MediaGoblin crowdfunding campaign: huge success!



Well the campaign is over.... and before anything else, I think we should take a moment to say: We did it! The campaign was a success! Thank you EVERYONE who supported the campaign! YOU ROCK!

Our goal, technically, was 60k, but we weren't running an all-or-nothing campaign, so we decided to set our goal higher than our "bare minimum". In several meetings with Deb and Will, I remember saying, "well, if we hit 40k, I think I'm going to feel really good about things." And we didn't just hit 40k... we got a bit above it, beyond 42k even (which, as I said in a previous blogpost, has nerd-significance in showing our place in the grander scheme of life, the universe, and everything)!

This is no small accomplishment and we should feel proud of it... we deserve to feel proud of it!

So you are probably wondering! What exactly did you all finance? How can you expect this set of money to be used? Well, let me tell you! Basically: you have bought a year (plus a couple bonus months, actually) of me working on MediaGoblin fulltime! I take this very seriously... more on how I think that will pan out... but first let's have a breakdown of how we got to that.

We weren't sure originally how much money it was likely that we were going to be able to raise. This is one reason why we both shot higher than the bare minimum, but also why the video said "hire our talented developers and designers" while the actual campaign page said "hire Chris Webber, and if we get more, pay for even more people to do work". Basically, it was really hard to tell before the campaign launched how much money would come in. Would it be 15k? Would it be 150k? We didn't really know. If we raised enough, it would be great to pay for multiple people to work on things. We estimated about a 25% overhead... 10% on the FSF handling the fundraising infrastructure and finance stuff (which is less than this has actually cost them) and about 15% on fulfilling rewards and a few other costs. So 60k was the estimated cost to pay me for a full year... which would mean 45k after everything else. That's a smallish salary for a programmer (especially without benefits in the US, and especially with my spouse in graduate school), but livable, and we budgeted that as possible. So when Deb, Will, and I discussed what we would shoot for as a reasonable goal, we thought 60k would cover things for me to lead the project fulltime for a year. But would we hit above that or below that? We didn't really know.

Of course, we actually didn't hit 60k, and I'm still saying that I'm going to work a year fulltime, plus! So how does that work out?

Basically, there are one and half months left of 2012. I won't be taking any payment for that, but I will be working (hence the plus). A lot of the next month and a half will be involved with fulfilling donation rewards and et cetera regarding the campaign, as well as wrapping up a few small prior obligations. So that's the plus. So what that also means is that I will be working fulltime on MediaGoblin for a full year... and a very literal full year in multiple senses!... from start to finish of 2013. (To make up the difference since this does not cover my full expenses I will be doing some contracting for a little bit on weekends.)

I consider this to be a real job with real duties, so to that end, I'll be providing biweekly updates once the "full year" starts in 2013. I haven't figured out where yet, whether or not it'll be here or on the mailing list or to the donors list or on the wiki or on my personal blog, but I'll be making updates somewhere. You financed this... you deserve to be kept in the loop!

And there's quite a lot to do. We've laid out a roadmap to 1.0, and it's quite a tall order. By completing this fundraiser it isn't the case that we now no longer need volunteer and community development... we actually need that now more than ever! (One of the reasons I left my job to focus on MediaGoblin fulltime was because we have a strong community, but it wasn't getting the level of attention it deserved!) Whether or not you want to help code, do graphic design, write docs, do testing, or something else... we could really use your help! So please do join us!

Thank you all! Thanks to you, 2013 really will be the year of MediaGoblin development! I know I am looking forward to it!

---

Let's bring this home, decentralize media sharing, and save the kittens.


Joar Wandborg and Aeva Ntsc of MediaGoblin and Jaisen Mathai of OpenPhoto working together on the future of decentralized media sharing

So we're nearing the end of our campaign. We've done a good job so far! We're at about 40 thousand dollars now. Our campaign goal says 60 thousand... we don't have to make that full amount, but we should take this opportunity to get as far as we can get! I think 42 thousand is a good and realistic number at this point. If we can get past 42 thousand, MediaGoblin might not be the total answer to life, the universe, and everything, but it'll be a good sign that it's part of it!

