Funding Krita

Published    8/25/2015

Even Free software needs to be funded. Apart from being very collectible, money is really useful: it can buy transportation so contributors can meet, accommodation so they can sleep, time so they can code, write documentation, create icons and other graphics, hardware to test and develop the software on.

With that in mind, KDE is running a fund raiser to fund developer sprints, Synfig is running a fund raiser to fund a full-time developer and Krita… We’re actually trying to make funded development sustainable. Blender is already doing that, of course.

Funding development is a delicate balancing act, though. When we started doing sponsorship for full-time development on Krita, there were some people concerned that paying some community members for development would disenchant others, the ones who didn’t get any of the money. Even Google Summer of Code already raised that question. And there are examples of companies hiring away all community members, killing the project in the process.

Right now, our experience shows that it hasn’t been a problem. That’s partly because we have always been very clear about why we were doing the funding: Lukas had the choice between working on Krita and doing some boring web development work, and his goal was fixing bugs and performance issues, things nobody had time for, back then. Dmitry was going to leave university and needed a job, and we definitely didn’t want to lose him for the project.

In the end, people need food, and every line of code that’s written for Krita is one line more. And those lines translate to increased development speed, which leads to a more interesting project, which leads to more contributors. It’s a virtuous circle. And there’s still so much we can do to make Krita better!

So, what are we currently doing to fund Krita development, and what are our goals, and what would be the associated budget?

Right now, we are:

  • Selling merchandise: this doesn’t work. We’ve tried dedicated webshops, selling tote bags and mugs and things, but the total sales is under a hundred euros, which makes it not worth the hassle.
  • Selling training DVD’s: Ramon Miranda’s Muses DVD is still a big success. Physical copies and downloads are priced the same. There’ll be a new DVD, called “Secrets of Krita”, by Timothée Giet this year, and this week, we’ll start selling USB sticks (credit-card shaped) with the training DVD’s and a portable version of Krita for Windows and OSX and maybe even Linux.
  • The Krita Development Fund. It comes in two flavors. For big fans of Krita, there’s the development fund for individual users. You decide how much a month you can spare for Krita, and set up an automatic payment profile with Paypal or a direct bank transfer. The business development fund has a minimum amount of 50 euros/month and gives access to the CentOS builds we make.
  • Individual donations. This depends a lot on how much we do publicity-wise, and there are really big donations now and then which makes it hard to figure out what to count on, from month to month, but the amounts are significant. Every individual donor gets a hand-written email saying thank-you.
  • We are also selling Krita on Steam. We’ve got a problem here: the Gemini variant of Krita, with the switchable tablet/desktop GUI, got broken with the 2.9 release. But Steam users also get regular new builds of the 2.9 desktop version. Stuart is helping us here, but we need to work harder to interact with our community on Steam!
  • And we do one or two big crowd-funding campaigns. Our yearly kickstarters. They take about two full-time months to prepare, and you can’t skimp on preparation because then you’ll lose out in the end, and they take significant work to fulfil all the rewards. Reward fulfilment is actually something we pay someone a volunteer gratification for to do the work. We are considering doing a second kickstarter this year, to give me an income, with as goal producing a finished, polished OSX port of Krita. The 2015 kickstarter campaign brought in 27,471.78 euros, but we still need to buy and send out the rewards, which are estimated at an approximate cost of 5,000 euros.
  • Patreon. I’ve started a patreon, but I’m not sure what to offer prospective patrons, so it isn’t up and running yet.
  • Bug bounties. The problem here is that the amount of money people think is reasonable for fixing a bug is wildly unrealistic, even for a project that is as cheap to develop as Krita. You have to count on 250 euros for a day of work, to be realistic. I’ve sent out a couple of quotations, but… If you realize that adding support for loading group layers from XCF files is already taking three days, most people simply cannot bear the price of a bug fix individually.

So, let’s do sums for the first 8 months of 2015:

Paypal (merchandise, training materials, development fund, kickstarter-through-paypal and smaller individual donations) 8,902.04
The big individual donations usually arrive directly at our bank account, including a one-time donation to sponsor the port of Krita to Qt5 15,589.00
Steam 5,150.97
Kickstarter 27,471.78
Total 57,113.79

So, the Krita Foundation’s current yearly budget is roughly 65,000 euros, which is enough to employ Dmitry full-time and me part-time. The first goal really is to make sure I can work on Krita full-time again. Since KO broke down, that’s been hard, and I’ve spent five months on the really exciting Plasma Phone project for Blue Systems. That was a wonderful experience, but it had a direct influence on the speed of Krita development, both code-wise, as well as in terms of growing the userbase and keeping people involved.

What we also have tried is approaching VFX and game studios, selling support and custom development. This isn’t a big success yet, and that’s puzzling me some. All these studios are on Linux. All their software, except for their 2D painting application, is on Linux. They want to use Krita, on Linux. And every time we are in contact with some studio, they tell us they want Krita. Except, there’s some feature missing, something that needs improved… And we make a very modest quote, one that doesn’t come near what custom development should cost, and silence is the result.

Developing Krita is actually really cheap. We don’t have any overhead: no management, no office, modest hardware needs. With 5,000 euros we can fund one full-time developer for one month, with something to spare for hardware, sprints and other costs, like the license for the administration software, stamps and envelopes. The first goal would be to double our budget, so we can have two full-time developers, but in the end, I would like to be able to fund four to five full-time developers, including me, and that means we’re looking at a year budget of roughly 300,000 euros. With that budget, we’d surpass every existing 2D painting application, and it’s about what Adobe or Corel would need to budget for one developer per year!

Taking it from here, what are the next steps? I still think that without direct involvement of people and organizations who want to use Krita in a commercial, professional setting, we cannot reach the target budget. I’m too much a tech geek — there’s a reason KO failed, and that is that we were horrible at sales — to figure out how to reach out and convince people that supporting Krita would be a winning proposition! Answers on a post-card, please!

Share on Google+0Share on LinkedIn0Pin on Pinterest0Share on Reddit0Share on Facebook0Share on Tumblr0Tweet about this on TwitterShare on VK

Be notified with new downloads and upcoming release information. We will not send you any type of Sale or Limited Time Only junk. Just the good stuff!