Hacker News new | comments | show | ask | jobs | submit login
App.net’s new direction (marco.org)
138 points by brianwillis on Feb 26, 2013 | hide | past | web | favorite | 80 comments

The first wave of accounts is up for renewal on April 14 — just seven weeks from now. So App.net has seven weeks to convince its early adopters that it’s worth paying more money for another year of service.

I'm pretty sure this is incorrect. Those of us that signed up for the $50/year deal got our subscriptions extended when the price dropped to $36. My renewal date is April 9th 2014 (i.e. more than a year away).

Setting aside that piece of nit picking, Marco's completely right that in order for App.net to be useful to both end users and developers it first needs to be popular. It's a chicken and egg problem in some ways (no users without interesting apps, no interesting apps without a lot of users). Opening the doors with invite-only free accounts goes some of the way towards solving that problem without the scaling issues that would come with unrestricted free accounts.

FYI - He corrected the original post to mark the correct date.

I think that the chicken and egg problem may be a little more fundamental. The major social networks (Twitter/Facebook) first had a compelling application and then later migrated to try to be more like platforms. App.net is trying to be a platform first, without providing a compelling application first. They need at least one killer app to convince people to sign up, and that is what they should be pushing to end-users.

They can push the File API/platform to devs at the same time, but who is their real customer? The end user or the developer? It's hard to figure out.

Who is the real customer on Twitter? Users who follow celebrities or celebrities who post?

Both. It's a marketplace for real-time information, and any marketplace always has both buyers and sellers. This is especially important for Twitter, as each person can frequently swap roles.

Right. And that's why it's not confusing or complicated for App.net to target both users and developers. Both are their "real customers."


Also for those of us who have Twitter accounts our subscriptions weren't extended. It may be a hard sell to renew. We'll see.

Wow, talk about discrimination

Sorry apparently I had Twitter on the mind. I meant Developer accounts. Which I have.

I wish there was more interest in developing on top of the tent.io protocol. Open and distributed should be the goal for the next phase of social networks.

Tent.io introduced a few cool ideas, so I'm glad it's part of the conversation, but it's definitely not the answer.

The developers behind it have a weird preoccupation with preserving all your content forever. For example, if you edit your Tent profile, people can still get all the old versions and see info that you deleted. And last time I checked, not only was this the default behavior, there was no way to opt out of it, either. And they claim this is a feature but nobody I have talked to wants it.

I also followed the Tent software's development for several months (initially I was extremely excited about the project). It broke frequently and progress on making it do what it was supposed to was slow. Plenty of features were promised ("we'll implement X next week") but never delivered even after months.

I don't think the developers really understand the problems social software solves, nor do they have a clear idea what their own project does and doesn't do (they seem to think everything should be a Tent.io app, which is crazy), nor are they particularly good at shipping. I've moved on, personally.

Do you know much about pump.io, which looks like it is the replacement for status.net?

Not familiar, sorry.

What about StatusNet? Just because you are more open and distributed, does not mean you will be more successful.

Part of what makes developing on tent.io a problem is that there seems to be no easy way to find other tent.io users that are not using the tent.is service, i.e, people who want to run their own server.

It seems that it's supposed to be distributed, but there seems to be no easy way to find users that are distributed.

We actually built a tool for that late last year: https://friends.tent.is/login

Currently it's email-based discovery, but other services using OAuth could conceivably be added.

The main issue with Tent is that it will have distinct problems scaling. It's fine with a small network, but as soon as you grow beyond that you run into trouble. In a worst case scenario, every user is on a different server that needs to be notified of an incoming message. If I have 100 followers, that's 100 individual requests to transfer just message. When a heavyweight like Gruber joins the service, that model just isn't sustainable.

Technically, C1M is basically a solved problem and now people are working on C10M. Economically, I like the idea of making social media experts pay their own way.

What's a C1M?

C10K is the problem of handling 10K clients simultaneously, I presume C1M is an extension of that.

[1] http://www.kegel.com/c10k.html

There's no money in that. Didn't you get the memo? The "let's make a new social network based on getting users" is "totally sustainable (TM)".

Also app.net's $100/yr developer account is perhaps not the best way to attract developers to the platform.

Our apps, whether they are dedicated app.net clients like Netbot or just apps that let you post content to a user's app.net account, add to the network effect that can help app.net grow.