Speaking about answers to things, I wanted to take a moment to talk about the opportunity at hand and why this all matters. We've spoken before about this a lot already... we've talked about why decentralization and federation matters, we've talked about what we're going to do to make that happen, we've spoken about the roadmap to 1.0, and even a glimpse of the features coming in the next release. So that's the world we're building, that's the future ahead!

But what about you? Why does it matter from your end? What can you do to help? Well, there are a lot of ways to help MediaGoblin... we have a huge task ahead of us, and of course joining our community and helping build that future is really important. But right now there's an easy way that you can help right now, right this instant: and that's to contribute to the campaign.

Your donation means a lot here. Whatever you can give, it helps. Even with as many awesome volunteers as we have... in fact, especially with as many awesome volunteers as we have!... coordinating MediaGoblin, building a coherent and beautiful media publishing experience, requires a ton of dedicated work, planning, engineering, and communication. By donating to the campaign you are helping power a year full of dedicated focus to keeping that machinery smooth and easy running so that no patch or contribution is left behind. There's a lot of grit-work involved in making the world's most beautiful media hosting future come true, and let's face it! We're up against some strong, well funded competition. It's a tough place, and as we've said in the campaign video, if we don't do something about it, it could be a real cat-astrophe (yes, sorry, it's a horrible pun).

But the reason we'll make it is because we are building the right world, the world of media hosting that the internet needs. And we have the internet on our side. We've got you on our side!



So show your support! Make the web a better place, for you, and for kittens everywhere! Donate now!



---

Upcoming Features in 0.3.2



Heya all! We had hoped in some ways we would be able to get out a new release mid-campaign... but it turns out the campaign itself has its own pull of energy. Well, we're nearing the end of it (you've donated, right? We could use your help!)... but we thought we'd give an update on the state of the code. Like the drawing above, it's not quite complete... but getting there! So in this post we hope to detail all the awesomeness that's on the horizon.



So let's get to it! The first feature is... yup! 3d model support! We've already blogged about this in detail,

But hey, why not throw in a bonus! Here's a higher res image of that print that came out per the video shown in that previous post:



Anyway! So that's one exciting thing. What's up next? How about API support! All you out there wondering when we're going to get the ability to upload from your phone, or your desktop, or whatever? WELL! Thanks to Joar Wandborg and the Icelandic government (who funded the feature!) we now have an API! This means that if you want a mobile upload application for your phone, the infrastructure to build it on MediaGoblin's side is falling into place. And if you want to build a tool that takes a screenshot and auto-uploads it to your MediaGoblin account, you can! (Why not!)

Anyway it's hard to create an exciting screenshot of an API, but Joar created a few demo applications, so here's some screenshots of that page on GitHub (gasp! centralized service!)



And also...



What else? How about collections! This next release brings collection support, which means that you can organize media into custom galleries! Want to sort things based on pictures of cute puppies? Maybe something else? Below is an example of Joar making a gallery of some Tears of Steel clips:



Anyway. That's a whole bunch of features! There's a bunch of other small things as well. For example, thumbnails are now much nicer looking in the current release. But we still need to do quite a bit of cleanup before we can package all these features all into a nice and clean release. Part of the reason this has been stalled though is the very same reason for this fundraiser though. Coordinating all this takes lots of work... we really need dedicated coordination and development time to pull this all together. But as you can see, we're building awesome futures... help us make them even more awesome! The future of MediaGoblin awesomeness is powered by you... so support MediaGoblin!



---

What's coming in MediaGoblin 1.0?



So the MediaGoblin fundraiser is in its last week, and we're coming near our goal of funding one year of dedicated MediaGoblin development. We're hoping over the next year to hit a point of MediaGoblin for the masses... AKA MediaGoblin 1.0! But wait... what exactly is MediaGoblin 1.0?

To answer that, we need to go over two things: what we already have, and what we're going to have. Let's go over the first list first...

Features we have (and will continue to improve on before 1.0):

• Image, video, and audio publishing

  MediaGoblin provides image, video, and audio publishing out of the box, making it a self-hostable and free replacement for services like Flickr, YouTube, and SoundCloud.

• Flexible storage system

  MediaGoblin has a flexible storage system built in. It comes with a local file storage option as well as integration with OpenStack's "Swift" storage system, so you can offload hosting to something like Rackspace Cloudfiles or equivalent. You can even add entirely new storage mechanisms.

