Hacker News new | past | comments | ask | show | jobs | submit login
Announcing the App.net File API (app.net)
91 points by anu_gupta on Jan 28, 2013 | hide | past | favorite | 33 comments



Dalton is very patient, damn frugal, and smart. His API programming team is off-the-charts talented (essentially 14 Stanford and Carnegie Mellon guys), and here's the important part - they get along very, very well. So, the team is tight, and thinking long term.

They've got a rock-solid cash position, and I wouldn't be surprised for Marc Andreessen to re-invest after 2 years, just to own a piece of this flex infrastructure for such a small $ amount.

Today's announcement is exactly what's next. Building more ways to "roll your own social network". The API is amazingly rich - the developers love it.

For example, messaging is far beyond Twitter's broken DM mechanism. The API allows you to DEFINE messaging protocols, and uniquely build a social network with a distinct message system. It's software-definable, via the API.

What's actually happened in the first 5-1/2 months is that the "core API" is finally finished. They've actually unbundled all the core social 2.0 infrastructures.

Watch for more creative "edge API" ideas (like this social Dropbox) to come out, now that the core team of 14 is freeing up to invent.


Let me speak in App.net's defense and say that I'm really enjoying the service. I know that's not really a popular opinion around here, but there it is.

Yes it has a stupid name, yes it has fewer users than twitter/facebook, and yes it's hard to articulate to your non-tech friends why they should spend $36/year on this. But it's becoming a pretty positive place to hang out, and most of the people I care about listening to are there.

In practical terms, this file API means that if you want to post a picture you don't have to use yet another here-today-gone-tomorrow photo sharing service with no revenue model, which I would hope means there's a better chance of your photos sticking around permanently (or at least as permanently as anything ever gets on the web).


> yes it's hard to articulate to your non-tech friends why they should spend $36/year on this.

Actually, I am a tech person, and I like to support App.net. I have not yet seen a reason to use the service.

> ...it's becoming a pretty positive place to hang out, and most of the people I care about listening to are there.

Ok. This is genuinely the first benefit I have heard about using it. Great. Please elaborate more. (I say this as someone looking for reasons to support App.net.)


There's people on App.net that he finds interesting to read, there's nothing to elaborate.

I participate a lot less on App.net, but I also find the community quite enjoyable, it looks like twitter circa 2007, mostly with opinionated tech people around.


App.net has a much higher signal-to-noise ratio in my experience, even among the same set of followers. Something about the culture of app.net seems to cause people to put more thought into what they're saying.


Curious how this any different than uploading your photos to Facebook and using apps which integrate with the Graph API to access your photos?

I get that App.net charges users money but I'm often baffled by that being misconstrued into users somehow having more control and ownership of their content. Specifically, how's that make my content portable?

Disclaimer: I'm the lead dev/founder of OpenPhoto and our claims are similar but we're open source and let users specify what storage service they want to user with our service.


> Curious how this any different than uploading your photos to Facebook...

Technically it's not: app developers are using another service to handle file storage. However, the difference lies in the promise from app.net that they won't arbitrarily shut off your app because they feel like it or because it somehow competes with their own service. Facebook has introduced fear into their developer ecosystem that they can't be trusted to build a popular product on top of. App.net is saying, "you can trust us, we are an infrastructure company..." like Amazon.

> Specifically, how's that make my content portable?

I suppose it doesn't. Any file storage service has some amount of lock-in, but people can create tools to import/export data to and from different services. Unless you have all of your data stored locally on harddrives in your house, you will have to use some service and trust their level of non-evilness.

OpenPhoto offers S3, Dropbox, and local storage out of the box (which is awesome), but even if I want to switch storage providers, I'm going to incur a switching cost (mostly paid with time to export/import).


I'm only mentioning OpenPhoto for philosophical reasons as I think these two services solve entirely different problems.

We support S3, Dropbox, CX, Box.com, DreamObjects and local file system. Two of which were contributed by the community. The point being, this list isn't controlled by any single person or entitiy.

> However, the difference lies in the promise from app.net that they won't arbitrarily shut off your app because they feel like it or because it somehow competes with their own service.

Perhaps I'm in the minority that doesn't trust promises of a company with valuable data (photos, for me). If it's not valuable data then why care at all (my tweets).

> I'm going to incur a switching cost

For hosted accounts we provide a migration option. All your photos are seamlessly moved from Dropbox to S3 and all your links and mobile apps work as they always have. http://i.imgur.com/wWUc5uA.png

That being said, someone should write an App.net file system adapter for OpenPhoto :).


I think the key argument to App.net is not open data but an open business model.


They charge users a yearly subscription. It's the same as any business model. Somehow they've pulled in "openness" and "transparency" into their marketing lingo and it's damn confusing.

I don't dislike App.net. A good friend of mine works there and I'm rooting for them and wish more online services would charge consumers.


The openness and transparency comes from the open business model, which obviates incentives to monetize users through ads and data-mining. It's still theoretically possible that App.net would employ these income models, but their entire platform is based on goodwill, and without it, no one will use it.

I don't buy the holier-than-thou rhetoric employed by Dalton, which also trivializes a lot of the complexities in running companies, but I think the value proposition is fairly clear - for nerds like us at least.

