Funding New Features -- Need some new idea I think

I love Ardupilot, of course, and stump up a small amount each month.

Given that the software development that I do is not useful for embedded development, I cannot really code much for the project. That said, how do I/we get movement on adding some features that are lagging?

For example, serial camera control with the Runcam protocol and Gopro control.

Can we crowdfund such features? It would be up to the project leaders to organize what could be offered and come up with how much it would cost? I do think a fair number of people would probably be willing to contribute if a mechanism was created. This could also apply to maintaining such features if they needed an overhaul.

In general it would be nice to come up with new models for how Ardupilot can be supported – short of being a commercial operator that can pay for features to be added. It would also be another way to interact with hobbyist users and understand what they want and value.

Anyone else think this is a good idea?

5 Likes

Hello,

Disclamer : that is not an official response, but my as team member.

This subject was largely debated inside the dev team during the last two years (at least). I think that we all want what your are speaking about but the main source of issues are : lack of management, lack of long term budget, lack of full time developper.
It a good chicken-egg issue : without stable funding, we won’t full time paid developper (yep I work for free here ;-P), without full time developper/manager we won’t have stable funding…

We already support funding of new feature, mostly with the partner program and also bug tracking. But generally we lack works force either for coding or review …

For new funding coming from the communauty, we have a funding commitee : http://ardupilot.org/dev/docs/how-the-team-works-development-fund.html . Maybe we should advertise more how to make proposal as those proposal are generally mostly fund by the project and sometimes with help from partner.

Another issue with funding feature is how to suppervised work ? We need somebody to code and somebody to watch over. But what append if it goes wrong ? Ardupilot has no legal right nor the money to attack a contractor (given that we made contractor contract for a feature). You see what I mean ?
I would love to work 100% for ardupilot, but for now, it isn’t possible.

I definitely can see the issue here and it really would require resources to manage crowd-funded developers. Mind you I really would not envision this as being terribly commercial. I would think it really a case of being used by the funding committee to spend on existing developers rather than just thrown out to devs for hire – that would be a non-starter.

The key being that these resources come from hobbyists who have pooled their resources. Serial camera control is a good example but there are others like it. It is basically asking users to pony up a little $$. I bet there must be a 100 people that would pay $20 each even to get Runcam control.

What about something like bountysource.com? Looks like there are some big teams/projects on there like IBM, and Mozilla.

I think a method to give a general donation should be the first thing to implement.
So why is there no simple “Donate” button on the home page?

http://ardupilot.org/copter/docs/common-donation.html#common-donation

Noted that we should make it more visible

I doubt anybody would find that. :slight_smile: In fact each finder should receive a prize!
Donation sent via paypal.

1 Like

Speaking as a monthly recurring donor of a few years, can you not get a bit more visible with donations. At the top of every page please just ASK people to consider setting up $5 or $1 a month to keep the lights on and features funded. As has been pointed out, this is absurdly well hidden.

And why not create a public badge for those who do so? This all plays into the very simple psychology of fundraising.

And who here cannot afford $5 a month for something many spend too much time messing with anyway – speaking purely as a hobbyist.

3 Likes

G’day Marc,
Good suggestion. I’ve raised an issue on the wiki referencing this thread, and will discuss with the team.
Thanks for your support!

2 Likes

Is there anyways you can add like the supporter membership or something like that. RC Groups and others have that.

I think this is something those of us who talk (mostly nonsense) about the industry can do more with as well as I’ll do my bit for sure. We need to promote this and Next week when I get a chance I’ll go and add the donation link into any video I have on Ardu, I’ll also make sure from now on I specifically mention about donation as well. Again we may need a poke on this but for sure more can be done.

I have always said that there needs to be a more visible donation or charging structure. Perhaps something again like a members only chat for monthly donators, more prominent request for cash on updates, while this is OS we all benefit from progress and that costs coin.

2 Likes

Did anything ever come from this? I was recently able to get my 2000mm quad up and flying mainly due to the instructions on THIS page and wanted to donate either to the person who created that guide, or the project in general. But, then as I was browsing around the site for who/where to donate to, I ran across the proposals forum and saw that it looks like close to $10k was being spent on flights and hotels for people to attend the upcoming conference. I haven’t been around long enough to know who the major developers are that contribute to the project often, and since all of those funds were approved by the committee, I’m sure they’re all well deserving of those funds. My question is, will 10k worth of ardupilot development come out of having these people attend?

Not sure if anyone is familiar with blender.org, but they are an opensource 3D animation program that started small and if I remember right, they would do fundraising campaigns for their conferences, specifying milestones for certain numbers of developers to attend and what types of things would be accomplished if those milestones were met. Does something like that exist for the upcoming conference? I searched around and couldn’t find anything, but it’s definitely possible that I missed it. I just find it really hard to pay for someone to fly to Australia to meet other developers and put on a Powerpoint presentation (not that it doesn’t have value), but that’s just my opinion.

I hope this post doesn’t come across as disrespectful, because I think the Ardupilot devs have done, and are doing some great things, and I really appreciate everyone who contributes to the project. But, when it comes to money, and even more specifically my money, I would love to be able to donate to specific things I would like to fund. Even if a percentage of that donation is designated for administrative fees like server costs, etc. If there are any developers out there who decide to create their own Patreon page and ask for monthly donations to fund things they want to expand on, that would be great too.

It looks like Ardupilot is getting bigger by the day, and @khancyr posted very good points, but I really think SOMETHING should be done to manage the project, but that’s just one hobbyists perspective. I think PX4 is now managed by the linux foundation, has that been considered to help at all? Just some thoughts.

James

I know a few people going. I reckon they have donated 100k plus of development work, maybe a lot more, just based on my billing rates (which is not embedded open source development). So yes, I do think it is worth it and very well deserved.