• Media types are extensible

  Images, video, and audio are just the beginning. MediaGoblin was designed to be extensible for entirely new media types. For fun, you can use the AsciiArt example media type, or you could even build something entirely new. Want to support presentations or e-books? Or maybe you want to have 3d model support? MediaGoblin was designed from day 1 to be extensible to fit the needs of whatever types of files you want.

• Commenting

  MediaGoblin supports basic, markdown-enabled commenting on media entries.

• GeoLocation / OpenStreetMap integration

  Does your camera have a GPS? MediaGoblin can put your photos on a map, so you can see where they're taken.

• Galleries and collections

  We show the usual galleries of a user's media, as well as special galleries for tagging, and we also support user-curated "collections" of media.

• "Attachments" (for media source files and etc)

  If you have files that you used to make the final piece of media you're sharing and you want to share those also, MediaGoblin has an option to allow for that.

• Beginnings of a plugin system

  We have the beginnings of a plugin system, with already a couple of useful plugins (flatpages, so people setting up terms of service or similar non-dynamic pages are easy to add).

Features that aren't developed yet but should be for 1.0:

• API (for desktop/mobile app integration and so on)

  We have an API that is close to being completed, but isn't yet. This should allow external application authors to integrate directly with MediaGoblin instances in various ways. This will allow for mobile phone uploading of photos and videos, or direct-from-desktop uploading, or integration with other services.

• Federation

  One major feature is the need to hook together separate MediaGoblin instances and provide a social experience that feels as clean as if it users were on the same site (similar to how users on different email servers can communicate to each other from different servers but it feels that things are on the same thing).

  We intend to use the meta-standard OStatus to provide the following federated functionality:

  • Ability to subscribe to other users across instances
  • Favoriting across instances
  • Cross-instance collections/galleries (think Flickr pool)

• Audio and video podcasting support

  With podcasting support, MediaGoblin will enter a new area of usefulness as people are able to use it to broadcast shows to subscribers.

• Admin tools

  MediaGoblin needs better tools for administrators to be stewards of their instances, both to promote things appropriately and clean out problematic content and administer users.

• More and improved plugins

  We have plugin support already, but more plugins need to be built. These include:

  • Comment spam prevention
  • Moving some extensions into plugins (OpenStreetMap, etc)
  • Limiting storage and upload space for various users

• Easier installs and improved documentation

  MediaGoblin needs to be even easier to install and develop. We can improve this situation with better packaging and stronger documentation.

• New media types

  In addition to the media types we have, new ones are on the way: 3d model support is close to being polished, and there may be other types added soon.

• More general polish

  MediaGoblin is well working software, but there are a lot of rough edges that still need to be smoothed out to make the experience better for users, such as cleaner submission processes, various adjustments to theming, and so on. We will continue to add shine to MediaGoblin as we approach 1.0.

Wow. Tall order! But if you look at the first section and the second section both, you can see we've already accomplished a lot. We can do this... but we need your help to make it happen! MediaGoblin has a wonderful volunteer community, but it takes a lot to coordinate... the closer we can get to our full goal, the better we can do to get a full and uninterrupted year of MediaGoblin development and coordination. Help us make all our dreams come true... Support MediaGoblin!



---

Donate to MediaGoblin, get a chance at free PyCon tickets!



Heya all! We've got an exciting announcement to make this morning! We got a generous promotion/sponsorship from the Python Software Foundation... the next 25 people who donate $200.00 or more will get a chance to get free tickets to PyCon US!

If you're into programming, and especially Python, PyCon is a conference that's not to be missed. I can say from experience: I've attended every US PyCon since 2008, and it is honestly the highest quality programming-oriented conference I've ever gone to. (I've even spoken there before and Deb, Will and I ran a poster session on PyCon in the past as well!) And this is a great deal: the individual rate for PyCon is $350 at regular rate, and $300 at early bird rate. Given that we're giving out two tickets to 25 people, that's a 1 in 12.5 chance you'll be chosen. Those are good odds!

