BRAVO!
I would try answer to the invitation concerning Nethforge for the module I did recently or in the past…One time more I would love to write in a better English…At the end don’t forget I’m not a bad guy, even if the dark side is attracting me.
The problem is not really to attract new developers, they will come with the light and the noise generated by the project, it is to create a sustainable development way for them…why they will stay here to create and maintain rpms.
If I look behind my little experience of contribs.org, we saw people who have done a lot of code, who have pushed the project really far, and now…you can find some of them answering an email, others don’t want to listen any more about contribs, and some are RIP 
It is a really private question why you stop to work in a free software project, and the answer may change with each guy…I totally respect their point of view, but I have my little idea.
They do not reach a sustainable way for the software development.
One common point for those who worked the longer time/are still active in the SME Server project, it is simply for the ‘money’
- They work for a company and the development was needed for their customers and after released in GPL.
- They funded their own company relative to sme server, providing services and software (dungog.net)
Concerning my involvement in the SME Server project, I provide rpms, code fixes, documentations, ideas, forum supports, freely besides to my day work and my family. All of this is used by people who play with SME at home, at work, earning money with something I gave. Everybody is happy and satisfied.
I’m happy but not satisfied.
I would like to share a bit about nethserver-crontabmanager, since it is asked to push it to nethforge. This software needs around 30/40 hours of development and tests for a first shoot, and probably many more for NFR and enhancements. I don’t know what can be the value in money, it depends where you live on the earth, but the standard measure of one week of work is the same everywhere 
I can push this rpm, and others to nethforge, I don’t mind, as proof of my willingness, the sources are on github, but I’m more concerned to find a way that will let to the developers, the motivation to find time and get creativity in the future. If I speak about money, and I really don’t like this word, it is because my ‘hobby’ needs it, the server at home that I use to develop and test, the server online that I have to pay each month, the laptop…etc…etc
How find a sustainable way, why continue to develop freely
- the rpm factory is a showroom to be hired by nethesis
- well known developer can be paid to develop a specific module for a company
- a company can sponsor the cost involved to your creation (hardware, server, beers)
- the developer may ask about donations to users, like remi collet does
- the developer can create his own repository and ask money for updates, it was the dungog model
- a market can be done, integrated to nethforge, this is the ClearOS model.
- … I’m listening you
I hope that you have understood that I’m not speaking only for myself, I’m just trying to speak loudly to find the code engine which doesn’t stop.
9 Likes
I couldn’t agree more! 
I propose a simple donation system to support at least external contributors and hosting for mirrors (and maybe all community stuff like wiki, official site, etc).
I know others have better ideas than mine, so stay tuned! 
1 Like
What’s about how to redistribute donations equally and/or by technical requirements (services, hardware, software ecc.)?
@stephdl, I took my time to try to answer, but I still don’t have a clear idea. We need to work on a good environment on a step by step basis.
I find myself in a difficult position, because I’m a both a contributor to open source projects and a stakeholder in nethesis. I’ll try to share some ideas, to foster discussion, but I think that this kind of problems are better tackled in front of a beer.
I see the nethforge repo as a kind of “certified quality” storage, i.e. packages coming from the forge have been checked by core developers and considered of high quality.
If someone has access to the forge, it means he demonstrated ability over time.
Before entering the forge, packages could live in personal repos, we could build a central store, where it’s clear how polished a packages is, who is supporting it, etc.
Also, packages can move from the forge to the os (i.e. become “official”).
We could offer a way to give the developer feedback on forge packages, we could extend the Software center.
I’d prefer that these features would be modeled by the community, leaving to nethesis only the implementation burden, i.e. we (the community) decide how things should work and then ask for help to some core developers to implement (I hope to have some power on them). 
Talking about money (it’s not a taboo), I’d like to hear more voices. Things should not go around money, but who makes money could share a slice.
nethesis is making money and is open to the idea of sharing.
My idea is to give kind of a token to packages promoted to core (promotion will be a community decision).
nethesis business model relies on customization, it could be shared easily, maybe through something like a market where who is looking for a feature can find a developer.
I’ll be happy to hear more opinions or critics.
Meanwhile, I’d like to wish you all a Merry Christmas.
7 Likes
Ehi @stephdl sorry for my late response but I took my time to clear my head.
Given that:
- Nethesis doesn’t wish to receive nor accept donations
- Every donation should be redirected to the NethServer project
- We have to ensure maximum transparency in money transactions
- Creating a foundation to receive and manage money is not an easy thing to do and likely it doesn’t worth at the moment.
There are two practical ways we can go at the same time:
- Reward developers funding features to implement
- Set up a simple donation system
Reward developers
www.bountysource.com could be a good proposal: straightforward, transparent, integrated with the issue tracker
How does Bountysource work? Users fund bounties on open issues or feature requests they want to see addressed.
Developers create solutions which close the issue and claim the bounty on Bountysource. Backers can accept or reject the claim If accepted, Bountysource pays the bounty to the developer. Easy, doesn’t it?
In this case, community members or companies interested might act as bakers.
Check the project FAQs and some examples:
Donations
We should set up a simple donation system through which people can support the project.