It's not enough to entice me to use the platform, because as nice as it is that the founders are profitable, the wider appeal to a general audience requires an electron microscope to find.


I agree with most of that. Though most services which charge users are disincentivized (is that a word?) to sell user data and display ads.

Besides having the API docs in Github I've never seen how it's much different from other companies that charge users.

Nontheless, I wish them luck because we need more self sustaining services on the web for consumers.


For one, we know they are - comfortably - profitable at the current price per user, so they won't get desperate enough to go in different directions. (They may do it out of stupidity, but perceived financial necessity won't be the reason.)

I'm sure it isn't easy for Twitter to make money, but their reputation is still in tatters, because of all they've done to affirm their position and guarantee what they believe is a healthy profit.

With most companies, we not only have to wonder how they earn or plan on earning money (Instagram! Quora! Facebook!); their business plan may not even suffice.

Tumblr is currently twisting and turning to monetize their platform more, because David Karp has refused to use a business model that is implemented at the expense of users. But good intentions don't guarantee a (sizeable) profit.


Different in that app.net is supposed to be fully accessible and is more "yours" then it is on Facebook.


> is supposed to be

That's my point.


Put it this way: Facebook has a very good reason to sell your data and limit API access; App.net does not. While there's nothing technical stopping them from suddenly targeting ads based on your posts or making sure that you can only use their official apps, they wouldn't make as much revenue as they're currently getting up front (how much does Facebook make per user?). Thus, their best course of action is to simply carry on as they are doing now.


I completely agree with that. I do think aligning users' interest with the company is a good thing. It's just the "marketing" of App.net that makes it feel otherwise.

For example, the heading of this post was "The promise of “unbundling”". It's vague but I think a little bit misleading. But I guess that's the point of marketing :).


As I haven't seen an App.net thread on HN for a while, and i've never really "stumbled" across app.net status updates or mentions, a question to its users: how's it been? Has the community and ecosystem grown? Do you plan to maintain your subscription for the foreseeable future? Do you see value in maintaining a corporate presence on the service?


> how's it been?

Very pleasant. The general tone is really friendly and engaging, and feels much less one-way, producer-consumer than Twitter.

> Has the community and ecosystem grown?

Yes. Not slow but not blitzkrieg either, just steadily.

> Do you plan to maintain your subscription for the foreseeable future?

Yes, definitely.

> Do you see value in maintaining a corporate presence on the service?

Not applicable for me.


Piggybacking on your question: isn't Tent.io a better answer to concerns about Twitter's control of its ecosystem?


It's not going great judging by this list of new users per day.. http://appnetizens.com/nupd

"Turns out App.net currently has approximately 20,000 users, of which a small minority seemingly dominates the conversation: Basch estimates that 250 users (1.25 percent) have so far accounted for half the posts." http://thenextweb.com/socialmedia/2012/09/07/off-slow-start-...

So once every 3-4 days a new active user is born..


Ah, that Appnetizens link is a very interesting one.

30,827 users * $36/year = $1.1 yearly revenue.

It should be interesting when the yearly subscription dates come up in August. 18,255 users are up for renewal in August. Well over half the user base.

Then subscription numbers will stabilize and App.net will need to deal with issues such as user base churn.


To be fair, that TNW link is from September 2012. The Appnetizens link is realtime though.

I'm an active user and thought I see few new accounts overall, it has picked up slowly over the past couple months mostly through referrals.


What makes this different from Dropbox, which already has an API[1] and paid plans?

> In this world, your photos are held in a data store controlled by you. If you want to try out a new service, you can seamlessly login and choose to give permission to that service, and the photos that you have granted access to would be immediately available.

All of this is already possible.

[1]: https://www.dropbox.com/developers/reference/api


This looks like exactly the lines the fellows at Unhosted [1] have been thinking along. They're working with the remoteStorage protocol [2].

If you're interested in this notion of decoupling storage from apps, I encourage you to check these out. Dare I hope that app.net is planning to work with this community?

  [1] https://unhosted.org/
  [2] http://remotestorage.io/


This is interesting. If you made a good API for resumable uploading and had nice oauth and signup procedures as well as apps for mobile ala google drive for accessing the files, I could see some people using this as a file storage backend.


If anyones looking for a quick open source project, we need to update our project to support the new API update: https://github.com/jdolitsky/AppDotNetPHP


If you get support for the file upload APIs you can ping me as we have considered adding App.net support to OpenPhoto (https://github.com/photo/frontend)


File storage? I feel like App.net should be trying for something a little more ambitious. I'm sure this is a very developer friendly solution to file uploading, but I don't really need a new storage service.


the funny thing is twitter or facebook could become a paid service anytime if it sees this as a threat. and people will pay for it


are you kidding ? users will never pay for twitter or facebook. As soon as facebook/twitter is a paid service people will leave it and find for alternatives. For paid APIs that's a different story but nothing is eternal on the internet.


If payment is made mandatory for those apps, it'll kill them. But making payment an option that enables more features or services would be a great business model for them (and less prone to backlash than selling users' privacy or plastering your site with intrusive ads).


The question is - would paying them prevent them from mining your data, or just stop showing you ads?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: