Ask HN: have an idea of "Twitter for developers", thoughts?

2 points by amirrajan 12y ago ↗ HN

We're developers. We can clone a git repo, and deploy to heroku, aws, etc. Why can't we have an open source, single user, twitter like system, where status updates are hosted by each independent developer (as opposed to a centralized app like twitter or app.net)?

To elaborate:

- You'd clone a git repo, and deploy this app to your favorite web host.

- There would be integration with twitter api's so that any status updates you post on your personally hosted timeline are also posted to twitter.

- Twitter API rate limits wouldn't be a problem, because you are technically the only user of the app.

- You would also get all of your twitter timeline, just through your personally hosted app (this will ease migration away from twitter)

- If you want to follow someone, you just enter the uri to their stream and you'll start getting their timeline updates (no need to follow them on twitter...you can of course still do this).

Key benefits:

- We won't be starting from scratch, status updates and timelines will still be on twitter. But eventually, a community will build up that's our community.

- Barrier to entry is small for developers. You clone a git repo, and push to a free heroku instance.

- free for the little guy... if you get 20k followers, you may want to buy web workers on heroku. But this a good problem to have, because: you own your brand, and can profit from that brand. not twitter.

- You own your data. not twitter.

- No single point of failure for status updates.

- This new distributed client can evolve past what twitter provides. And it'll be driven by what we want, through contributions to the central repo.

I've already started work on it: https://github.com/amirrajan/sortis/blob/master/README.md

5 comments

[ 4.2 ms ] story [ 21.4 ms ] thread

mattboehm2 12y ago ↗

So... centralized logging?

amirrajan 12y ago ↗

Did you mean to say decentralized logging?

Never mind, I misread what you were proposing. The biggest concerns I can see is:

* I doubt twitter will let you just take their API * No matter how much you try to simplify it, managing this is going to require work. It doesn't seem worth the work just so you can own your 140 character blurbs. I can see businesses wanting twitter on their intranet, but there are already many solutions for this.

I haven't looked too much into it, but you might want to look at the diaspora project as it shares a lot of these concepts.

loumf 12y ago ↗

I don't have a problem that this solves. I don't consider tweets to be important data -- the value of a tweet goes to near 0 after 24-hours.

Problems I do have

1. Noise to signal ratio in reading tweets

2. Having ideas for what to post

3. Getting my attention drawn to important tweets

4. Managing interactions

5. Increasing relevant follower count in organic ways

I have solved some of these with custom clients, lists, ifttt scripts, and buffer

The work I've done on the app so far has centered around managing tweets I've favorited: exporting, searching, tagging. Thanks for the list. I'll keep these in mind when incorporating new features.