Developers should be allowed to build against the platform at the same price as regular people pay to become members in my opinion.

I'll be watching app.net and seeing if they change their stance on this in the near future.

Especially with such a limited API.

App.Net is literally paying their developers money, though. If you build a useful tool, you get money directly from ADN. I don't see why paying for access to the API and, more importantly, support, is a problem.

Apple charges $99/year for their iOS or Mac appstores, which has a far greater reach. So perhaps they've set their price too high.

The $99/year is mostly just away to filter out some of the noise from their review queue and tech support I think. I don't think the revenue from it matters at all to apple.(i'm sure if they thought they could get more quality apps by dropping the subscription, they would. You can still dabble with iOS development without the subscription.

I don't mind paying -some- price and I gladly fork over money for a good product or service. I bought Coda for $99, and MarsEdit for ~$40 and countless other great software.

However, paying $99 for the right just to develop against an API so I can add it as an option for an app I'm making doesn't compute for me.

I'd probably not use the service enough as a regular user to warrant it either.

I think the problem is possibly that most people don't want to pay to get into a silo when there are so many free silos to choose from.

It's certainly why I haven't touched it. I suppose not having ads is nice, but it still seems to have the same fundamental problems that FB/Twitter/G+/et al have.

I really want something like tent.io to succeed (and when my time frees up shortly I'm considering playing around in that space) but I don't quite think anyone's figured out how to make these open frameworks be as user friendly and seemless as these convenient friendly silos.

App.net seems like the answer to a question, just not a question many people are asking.

Which worries me, I have a different business model with a similar problem: getting pay people to pay to use stuff that they can sorta-kinda use for free.

As do I. Well. As did I.

I've pretty much canned the idea at this stage, because I'd love to build this thing, but the three funding options seem to be angel, subscriptions or ads. The first and third make me behooven to someone else; the third would also require (if I wanted to do it right) writing a ad-engine backend and actually hiring people to source ads and crap like that.

Subs would be awesome, but since my app neither makes you money or gets you laid I don't see anyone paying for it.

When I get my stuff cooking, I'll be able to help people just like you.

All I need is zillions of dollars and for a 30 month patent process to run a smidge faster.

I don't understand the problem with Twitter ads.

A simple script gets rid of all Twitter website junk quietly and automatically. No Promoted tweets, no "You may want to follow" nags, no "Now trending" junk. Nothing, but a stream of tweets from those I follow.

I realize that making people pay to use the service is likely to attract a different user base, but the lack of ads doesn't really move a needle a bit.

I just don't think anyone wants yet another social network, with twitter, facebook (and google+ muscling its way in) no one is looking for another way to contact old friends or meet new ones... especially not at $36/year, as we saw with Instagram most people don't care about privacy or ownership, if you have a compelling product as long as it's free you'll get users no matter what your privacy policies are.

App.net is really an experiment to find out if this hypothesis true, or if there can be a viable business model selling to people who do care. Marco doesn't need everybody to sign up if he can reach profitability in the margin between the "most people" who don't care and the handful left over that do.

Great explanation, I guess it's similar to the Orabrush, even though most people don't care they realized that over 20 million would buy such a product,which is a large audience...

I'm so used to the boom or bust model, especially when it comes to SocN that it never occurred to me that they'd be okay with niche

I really wonder what the true costs of running/maintaining a Twitter-clone per user would be. Is it anywhere close to a few dollars per month per user?

To me, it feels like it should be a tiny fraction of that, but perhaps my judgement is off?

I mean, compared to say youtube, you have to deal with short strings of text...

Keep in mind that in a freemium app each paid user needs to pay for ~20 free users. So yeah, it should be on the order of 10 cents/user/month.

That's true.

But if it were cheaper, wouldn't a lot more people be able to sign up too? At $5/month, it's almost what you pay for Netflix. It's probably more than most people pay for their bank accounts.

They might be able to sign up if it was cheaper - but it doesn't mean they would. Cheaper isn't always "enough".

I'm personally really excited to get an invite! I think that invite systems are a great way of doing a controlled rollout while building hype.