That’s a simple proposal, but we can find other ways:
Then, we can redistribute such pool of money equally and/or by technical requirements: services, hardware, software, mirrors, merchandising, community initiatives. Or even simply supporting issues on bountysource itself.
We have to decide who can manage this pool, for instance, a group of chosen and trusted people.
What do you think? @Ctek @islipfd19 @robb @sitz @Ctek @Nas @dz00te
3 Likes
Hi Alessio,
I like the idea of bounty for features! This system is not new and it works well in other communities. Plus it brings value to the NS in my opinion
I have some questions 
and hope that each can chip in
- How can we make a test-phase?
- What features can qualify and how ?
- Will be a fixed amount per feature or some features will have more weight? (who/how the weight of the bounty will be calculated) ?
The amount will be based on the difficulty of the tasks.
Anybody will be free to create a bounty with any amount.
Potentially, you can claim a bounty on every features/issue
We need to open them on github because currently Redmine isn’t supported by bountysource, so we can’t do it soon I think but we make it happen.
1 Like
These are the links
there are some GitHub issue tests I’ll remove in the future…
Bump! I am keen to hear what others think, this topic is extremely important for me
Hi guys,
the proposal of Alessio sounds good for me, I don’t know bountysource.com but I will give it a look.
I’m agree with the idea of creating a group of trusted person that can decide how to manage donations and or resocurces, all transactions could be trasparently shared with the community.
Since I am more familiar with the ubuntu community than the CentOS community, I propose to more or less copy the method used by the ubuntu community. There are several teams, as are with NS. Each team has a certain amount of volunteers that are chosen by the community. There is a x-year mandate for each election and after that, the person can be a candidate again for re-election. The election is like a campain where the candidate must explain why he or she is the best person for the job. Then the community can decide.
For important jobs like managing the community money, there should be a abuse proof set of rules, so no individual member of the treasure comity can spend or take money. All money spent must be transparently documented.
Also legal issues must be met. Can someone be held personally responsible for income or spending community money? What about taxes to be paid? Or if people are getting money from the government because they are (temporarily) without a job, will being part of the treasurey team have impact on their personal income?
Maybe a set of (financial) skills must be met in order to apply for such a job in the community? Maybe this sounds overkill for now, but as soon the community gets bigger, like Fedora or Ubuntu communities, this can be a big responsibility.
To get started, I propose the community team (@alefattorini, @Jim, @vhinzsanchez and me) take this responsibility until all things that are needed are arranged and an election can be held. The main task for the community team in this will be to explore all the legal issues and the set of rules that are needed for a healthy and transparent treasury team.
Please comment on this. Any other ideas?
It looks to me a bit overkill for us, at least at the time. Only to me?
Regarding this Salt looks pretty easy to administrate, do you have any other suggestion?
I can do the job but I’d like to surround myself by trusted and selected/elected people.
@stephdl with boutysource and donations we could address thess goals suggested by you:
- well known developer can be paid to develop a specific module for a company
- a company can sponsor the cost involved to your creation (hardware, server, beers)
- the developer may ask about donations to users, like remi collet does
What do you think to start to raising funds for it creating a bounty? I’m interested in this. When you have raised enough money you can start to develop it, having your expenses covered.
I’m very curious to hear your opinion.
Certainly that can be great.
Bounty program for features it is amazing, but it has a lot of the pros and cons.
Frankly speaking, this program will stimulate our internal Dev team and it is one of the major pros, but as for involving newcomer developers it is very controversially.
As for paid features, we have already discussed this point on the forum or in Hangout with @Ctek and others, so bounty for this is appreciated.
As for involving new developers, first of all, we should spread our GitHUB page all over the world and developers should use our product by themselves, they should feel our product and only after that they will contribute.
Somth like that https://github.com/fail2ban/fail2ban/pull/1288, or like our @Ctek, he feels that it is lack of backup module and he made it
First of all, we should improve our Basic functionality like:
Mail (DKIM, transport maps, relay maps and so on), Fax - move to Avantfax or translate existed Fax2web, DNS/DHCP improvements, HTTPD service management.
Next step it is paid features (bounty), we should not put the priority on paid features than on basic functionality because we will drown.
Let me summarize:
We need a manager who will check dev team possibilities for paid features development (@alefattorini ) , all paid features should be completely discussed.
We need a donate button in the community page, this donation will spread for community needs like prizes, t-shirts, and promotions.
This is my point of wiev.
2 Likes
Totally agree.
Else, will be an avalanche of modules/features which may have nothing to do with the basic functionality of NS (UTM, Mail, DC, AD - when Samba4 will be available) and those modules will must be paid (rewarded).
In my opinion, even extensions of basic core features can be part of a bounty.
From my own experience, it can be (really) hard to work on core feature…You might lost a lot of valuable developer time…maybe bounties for specific modules could be more appropriate if you want to attract new developers.
2 Likes
Bounties can be for core and specific modules.
Adding a feature to core code can be a bounty that some developers will want to get and others no.
The developers can chose from the bounties available and get them if they want.
We can see how things evolve, after some time. I think is too early to tell what will work and what not.
3 Likes