And of course, either way you're already helping out a good cause by donating to MediaGoblin. You're helping the dream of decentralized media publishing come true. And, as it turns out, you're helping prove that Python web applications can help save the world. What more could you ask for?

Well, you could maybe ask for more rewards... at $200 you've already passed the $100 level:



That means that in addition to a chance to get a free trip to PyCon, you already get a t-shirt, a postcard, a sponsored commit, a virtual hug, and stickers! But if you're really generous, you could go up just $150 more:



At that point you get a 3d printed figurine of our mascot, Gavroche... plus a t-shirt, a postcard, a sponsored commit, a virtual hug, stickers... and of course your chance at a ticket to PyCon!

So what are you waiting for? Support MediaGoblin and help make the future of free and open source, python-based, decentralized media hosting come true... and get a chance to go to the world's greatest programming conference in the process!



---

We made the 10k matching grant!

Hey all!

This is just a brief blogpost to announce that we made 25k, which means we met the 10k campaign matching!

This means that yes, the forementioned magical 10k drop is going to happen. However, it's also the weekend, and so presumably that will happen when the FSF gets into the office.

But the answer is: yes! We did it! We made the 10k matching thanks to you! So thanks to both our anonymous donor and to you for making that happen. You rock! :)

EDIT: It's updated! (Within ten minutes of me writing this post, even! :)) The 10k donation is now reflected in the progress bar :)

---

MediaGoblin (and OpenPhoto!) at Federated Social Web 2012: a recap

This last Friday, MediaGoblin contributors Joar Wandborg, Aeva Ntsc, and myself all went to the Federated Social Web Summit 2012. And we weren't alone! There were many brilliant minds thinking about federation issues at the summit.


Jezra on the left, David Crossland of Open Font Library on the right, and I forget whose back of the head that is.

There was a lot of exciting conversation going on generally, from conversations about standards like dialback authentication, to various applications implementing federation, to various conversations about how we can make the user experiences of federation and etc better. If you're interested, you might be interested in checking out the Etherpad document that some were annotating during the summit.

But what I was really interested in discussing at the summit was how to get federation working nicely in regards to media publishing. And I'm happy to say that we had a wonderful set of conversations about that! In fact, the conversations were made even better from the presence of Jaisen Mathai of OpenPhoto!

Sometimes people ask if MediaGoblin and OpenPhoto (which were both announced within a short time period) know about each other. They sometimes wonder... are you bitter enemies? Do you battle? If the project leaders in a room together and shook them up, would they break into fisticuffs?



Actually, the answer is that we do know about each other... Jaisen and I have been in communication since fairly early on and are both in support of each others' projects. We see ourselves as allies trying to tackle similar problem domains from different angles. And one thing we're both interested in is getting our projects to be interoperable as in terms of federation!



There were several major issues we wanted to discuss in making federation nice between instances:

• Subscribing to users across instances
• Favoriting across instances
• Cross-instance collections/galleries (think Flickr pool)
• How to make presentation look good across instances
• Making things private, and different types of actions/permissions

That's a big set of problems, and not super easy to solve! I typed up a bunch of notes from our conversation... here they are, distilled a bit:

• Okay, good news first! For the most part, subscribing to instances already works fine with PubSubHubbub. It works fine for existing "subscription" type federation, and will probably just work fine for subscribing to other peoples' galleries here too. You can see here a thread with me talking from identi.ca to netziens of rainbowdash.net (a brony community hosted on StatusNet) who suddenly realized I used them in a federation example in the campaign video.

  

  When you think of the subscription needs of showing a gallery with thumbnails, such as on MediaGoblin, this same type of system should work well enough.

• A bigger concern is, "what to do about mismatched thumbnail sizes?" and other display issues. OpenPhoto currently takes the approach where you can request any image size you want and it'll make and cache it for you on the fly. We can't really do that for MediaGoblin: this doesn't really work for audio and video (and while I like the idea I'm still not sure it's a great one for images either, especially not given MediaGoblin's processing infrastructure). If we want galleries to really look nice across instances it would be good to agree on some "common" thumbnail sizes and etc, as well as having adaptive CSS for the cases where a "good fit" thumb is not provided.