Ardupilot is pretty informal. You get into very difficult territory when you start questioning the value of some things given market value some people give for free. I know you did not mean to be disrespectful but I am sure others will respond in a similar vein.

That said innovative funding discussions are good.

If that’s the case, then it is very well deserved for sure! Just to reiterate, I have zero knowledge on who put in what time and how/when they should be compensated, as that’s the committees job, I’m just pointing out what I see and what someone else will see who is contemplating use/donations/funding, etc. What I’m hoping to get across is that people might be more inclined to fund things, or donate to the project, if they have a better understanding of where or what the funding is doing. And that info might be readily available somewhere already, but I don’t know where it is. Also, @james_pattison, I couldn’t find that donation link without your post. Once I did I was happy to make a donation. Thanks!

1 Like

I do think funding model debates are good. I was just offering my limited perspective.

And the way things are going in the US and elsewhere it is likely going to have to be up to professional users to sort out funding out as I am not sure many hobbyists will be around in 5 years time.

1 Like

The issue is that nearly all of the top devs here are simply diverting what could be paid hours to the project. I mean how much do you think that some of these people could charge for consulting on proprietary work.

G’day James,

Feedback is always good - no criticism taken in any of it.
I’ll try to address each of your points, but please feel free to request more information or clarification.
Increase visibility of donate button: there is a “Donate” button on the wiki homepage/sidebar now (https://ardupilot.org/ardupilot/), but I’ve left the issue on the wiki open for now. The main website is built on Concrete5, which I have an irrational dislike for and am lobbying to change to something that is more maintainable (I want to roll both the website and the wiki onto a common framework, but it’s a pretty significant undertaking and another problem entirely).
Tuning page: credit goes to @Leonardthall for being the wizard behind that. I suspect the positive feedback will be reward enough for him, but feel free to drop him a message.
Conference funding. I think you’re right to question the amount we are spending on travel to the conference this year: I can understand why it might appear disproportionate, and it is more than we have done in the past. What we have found is that we get a substantial productivity surge from the team in the lead up to, and at, the Conference - so the entire community/user base benefits. If it was costed / billed out at commercial rates it would be many multiples of the expended travel costs (as Marc suggests). Apart from that, getting as much of the team together as we can lays a great foundation for the following year of development. For any team, the ability to build relationships is critical, and whilst we live and work in a digitally connected world, the ability to actually meet and understand someone is priceless.
We didn’t assign formal targets/milestones to the devs to qualify for funding, although those who have been subsidised were all required to submit a proposal - some are leading panel discussions, others presenting, demonstrating, etc. There has also been specific sponsorship for the Conference, and we have assigned some GSoC funding to it - it isn’t all coming out of general donations.
Regarding funding of particular issues/features, we investigated a number of options (including Patreon), and haven’t reached a concensus. It seems each of the options have different mechanisms to hold donated funds, and approve release of them, and we haven’t found one which the team is comfortable with that also demonstrably meets the requirements of a 501© NFP. It will be a discussion topic at the Conference though, as the need/demand from our Community is clear. There is nothing preventing individual devs creating a Patreon or similar campaign, but it would need to be clear that it was separate from ArduPilot.
Re legal domiciling of the Project, a little history: DroneCode within the Linux Foundation was created by ArduPilot. I’ll let you google the rest of that particular story, but ArduPilot is now part of SPI: http://spi-inc.org/projects/. SPI provide the governance and financial framework for ArduPilot, with the majority of our funding coming from the Partners Program.
With around 65 members, the ArduPilot Partners Program could well be the worlds largest commercial community for open source autopilots.
@Marc_Dornan makes a valid point re the relative informality of the internal workings, currently. We meet the governance requirements of SPI, but there is hesitance about recreating a corporate entity like DroneCode, due to what happended last time, and the cost overhead (it requires management staff to operate that type of model). At this stage we don’t think it’s the best use of funds, but that position may change over time.

Wow @james_pattison, thanks for addressing all of those points!

As for the donate button, I see it now, and i’m sure it has probably been there for a bit. It must have been hiding in plain sight the whole time! :grinning:

I sent Leonard a personal message with thanks.

Yes, looking at the proposals forum, which I’m assuming all of the proposals are placed on, I see the funding for the 2019 conference last year was approved, but no travel. Now, for 2020 I see 5 requests for travel expenses. It just LOOKS odd, even though there’s justification behind it and the community benefits from it.

I understand the benefits of meeting people in person as I can directly relate to it from my work experience. I’m part of a team that helps maintain/manage the software the local sheriff’s department uses to write reports and keep track of crime statistics. The vendor who provides the software is based in Florida, and we’re located in California. Multiple members of the vendors software development team are spread out across the U.S. so regular virtual meetings are a necessity. There is a conference the vendor puts on once a year, but they charge all of their customers to attend and put on different presentations and forums over the course of 4 days. During that time, we can connect with the people we have virtual meetings with and put a face to the name which like you said, is priceless.

I look forward to hearing about any developments on the funding of specific issues. I’m sure meeting the requirements of a NFP is difficult, but I think it could definitely benefit the community and project. If someone doesn’t beat me to it, I’m hoping to put up a bounty on getting Mateks Optical Flow and Lidar Sensor working by putting my money where my mouth is.

I’ll definitely try to find more info on Ardupilot and Dronecode. It seems like there’s quite a bit of history (not all of it good) between some of these projects. I frequent the PX4 forums too and like to keep up on who’s doing what.

Informality is great, to a point, in my opinion. It’s definitely a delicate balance between keeping the informal nature of the project, and trying to make decisions that typically rely on a formal business model. Kudos to you for putting in the work on it.

Thanks again for taking the time to write such a detailed breakdown for me, I really appreciate it.

James

1 Like