Does Open Sourcing Niche Web Apps Work?
Recently, we've seen a couple of our favorite web apps go open source. The code for FF To Go, the popular mobile client for FriendFeed developed by RSSMeme creator Benjamin Golub, is now available under the MIT license and Snackr, the RSS AIR app we adore has posted their source code on Google Code. Is this the start of a new trend for niche web apps? And what does this mean for the future of their development?
Why Go Open Source?
Of course, we can't talk about the trend in open sourcing web apps without mentioning the big news about Reddit. Recently, the social news site decided to open up so that you can now build your own version of Reddit. The difference here is that Reddit is large enough to not be considered "niche" - it has 4.5 million unique visitors per month and has grown 1000% since the acquisition by Conde Nast's Wired Digital division.
However, when compared to competitors like Digg and Yahoo Buzz, it's clear that Reddit is the underdog here. If anything, the decision to go open source was a move to compete in a way that Digg and Yahoo cannot. When speaking as to what they hope to accomplish by the move to open source, co-founder Steve Huffman said that they're hoping users will tweak what they want changed and add new features. The beauty in this is that even if users develop new features, they can't necessarily be used to go and build a competitor to Reddit - because the code is open source, Reddit could just implement those features themselves.
Small Apps Open Up
While open source might be a good thing for a community the size of Reddit, there's still concern that when smaller web apps go this way, it means the development cycle is going to slow down.
Take for example, Enso, an app which provides an alternative way to interact with your computer, similar to what Launchy offers. In March of this year, this niche app also went the open source route via the revised BSD license. In this case, the move was most likely due to the fact that the Mozilla Foundation hired away three of the principals from Humanized, the company behind both Enso and Songza. But immediately upon the announcement of the move, there was concern echoing in the comments of the blog post:
Will the Humanized folks still be leading the charge on Enso? or are leaving it adrift?
Humanized, please don't leave Enso adrift.
As to whether Enso has indeed been set adrift...well, no one from the Humanized team ever responded in the comments section and the Humanized blog is now promoting Mozilla. A May 6th entry even let everyone know Mozilla was hiring and provided an email address to apply. In addition, four other of the May blog posts were about Firefox.
When it comes to FFtoGo, though, creator Golub assured me on Twitter that he's still developing the app, but he hopes others will start sending him patches soon. (We hope so, too!) Snackr also is currently going strong, but they only went open source on June 5th. Still, they've already posted a couple of test builds on Google Code, which implement new features like the ability to star new items, change ticker transparency, customize how old the items displayed can be, and more.
Even Bigger Apps Can Lose Momentum
However, the concern for the lack of development isn't entirely unfounded. Take for example, another open sourced app, this one from Socialtext, maker business social software. In July of 2006, they released Socialtext Open, an open source version of their wiki software. But it's a good thing they didn't stop development of their own because, by the looks of it, Socialtext Open has only had one patch in 2 years:
Excited or Worried?
Those are just a handful of examples of apps going open source, but there are many more. Of course, this isn't to ignore the fact that Sourceforge does quite well, it's just to raise the question - does open sourcing a niche web app really encourage more development? When the community is so small, if the creator doesn't continue to work on the app, how does the app really fare? And has a favorite app of yours ever been abandoned?
Photo Credit: Free Beer by Henri Moltke
ShareRelated Entries
1 TrackBacks
TrackBack URL for this entry: http://www.readwriteweb.com/cgi-bin/mt/mt-tb.cgi/4304
Comments
Subscribe to comments for this post OR Subscribe to comments for all ReadWriteWeb posts
Great contributions, keep it up!!!
There is much more to open sourcing a project then just getting more help.
1) It is the ultimate in transparency. People have no reason to fear using fftogo because the source is right there for everyone to see.
2) It is great for communication. When I make a change to fftogo everyone will see it (a RSS feed of my changes gets pulled into http://friendfeed.com/rooms/fftogo).
3) It is a learning opportunity. Anyone interested in Python, Google App Engine, Django, or FriendFeed could benefit from just looking at the code.
I don't mind if nobody else ever helps out; honestly I care about the points above more than getting help.
Those are good points, Benjamin, and I know the community is glad that those are your reasons for doing so! :)
Posted by: Sarah Perez
|
June 26, 2008 12:43 PM
I think one of the main benefits for a niche app to go open source is to get the input and feedback from a much larger community. One of the problems developers have working on niche apps is that its hard to connect with other developers who are working in the same niche. Going open source solves this problem to an extent. The other benefit is that since these apps are niche, they likely wont get over run with too many contributors which is a common problem with very popular open source apps.
As I understand it, open source projects have two models that work:
1) Create a nice app that doesn't need to make money. This app is used by multiple teams around the world, all doing their best to maintain it. Linux started this way. Joomla is a nice living example.
2) Go head to head with Microsoft, trying to cut into their monopoly. Then, get a sponsor like Google, IBM or Sun to fund you. This is what Ubuntu, Firefox and Openoffice are at now.
Personally, I liked #1 better, as #2 just helps other giants get richer in a different way.
I don't see any web application, that doesn't fit one of these definitions do very well.
We release the source of GroupServer, the software that powers our web and email app at OnlineGroups.Net.
Though our time-between-releases has been large, we have no plan to off-load development to the community. Of course we aim to build contributions (Scott, being over-run with contributors would be a good problem to have), but we have further reasons for the move.
Like Benjamin, we know transparency builds trust. We want our customers to know what's under the hood of the system that they entrust their data to. In fact, we want them to use our service by choice, not because it's their only option.
Releasing the source is also consistent with our philosophy of collaboration, where it makes sense. After all, collaboration is what our software is for. We benefit from open source tools ourselves, so it makes sense to contribute back. It doesn't cost us to give away software, and the more we give away, the more people will come to us for what we're good at. We do aim to build our consulting and support services business.
Feedback is often as useful as development input. We get plenty of interface feedback from users, but we couldn't get feedback about the back-end, without releasing it. We want to make significant, widely installed software. Competitors to our software service are not a threat, because they are just as exposed to competition as we are. The bigger the pie, the better our piece can be. Amir, whether we can "do well" like this, depends on how you define that.
Thanks for the survey of web apps releasing their software, Sarah. It would be interesting to see a list of companies successfully providing software services with open source software. Automattic, SugarCRM…?
We do a open source cms on the .net platform - umbraco
-which was open sourced about 5 years ago.
Now there were a number of challenges:
#1, it was a typical small agency cms, without a community, and not a very large user-group. So there was nearly no community backing.
#2 it's .net based, and the .net developer culture has never been that into the whole sharing and community thing.
And the only reason this was been a success (with 50.000 active umbraco installs it is a success) Is because the core has been persistent and believed open source was the right way to go. For the first many years, nearly no patches were committed, and no-one joined the core team.
So what we learned is that, to open source a project, you need to have the same commitment and persistants as you would have in a closed-source project, if you don't, you are probably open-sourcing for the wrong reasons.
As Dan from Groupserver mentioned in his comment, there has to be a sweet spot that takes the best of what opensource has to offer and mashes it with the benefits of a proprietary model. I posted about it last year as a response to readers who'd seen my concept about combining SaaS/ and opensource to both provide a valid business while at the same time leveraging community.
Seems that the opensource and proprietary web apps are to a certain extent trying to eat each other's lunch instead of concentrating on the "bigger picture"
This is very important, and a point missed by the blog post. You don't open source by setting software adrift and hope that someone picks it up. You do the same things as if it were closed source and if someone contributes, fine, if not, then you are no less off than before.