• For this all to really work, taking actions on remote instances has got to get easier. For example, subscribing to users on another instance is still a huge pain in the ass on StatusNet instances. For example, subscribing to someone on another instance is a huge pain in the butt... if you're on another site, you have to tell it where your user is before you can do anything:

  

  And even after that you have to confirm on your own site. This is an awkward user interface, even though it works for these things. But it doesn't work for everything... if I just subscribed to a user and they had an old post that I'd like to favorite, I can't do so because it isn't on my timeline. And even if it is on my timeline, I have to dig through the history to find it instead of just favoriting it from the site that I found it on. That's a frustrating user experience... but like many user experience frustrations with federation, it's frustrating for a reason: it's a hard problem to solve.

• Luckily it looks like there's a good way forward to making this easier. There's work being done in the proposals of WebActivities and WebIntents which are standards which should help in the process of saying "When I do X task (such as favoriting, subscribing, or adding to a collection), I want to use Y service". There are a whole bunch of tasks that will need such functionality to have a clean implementation.

  I'm not sure this solves All The Things; I'm not sure how you'd get an indication of whether or not you've already favorited such a thing on a remote site without having some sort of security vulnerability or privacy issues, and I haven't looked at the spec enough to see if that's exposed. But it definitely looks like it's going to help things go in the right direction, and we'll need it for a lot of stuff to feel smooth: subscribing, favoriting, adding to collections, etc.

• How to handle private sharing across instances? Probably something like Salmon might be right, but it's a tricky problem. The topic of "what happens if someone makes something available to you but then retracts that" came up. Frankly, on the web, it's impossible to fully retract something and trying to force it often makes the situation worse (AKA the Streissand effect). It's possible that servers might publish a deletion notice and that could be interpreted as a "please" remove issue, but that's probably not enforceable. Furthermore, there are a whole series of issues with implementing privacy right. We agreed this is something that we'd like to do, but that getting media federation to work at all correctly publicly is a bigger first priority.

• There was a bit of discussion about how to handle some other actions, such as Jaisen suggesting a button to mirror content onto your instance. (I thought this was interesting; in the past, a good number of people have suggested that we implement mirroring the content of people you subscribed to, and I've said that I think this is a bad default but that we could have a "diskgobbler" plugin in the future or etc. This solution actually sounds more interesting to me.) There might be additional actions too, such as "get a print", or even with 3d models, "send to a botfarm to 3d print" (well, that requires some future 3d printing botfarm infrastructure that doesn't exist yet, but I know Aeva is thinking about it), or maybe even "play on a stereo TV at home". These are interesting ideas, and not necessarily directly federation related, but it was interesting to discuss how they might be done by having some activity represented in an Activity Streams feeds and how webactions/webintents might play in.

Anyway, we left with a lot to think about, which is good given that federation is a big next step on the agenda once the campaign wraps up. And events like this are great opportunities to connect... in this case, it was both great to meet up with other MediaGoblin folks (this is the first time I'd ever met Joar in person)!



Furthermore, if federation is going to succeed, it's going to require the hard work and collaboration between a bunch of projects. So I'm glad that Jaisen of OpenPhoto was around for us to have such conversations!



While we're at it, I might as well remind you! We're in the middle of a crowdfunding campaign, and all the above is an exciting future, but its success depends on whether or not we can get focused development in to MediaGoblin. This is why I'm asking you: if you haven't yet donated to MediaGoblin, please do so! The future of media decentralization is in your hands!



---

MediaGoblin now with 3d support: MediaGoblin community and Lulzbot team up

I have some very exciting announcements this morning. Basically, thanks to generous support from LulzBot and one awesome MediaGoblin contributor Aeva Ntsc, we now have 3d model support in MediaGoblin! This awesome news means that MediaGoblin becomes more useful for 3d artists and is well on its way for those looking for a free software, customizable alternative to something like Thingiverse! (Goblinverse??)

I have plenty more to say, but demonstrations speak louder than words, so check out this video:




Thrilling stuff, right? What's also thrilling is just how fast this all came together. Last Sunday Jeff Moe of LulzBot made a generous donation to MediaGoblin's campaign and said that whoever added 3d model support to MediaGoblin would get their LulzBot AO-100 3d printer as a reward (which, by the way, is the first ever FSF certified piece of hardware). Before we could even make a public announcement, Aeva blasted through the implementation... in just two nights! It's pretty cool... it uses Blender to render the stills and has a WebGL browser preview using thingiview.js (an LGPLv3 based JavaScript library). Anyway. Thrilling stuff.

I think this news was exciting for everyone, but especially for me! When I give talks about MediaGoblin I frequently say "the media types layer is extensible, so you could even add support for something like 3d model support if you wanted to." I didn't expect us to get this so soon... I guess I'll have to come up with a new line. I guess that's the best kind of problem to have!

Code is not yet merged into master but you can check it out here if curious. We are hoping it should land in the next release.

Oh, and did I mention that we're running this crowdfunding campaign? Every bit helps, and we've still got a ways to go! Contributions are currently being doubled... so your support is welcome. And tell a friend! The world's most beautiful media publishing future is counting on *you*!



---

Amazing MediaGoblin fanart by Justin Nichol

Something fun today... last night I got contacted by Justin Nichol, who is self-described as "an entertainment designer, game designer and free culture advocate". He said he heard about the project, thought the mascot was cute, and wanted to paint it in his own style. That didn't prepare me for the awesomeness that awaited me when I opened the attachment:



Well, what can I say. Beats the heck out of anything I can do! I'm pretty blown away, and quite frankly, both humbled and flattered by having my character drawn in a way that so massively outpaces my own skills.

Justin says the above image, as well as this dark background version, are both released under CC BY-SA 3.0. So feel free to remix and reuse as long as you reciprocate under the terms of the license (be sure to attribute Justin Nichol!) Justin also uploaded the image to his CGHub account, so you can also see it there.

Oh, and on that note, I encourage you to check out Justin's other stuff. He's got some crazy awesome stuff on his CGHub account. Even cooler: he's got a game website called Black Flag Games, which appears to be tabletop free culture games. Pretty cool... and important work! I happen to think that games are important to free software and free culture and when I worked at Creative Commons that was an agenda I tried to push forward. So check out his stuff! Looks like they've got some cool stuff going on...

Oh, and speaking of cool stuff, you know we have a crowdfunding campaign going on right? Now's the time to donate since contributions are currently being doubled. And I guess I'll put the "Support MediaGoblin" image below... even though it noticeably pales in comparison to the above. :)



---

How to respond to a YouTube cat-astrophe? Decentralize the web!



In the video we made for the MediaGoblin campaign, there's a part of the video which says: "What would happen if YouTube went away? What would happen to cat videos on the internet? It would be like a cat massacre." People seem to really respond to this part of the video, which is good (though they usually ask me how we resisted the pun "cat-astrophe"... I guess with the title of this blogpost, we finally gave in). And every now and then we get a reminder that this isn't just a vague possibility: these things can... and do... really happen.

Today, for a few minutes, YouTube went down. For a brief moment in time, millions of cat voices cried out in terror and were suddenly silenced. Now, granted, they came back a few minutes later. But within the same short interval that I heard about YouTube going down, various programmer friends of mine started complaining that they couldn't get any work done because GitHub also went down due to a DDOS attack.

Is this because YouTube or GitHub are badly run, or that the companies that run them are inherently evil? No, I don't think so. But there is a structural problem, one that's the case with any major centralized service: when that service goes out, it takes everything it hosts out with it. This is a reminder that these types of institutions, even when run by brilliant and wonderful people, have inherent flaws as they become large, centralized behemoths. Even the nicest, most well run of centralized behemoths can fall. And will.

And as we point out on the campaign page, it's entirely possible that your favorite large, centralized service could go away permanently. In fact, some day it probably will. Geocities might have seemed like a joke to everyone by the time it disappeared, but in 2000 it seemed like a huge institution that would never go away... but then it did. Maybe some day YouTube or Flickr will cease to be profitable, and then those will go away. It could happen... it nearly happened to Google Video.

What's the cure? Bring the web back to be a decentralized place, the way it was intended to be. This isn't an easy task, though: services are getting larger and more complicated, and need a lot of special expertise to get them working properly. The good news: we are already working toward that future. Could you help us out?



Thanks for all you do,
  -- The MediaGoblin team

---

« Page 6 / 8 »