Open source twitter | Hacker News Reader

Open source twitter

identi.ca

I launched Twoorl (http://twoorl.com), an Open Source microblogging platform, about a month ago.

Cool, I'll have to give this a try. While the microblogging space is pretty crowded, I think the basic microblogging framework (short messages, ability to follow friends, mobile support, etc) could be re-purposed to do some other cool stuff. Having open source code to play with (whether it scales or not) definately makes it easier to get an idea off the ground and see if it sticks.

A few ideas:

- An interesting story telling format where people can "follow" a story and the author periodically posts updates from the perspective of the story protagonist. Isn't part of the appeal of twitter to follow the lives of friends? Could this same format be used for fictional story telling?

- Mash-up w/ Yahoo Fireeagle and Google Maps and you could have a mobile service for commenting on real world places / businesses and getting mobile updates when friends comment on something or someone comments on something that's near you.

- A twitter for images/photography (make a firefox extension so that users can easily favorite a photo on any website with one click and it automatically gets added to their stream).

Open source apps like this are great because they make it easy to try out an idea without investing too much time in it. Then if the idea then takes off you may want to think about re-coding a more stable version from scratch...

Okay - so now pretty much anybody can create a web site like Twitter. Do any of the open source apps have a full featured API like Twitter's?

The next step, it would seem, would be to "decentralize" it and get it to "just work" like instant messaging services.

Perhaps then Twitter could just abandon its web site all together and just focus on making its API as robust and scalable as possible?

If done right, I don't think the API would be hard to add in.

The real challenge is making sure that it can scale out to the number of updates and users as twitter has. Making something that looks like twitter isn't that hard.

True. The API is the easy part. Scaling is where the real test will be.

Decentralizing and API are coming soon apparently. The decentralized aspect through http://openmicroblogging.org/

Twitter as a platform was a shitty idea from the start. Imagine all blogs/subscriptions handled by one company. A group of big players and then the ability to host your own will make microblogging stable and popular.

I find 0 value in Twitter as a "microblogging" application. I know others do, though.

I find value in Twitter where it really shines: SMS integration, Jabber integration and great API.

For my particular use case, Twitter is a (very) different kind of instant messenger rather than a "microblogging" application. For me, none of these apps (dis|re)place Twitter.

I haven't tested it out yet, but according to RRW, it does have Jabber integration: http://tinyurl.com/3r7s9o I know nothing about programming for mobile devices so I'm not sure how easy / difficult SMS would be for them to accomplish. But, being open source, a strong API is inevitable as long as they can get people to contribute.

But then you also need the users if it's to be a successful communication tool.

I don't know, call me a skeptic but I think half of these clones are destined to fall on their face and the other half are destined to fade into obscurity.

There needs to be a compelling reason to switch from Twitter to this app/platform. And I'm not sure being Open Source or requiring posts to be under a "free" license is enough.

The "compelling" reason for most people seems to be Twitter's lack of stability. Personally, I've decided to accept and deal with it for now and give them a chance to catch up. However, two of the people I follow decided to jump ship, and it would be dumb of me to stick around if none of the people I followed were around anymore... Hopefully that doesn't happen.

I've tried to download the source and all I get is a page that says "It works!".

Has anybody found the source?

At the time that I clicked the 'Source' link, it told me they run the unmodified Laconica code, and directed me to the Laconica site: http://laconi.ca/Main/Source

GPL3. Blech.

Actually, no, it says on the source page that it is licensed under the GNU Affero General Public License Version 3.0, which is significantly different than GPLv3. So much so that it even has a different name.

What's wrong with free software, anyway?

Nothing is wrong with free software, I just think that if you're going to make it free, make it free, don't limit it like this:

"It requires the operator of a network server to provide the source code of the modified version running there to the users of that server."

I realize the Affero GPL3 and GPL3 are different. They both have the same limitation so I shortened the name in my comment to simply "GPL3". I understand the reasoning behind the clause, but I don't agree with it for a number of reasons.

You really shouldn't comment about GPLv3 if you haven't bothered to read it. GPLv3 says NOTHING even _remotely_ similar to what you quoted. In fact, the only real difference between GPLv3 and AGPLv3 is that AGPLv3 has exactly the requirement that you quoted above!

I don't care if you like GPLv3 or AGPLv3, but at least get your facts straight.

Actually... I have read it and that was my understanding of it. If I've misunderstood it then I'm open to corrections, I'm not a lawyer.

This illustrates yet another reason I dislike the GPL3--understanding it requires reading not only the license itself but also reading explanations and interpretations. This is made apparent by the thousands of search results leading to content that attempts to explain the license.

Reply on news.ycombinator.com