In November 2012, the MediaGoblin team already ran their
first fundraiser. The objective was to get Chris Webber to work on MediaGoblin full-time for a whole year. Chris kindly agreed to answer a few questions about the outcomes of that campaign and clarify some of the development plans.
Hi Chris! What are the results of the previous campaign? You got less money than you needed, but agreed to do the planned work nevertheless. How did it go?
We did get less money than the full goal, though we did know this might happen. At the end of the campaign we ended at $42k, not the $60k we had set as the goal, but we still considered it a success and
Part of this was because we knew that we could set a higher goal and still do okay, if we didn't make the full amount, since we weren't doing this Kickstarter all-or-nothing style. So, this made it possible to fund a year's worth of work, just on a very tight budget.
Overall that work went well, and we got a ton of stuff done across five major releases. Lots of hard work! We're very grateful that the community helped us make this happen.
We also were fortunate to be able to
participate in both Outreach Program for Women and Google Summer of Code.
This helped us out a lot. We had some really amazing students. Basically all the projects were successful, and the projects that haven't been integrated yet are very close to it.
What do you think you've learnt from the first campaign, and how are you going to build upon that experience this year?
Unfortunately, the "we didn't get 60k but it was still a success" message doesn't really explain clearly to users a few things. How much really do you get out of the amount of money you put in? How much is really needed to do what? This is something we're improving this year with the "milestone" unlock system:
Of course, we will accomplish more than just those details, but I think this makes clear, what a lot of the big picture goals and plans are, and just how much people would get out of it, should the funds be raised.
Although fairly informal (details are hard to work out until we ensure the funding is there), the goblins more or less represent the number of resources I actually hope we can get into the project at each level.
When we ran the first campaign, my co-conspirators Deb Nicholson, Will Kahn-Greene and I all talked about having stretch goals, but it was hard to plan out what how much resources would actually mean. Now that we've done this once I feel confident in our plan there.
Another directly-campaign-related thing we learned was how much impact the video had. People really got into it. I had a bunch of people respond to me that they weren't clear why they should care about these issues, and then after they saw the animations in the video it became really clear to them.
I was happy to hear that. There's a secondary agenda there in the video to not just get people to donate (though of course that's a big goal), but also to get people to understand and care deeply about the issues at hand. I also really would like to see messaging about computing freedom issues improve and become more accessible, so hearing people saying that we seemed to be succeeding in that goal. That made me really happy.
This year I actually took a week to stay with friends Bassam Kurdali and Fateh Slavitskaya of Urchn / Tube fame, and they helped greatly with planning the animation and revising the script. I think it paid off.
That's not to say the video was perfect... but I'm happy to have it both achieve the direct goal and also my other not-so-secret agendas of messaging.
Another huge lesson from the campaign is that free software enthusiasts are (I suppose unsurprisingly!) our most enthusiastic group of supporters. I'm still interested in getting people outside of those communities engaged, but it's great to see that free software people tend to put their money where their mouth is and really care about these. That's good... we want to do right by them.
Oh yeah, and this year people can donate in bitcoins. The address is right there on the donation page! ;)
The campaign's page focuses on federation, wrapping up 1.0, and podcasting for step 1, privacy features are among stretch goals. Why this order of implementation?
Federation is first and foremost. We had good progress on this through Jessica Tallon's work through Outreach Program for Women, but federation has been one of the main reasons for MediaGoblin from day one, so we can't hit 1.0 until that work is done.
Furthermore, we're expecting to base the first half of privacy work off of the federation work. The Pump API has some methods of handling private sharing, and we'd like to base our private sharing support off of that.
The second branch of privacy support is to implement an encrypted backend via Tahoe-LAFS. This is something we've wanted for some time.
The idea is that people might want to host their media in some remote (I hesitate to say, but "cloud") storage service. The problem is, that's totally vulnerable for spying on users' activities. But with Tahoe-LAFS, the storage server itself doesn't even know what the contents of the media it is hosting are.
That, combined with private media sharing could be really excellent. It's not as high of a priority as the rest of it, so it's a bit higher up the milestone list, but it's something we'd love to have.
Can you think of any notable Mediagoblin installations you've been made aware of for the past year?
Most of the installations have been pretty small for family and friends, which is actually pretty nice to see in its own way. On that front, my favorite is still MediaGoblin contributor
He took that as an opportunity to build a very cool looking MediaGoblin theme called "Sandy 70s Speedboat". We hope to build that into MediaGoblin itself soon. :)
The FSF ran a MediaGoblin campaign for
hosting their videos from LibrePlanet, and that was pretty exciting to see.
Do you think you MediaGoblin is ready to be deployed by virtually anyone?
It's still more complex than we'd like it to be to install. This isn't actually our fault I think. We've worked pretty hard to improve the process, but installing modern web applications tends to be a challenge.
Most people don't see that though, because they aren't the ones running the software... some team of devops are! But free network services don't have that luxury, so it's a challenge we have to tackle.
To that end, you'll notice that some of our milestone goals involve improving deployability!
How much has the project changed since the last campaign in terms of organization and ecosystem? More new developers? More activity from existing developers?
Developers are steadily increasing, yep! The last year was really good for getting a bunch of new people in. There were times over the last year where so much was going on, it was hard to wrap my head around all of it.
That's slowed down a bit, because managing that level of contributions is basically a full time job, and I've had to plan for how to fund that job again. Hence the campaign!
It's a real opportunity though. If this campaign is successful, I suspect the next year will blow the last year out of the water as in terms of activity. I'm looking forward to it! :)
Last year there was that crazy idea coming from Tube project team: turning MediaGoblin into an asset management platform for VFX houses. What's in place, what's still missing? And what's the priority of this idea?
There's definitely a lot of interest in this from the Blender and open movie community. I have some ideas around it, but I think it will require some more infrastructure.
If we do it though, I want to do it cleanly: MediaGoblin should always be first and foremost a media publishing tool. We should shoot to be the best of that first, and then we can possibly build something cool like this on top of it! :)
If you think MediaGoblin is an interesting project to support, head over to the