I think we're about to see a new generation of apps that take advantage of user-owned file storage. For years, native apps have stored all of their data on hardware users own. Now that everyone basically has some kind of always-on file server, I think we'll be able to do similar architectures in the cloud. Imagine if we built the next Instagram by storing all the photos entirely in the user's Dropbox and just have a heavy caching layer on ec2. I'm working on a platform right now that should make building this kind of app pretty easy.

I think there's room for all of the competitors in the space to benefit from this paradigm shift.

I mean, I was excited to get an invite to google wave. Then I couldn't figure out what to use it for, didn't have any friends using it, etc... which has been my experience with app.net so far as well.

I find it a little ironic that you view dropbox as "user owned file storage." I don't have a dropbox account and don't want one. I'm happy to pay for (and I do pay for) a separate service that gives me webdav file access, but it seems like it's less and less common for programs/apps to support that. (please support that in your platform!)

I feel like I own my storage on Dropbox to about the same extent that I own it on S3. Do you not feel like you own cloud storage? Is it because we're just renters? I'm really quite curious what you find ironic.

A big part of "ownership" for me is the ability to replace the service with my own machine, whether I actually do that or not. So (for example) I'd have less ownership over the gmail account <whomever@gmail.com> than one with a custom domain, even if I chose to forward everything to gmail or use google apps. I haven't looked into it at all, but my understanding is that it would be hard or impossible to point applications to a self-hosted dropbox replacement; compare that to, for example, github where I could replace it if I wanted to fairly easily (at least for the small, personal projects I keep there; this probably becomes less true for larger projects, so take it as an example of my thought process rather than a statement about the "ownership" of data on github).

I just cancelled my APP.net account two weeks ago. I tried the service for over six months. I was paying monthly.

At first, I enjoyed the service. But, I gradually stopped participating. Perhaps it was the lack of new people to follow. I might retry again, if more people use the service.

The best direction for App.net would be to make developer accounts free.

You are building a platform, you want lots of apps. It doesn't make sense to charge developers $100/year. These are the people who are going to innovate on your platform. These are the people who will make your platform more visible by integrating it into various services.

The company I work for has Facebook / Twitter sign up. I use app.net and I would add a signup option to our website if I did not have to pay $100/yr to do it. Not when perhaps a handful of people will use it and I will see no return on this $64 additional investment.

I agree with this. I should not have to pay $100 to have the right to develop on their platform.

I've been pinging users on buddycloud, about why they signed up. Main reasons: they don't want another silo / want data portability and they like the idea of beign outside a company controlling their social networking experience

Of couse I'm biased, but permit me to dream: it would be great if App.net would consider building on an open and distributed/federated design like tent, status.net or buddycloud. The App.net guys seem great at design and building an ecosystem and the syngergies would be a really interesting alternative to Twitter/FB.

I would like an invitation too, but rather than have a hundred comments like this I made a google doc to consolidate them:


Just post your email there if you want one, then delete someone's email if you sent them an invite.

(Mods - this strikes me as more organized, but you can remove it if you don't approve)

Thanks in advance!

I like everyone else was curious when App.net was first announcing this pay-for-Twitter clone awhile back. But I admit I didn't really "get it;" I thought the idea of a Twitter that's not Twitter was possibly a very interesting idea. But I always thought the idea of it being more a platform seemed to not make much sense.

Now, I have to say, I really don't get it at all anymore. What it is thing about now?

I actually thought alpha, the name chosen for the app you see on app.net website, an alpha release of app.net, a Twitter clone. And I thought app.net was a bad name for a Twitter clone, until I realize they are trying to build an app platform.

I had a really hard time figuring out their value proposition, too. In fact I'm still not exactly sure what it is. I've given up and moved on.

I'd love an invite if anyone has an extra! Email me if you can!

In exchange, I'll buy you a beer (or coffee) if you are ever in LA. Or if you are in SF I will be there next weekend, and I'll be you a beer (or coffee) then if you'd like!

Sent. Give the coffee to a homeless guy.

Thanks! Will do :)

Edit: Never mind, got one.

Does anyone have a rough idea how many users are currently on app.net? Considering the audience is primarily developers and technologist I thought there would be some applications (particularly in the startup community) that could use it as an effective use acquisition channel. But it really depends on how many users there are, 5,000? 50,000? 100,000? more? I'd be interested to know.

There are over 40k users as of today.

