetization plans beyond this enterprise version, but none of them change how the community is structured and how Salt is developed and made available, and none of them involve making Salt code proprietary! The primary goal of my company, Salt Stack, is to make software that is fantastic, to make software that will endure, and make software that can meet the needs of the widest possible audience while still being easy to use and easy to understand. A proprietary enterprise fork would greatly damage our ability to meet these goals, Proprietary software dies, it does not grow with companies and customers' needs, and it kills innovation.

I want to apologize that I have not been as attentive over the last few weeks, the business demands have been very high, and there are only about 18 hours I can work in a day while maintaining some degree of sanity. I can't thank you all enough for your help and support of Salt, as I originally hoped, making Salt open source really has enabled us all to create a fantastic system, it is something that I am deeply convinced could and can never happen in a proprietary setting. Since the updated training course is almost ready, salt cloud 0.7.0 (it is awesome, stateful management of public cloud vms!) is nearing completion, and some of the company needs are getting wrapped up I will be back in the Salt saddle about 40 hours a week again soon!

- Thomas S Hatch

- CEO, Salt Stack

[Ben Hosmer]

This should go up on the wiki!

--
Ben Hosmer

Brevard Tech - Brevard County Florida's Tech User Group

http://www.brevardtech.org

Drupal Developer

http://www.radarearth.com

321.252.9581

[Thomas S Hatch]

Hopefully this is a satisfactory response, I think the story should go in the docs, I will put it in there

[Sean Channel]

Thanks very much, Thomas! I hadn't really thought much about the back story.
Salt is just what I was looking for. I would have to go 3 different directions
to say more.

Let me take this moment to also thank Thomas and the community for not going all
cutesy and naming _every_ component after food/spices/utensils, etc. like some
other projects I know of. Tedious! ;)

_s

On 08/10/2012 01:40 PM, Thomas S Hatch wrote:
> Hopefully this is a satisfactory response, I think the story should go in
> the docs, I will put it in there
>

[...]

[Tom Vaughan]

On Friday, August 10, 2012 12:47:14 PM UTC-7, Thomas Hatch wrote:

Now there are a lot of monetization plans beyond this enterprise version,

A salt master service?

-Tom

[Thomas S Hatch]

Maybe I should not have said anything :) but something like that is on the list of ideas, it is a big bin of ideas really :)

[Matt Ryanczak]

Thanks for the detailed explanation! I'm really looking forward to a
"release" version of Salt in the future. I imagine that will help drive
adoption which is great.

~Matt

[Thomas S Hatch]

Yep, it is good business too, things that drive adoption drive demand and establish a larger need for expertise :) 

[Kent Tenney]

On Fri, Aug 10, 2012 at 4:30 PM, Sean Channel <penta...@gmail.com> wrote:
> Thanks very much, Thomas! I hadn't really thought much about the back story.
> Salt is just what I was looking for. I would have to go 3 different directions
> to say more.
>
> Let me take this moment to also thank Thomas and the community for not going all
> cutesy

Me, I'm all about cutesy, my current fav is _saline_, as in
'Salt, the saline solution'
or
'Smooth saline with Salt'

[Sebastian Kreutzberger]

Thomas, thanks for the explanation & background story! 

I suggest maybe getting revenue through certifying Salt administrators (or however you call them) and through paid classes all over the US and EU. I don’t know where you are living, but if you are in the Valley I strongly recommend speaking to the guys at Appcelerator (http://www.appcelerator.com). They also have a very rapidly groing open source project and their business model is (besides offering analytics) to host classes (they cooperate with an IT training company) and people can become Certified Titanium Developer (http://www.appcelerator.com/training).

A certification would also help me in recruiting, especially in the mid-term, because then I could just hire administrators who are Salt certified. 

Best regards,

Sebastian

[Thomas S Hatch]

Right now we do have a training program, a three day training (I am teaching one right now).

All of the training we have done so far has been in house and it is starting to really pick up.

We are also working on a certification exam which will be available very soon in addition to the training program. So yes, this is in the works.

Right now Salt Stack does offer commercial support for Salt and training. Certification is on the horizon and we are working on lining up our first public training classes soon, the current target is later in September and/or early October.