I think app.net should be working on getting actual implementations of their file API out there (Instagram, every photo editing app, etc), yet that's another chicken-and-egg problem: what company will encourage their own users to rely on an unknown third-party service?

This sounds like a problem that can be fixed by throwing lots of money at it, but that doesn't fit their business model...

yeah no point! at this point its kinda mute... to be useful you need to have your friends in it! and the way its setup its not easy or user friendly to have people join, something like gmail gained traction with invite only because gmail is useful by itself you dont need anyone else using it to make it more useful, at this point it really doesnt matter if its dev friendly... twitter might suck right now for any dev, but app.net limited scope puts it in competition with way more tools than just twitter and facebook's of the world, if you are this limited, you might prefer to go use something like tumblr or do it yourself with your own hosting, why use app.net?

Does anyone else have an extra invite for this platform? I'm unfortunately a little late to the party but I'd love to check this thing out! Thanks in advance, I'll do what I can in exchange to help you back.

Email: 6231.32@gmail.com

I'm also interested in an invite (saas(at)omarabid.com)



sent (your providing an email address made it quite easy...)

May I have an invite as well? my email: danprime@gmail.com


There you go!

If they want users, why not more payment methods?

I really don't want to enter my credit card information on the page of a startup.

But now it seems to be a little too late. Good night, sweet service.

App.net was an interesting attempt at taking on Twitter, but failed.

However there is a real challenger waiting in the wings. Its name is Buffer.


If anyone has an extra invite, can you please send me one at duncallmeeel@gmail.com. Thank you~

So I don't get it. The crowd funded this project and the crowd has to pay to use it?

The initial funding came from people buying a year's subscription. Once the initial subscription is over, they will have to choose whether or not to continue paying.

maybe i'm just not `getting it`, but what benefits does the app.net platform offer _exactly_ for developers? I could be wrong, but usually when services go from paid to free, its a sign.

Switch to Buddycloud the true free social networking platform http://buddycloud.com/

can anyone invite me hskoder@gmail.com will be highly obliged

Allright , i'm a user , what does App.net have that twitter or facebook dont ? why should i use it ? instead of non-paying social networks?

I feel like this gets gone over every time App.net is brought up. The value proposition is that rather than trading your privacy and your attention for a free service you'd rather pay and retain ownership. For developers, the value proposition is that you have a service you can treat more like a public utility than like a private company's secret API that they could revoke access from on a whim. Eventually, if the second part works out, there will be applications that only work on App.net and they will also become part of the value proposition for end users.

>The value proposition is that rather than trading your privacy and your attention for a free service you'd rather pay and retain ownership.

If that's the value proposition, than it's a terrible one. There is nothing inherently different about App.net that would protect your privacy or guarantee you ownership (of your tweets?). They can try and argue that position while they are tiny and inconsequential but it will ring hollow once (if) they get big.

I think there's potential in a service like App.net, especially if they cater to their audience (developers and start up culture in general) and are willing to go for a steady organic growth. Developers are a great group to target because they are fashionistas and they love to tie themselves to services that increase their 'street cred'. Now that GitHub is mainstream, the time is right for 'the next thing'. If App.net tries to be Twitter with a better EULA they will fail.

I've been sitting on the sidelines and looking at App.net, and never pulled the trigger on a subscription, but I may check them out now with an invite. I really hope they do well. There's potential.

> There is nothing inherently different about App.net that would protect your privacy or guarantee you ownership

The concept here is that by making their money from the users directly they align the users' interests with their own.

> it will ring hollow once (if) they get big

I suppose it's possible that they could decide they're not making enough money and convert to an ad revenue model. The problem is that there is already an ad supported service which is quite similar and free, so if they did that they would essentially lose all their users instantly. They'd also be violating their own core beliefs. I think they're a lot more likely to shut down the service from lack of users than they are to "sell out" in this fashion. I mean, there's nothing stopping Microsoft from shutting down the Windows division and focusing on Office for Mac, except that it would completely undermine their brand, and they make significant money from the Windows side. It's the same situation here: switching to an ad model undermines their brand and loses them their paying customers.

Besides that, Marco's goal is to achieve profitability with a small userbase. I don't doubt that it's possible to do that, and it undermines the "they'll screw you over when they get big" concept because they are not, like most startups, trying to reach ridiculous scale before figuring out how to make money, which seems to leave the ad model as the only possible option.

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