Why We are Moving Away from Facebook as a Platform | Hacker News Reader

Why We are Moving Away from Facebook as a Platform

blog.zuupy.com

105 pointsfezzl45 comments

Facebook seems ripe for disruption, or at the very least it shouldn't be that hard to compete with it as it was a while ago. Former early adopters and addicted users are getting tired of it and want something new, better, less spammy.

It's also in the nature of early adopters to move on by the time a product or service becomes too mainstream. They just need to see something else that is compelling first. But I think many of them are already using it much less than they used to.

I think a true distributed P2P social network would be very interesting for them, but it would also need to get the execution right. I haven't heard much about Diaspora lately, but something interesting came out a few days ago with BitTorrent trying to make a P2P social network around torrents. I think that's an intriguing idea, and it's a bit similar to what Opera tried to do with Unite.

As a side note, I think all social networks so far have had one major design flaw. I've joined other social networks before Facebook and they all seem to suffer from the "too many friends" problem, which does become a problem once you start having hundreds or thousands of friends.

I think there needs to be 2 categories of "friends": true Friends and Followers. When you accept a request it should make that person a Follower by default, but it should give you the option to make him a Friend also. So many people add others on Facebook now, when it's just so they can see more info about them or their private pictures, with no real intention of "befriending" them later on. That's why I think most requests should become followers by default, and only choose to make them "Friends" when you really know the person.

I think Twitter's model works a bit like that. You can have thousands of followers, but only a few "friends" where you basically follow each other. Just something to think about for whoever wants to build the next social network or whatever, so they don't repeat this huge mistake that almost every social network has made so far.

In the real world, someone who follows you that is not a friend is a stalker. I suggest that the "too many friends" problem might better be described as the "too many stalkers" problem. How many is too many? One.

The real problem is that we want to map our existing relationships to online tools without redefining them according to some nebulous model. Nobody has offered that capability yet. Friendship is intimacy. How can you achieve that in an environment where anyone can invade your privacy to some degree?

No one want's to be a second tier friend.

But it seems to me most people have up to 10 close friends, up to 100 sort of friends (the colleague you like and talk to daily, but wouldn't ask about your sexual problems), and then a large number of people they want to follow or not fall completely out of touch with (second tier friends who you no longer work with, but might ask a favor at some point).

I've found that people get upset if you don't ad them on facebook. It marks them as unimportant to you. I doubt they'd like to be categorized as a second tier friend, or demoted to third tier.

I'd say rather than creating categories, it would be better to allow users to make groups of friends, and select which group gets which news. That way you could say "Oh I'm sorry I didn't invite you to my party, I had you down under work people"...

Are most people really so insecure that they would be bothered by not being a "top tier" friend to everyone they have ever known? Surely you can name a person with whom you have a cordial acquaintance, but are quite obviously not in their circle of "close" friends. I my case that's the big majority of people I know. So what?

Well it's fine if it's a symmetrical both of you think each other as same tier; less so if you'd disagree (e.g. some bosses think all their subordinates are their friends and would want them at their parties).

www.dreamwidth has built a two-tier system.

You can 'subscribe' to someone, in which case you are reading their public posts. They can subscribe to your public posts too.

And, you can 'grant access' to someone, and now they can read your private posts.

There is also an access control model for posts.

I think that renaming the relationship from 'friend' is a good thing. 'friend' is getting semantically overloaded.

A good social network needs to build real life grouping like school friends, office colleagues, relatives, family etc. Sharing of photos, updates and read permissions by group needs to be built in. But the big question is can they compete at this time with FB. How many people will switch just for this reason.

...and acquaintances, enemies, people you're mad at, ex-whatevers, businesses you refuse to do business with.

And then when we start to admin all these relationships online, we will realize, "Why don't I just do this in-real-life?"

also would be nice to have overlapping tags instead of discrete lists. You could have your startup people / people interested in ${tech} / people you'd invite to your birthday / people you'd share photos with.

Also, one thing I find annoying in Facebook newsfeed noise is for multilingual friends getting their non-English updates. Language groupings would be nice (or auto-tagging based on language detection (or autotranslations)).

I like the tag idea. You could more easily group people under tags compared to lists, and when you'd post something you would just click on a few tags. It would at least be better than what they have now.

Having a hierarchy of friends may be in your interest but it is certainly not in the interests of the people hosting the social network. They need the social graph to be single level either to make mining the data easier or to get to expose your information to as wide an audience as possible.

This is my main gripe with Facebook, the lack of proper entity modeling.

Use Facebook and your only option is to consider somebody a "friend" entity.

Modeling relationships with only a friend entity is pretty poor, IMHO.

A nice idea, but I think it wouldn't work. I think people would be reluctant to add people to their second tier, and you'd see "friend inflation" where everyone was on the top tier.

The 'second tier' people don't need to be told what tier they're on, necessarily. There does need to be more thought put in to how relationships are categorized and info flow from those relationships displayed and managed. But all of that can be on one party's end - the other party just needs to know that I'm aware of and acknowledge the relationship - they don't need to know what I think of it.

Every single story about facebook has a post like this, and I'm so tired of reading them.

First of all, why would you even consider competing with Facebook? You do realize that outside of the valley, there are companies not packed to the gills with ruthless nerd geniuses that would be much, much easier to disrupt? Facebook isn't even ossified like Google or Microsoft. They'd just reimplement your killer feature in one of their hackathons.

P2P social network won't work because it's way too hard to do any analysis on the data of the system as a whole. How would you implement a newsfeed? It would be a nightmare.

Regarding having two categories of friends, it's already doable with friends lists. You can do exactly what you are saying with a friends list for acquaintances/grannies/whatever. Or you could just not accept so many friend requests.

How about reason 0: You depend on a platform that can shut you down in an instant without any justification and without possibilities to reactivate your app?

this argument comes up over and over when talking about facebook APIs etc, but this is an inherent risk in all kinds of online and offline applications: from the electric power company to the ISPs, to browser vendors, to facebook, there are all kinds of dependencies you have to rely on. True, facebook is more likely to have conflicts of interest with your web app. But the main issue is that facebook doesn't yet have a financial interest to maintain a healthy app ecosystem.

Electric companies, ISPs and browser vendors have competition. When Facebook is a dependency, it becomes one without easy alternatives.

To be fair, facebook platform was a mess even when they had competition from myspace.

I can sign contracts with electric companies and pay for SLAs from ISPs.

If browser vendors simpy start disabling browsers, we have much bigger problems.

Facebook's interest is in policing apps, and acquiring/crushing/copying the relevant ones.

> If browser vendors simpy start disabling browsers, we have much bigger problems.

There is enough competition in the browser market that selectively disabling certain sites will lead to a big loss in market share, and would be an exceptionally stupid move.

Crushing apps is not a big problem (Zynga still earns more than facebook itself). The problem is we cannot sign an SLA, or have any other contractual relationship with facebook, even though they take a 30% cut or ouf virtual currency sales.

Zynga still earns more than facebook itself

Citation please? I think this was true in 2009, but hasn't been the case for quite a while.

You 're right wrt earnings (zynga had $0.85B vs facebook $2B in 2010). Zynga will probably earn $1.8B this year (http://blogs.forbes.com/afontevecchia/2011/03/02/zynga-revea...). I can't edit the previous comment, though.

Still, for a company that is essentially a parasite on facebook, the money they earn is staggering. They could even earn a lot more if they used advertising on their games (they stopped using ads in 2010).

"even though Facebook actually has 600 million active users" <-- does facebook have 600 million actual active users or 600 million accounts?

Active users

I'm not sure if I can believe it since websites tend to inflate their own numbers to look bigger. Is there any independent verification of such a number?

They described it once but it's definitively active accounts. I think it's something like any account that logs in at least once a week.

The much more interesting number is Twitter. Only 15 millions Twitter accounts are active, out of 75 millions.

src: http://thenextweb.com/socialmedia/2010/01/26/15-million-acti...

I think it was more like a month.

Former Facebook intern. The number they track is number of accounts with activity in the past 30 days. Logging in to Facebook or performing some other action associated with your account is sufficient to trigger this.

So ... I ask not being a FB user myself: if I have some kind of facebook "toolbar" (I assume there is one?) installed in my browser, and it's configured to automatically log in, then I am an "active" user even though I may not be interacting with FB in any other way?

When you say "performing some other action associated with your account" does that mean:

1) you can perform actions associated with your account without having been logged in? If so, could you give an example.

2) someone else can perform an action, and that would mean you would be considered active even if you hadn't logged into Facebook in over 6 months. If so, could you give an example?

I interpreted 'logging in or performing some other action' as 'we do not want to count users as active when they stop using Facebook without logging out'

Is just visiting a webpage with Facebook connect or social plugins enough to count as being "active"? Or do you need to visit facebook.com itself?

That depends on how you define 'active'. A quick Google gives me some Facebook published statistics [1] which say that more than 250 million people log on every day. Goldman Sachs seem [2] to say "600 million+ monthly active users".

[1] http://www.facebook.com/press/info.php?statistics [2] http://www.businessinsider.com/facebook-has-more-than-600-mi...

For Facebook "active users" always means "users active within the last 30 days."

Reason 3 is why I am constantly advising friends not to take "promote your business on Facebook!" classes, especially when there are other, more direct avenues available to them. I've come to the conclusion that social and marketing are an awkward mix, and most people route around attempts to inject the latter into the former.

On the other hand, social media is a great way for your customers to tell their friends if they've had an exceptional experience with you. So I believe making it easy for your customers to talk about you is far more valuable than building on Facebook itself.

There are many more problems with FaceBook IMHO. That doesn't mean you should move away from it simply just use it to your advantage.

(Shameless plug) http://000fff.org/how-to-think-like-facebook-and-twitter/

Hold on a sec - replacing facebook with something else is not that simple and i'd venture probably unlikely. there's a huge network effect that facebook has in place. it is not just for games. it is starting to drive traffic to various news outlets: (http://www.pcmag.com/article2/0,2817,2385095,00.asp) and it is probably just getting started. There is something substantial to facebook that makes people come back. And naturally as a company is growing and adding more features, the API is going to change. I personally use it a lot to share news articles with my "friends", but the ability to now share those articles with only certain friends, directly from the news website - which hasn't been implemented everywhere yet, is pretty interesting and shows that facebook realizes that there are friends and there are "friends".

It would be far easier to have disrupted something like Google, which doesn't have the same network effect as Facebook - but even that has been very challenging for anyone to do.

So you can use it like Twitter?

Their platform changes are mostly about reinventing the wheel. They dont even introduce important features, instead they strip them ( profile boxes, notifications, profile tabs etc)

I agree. The Facebook API is a mess. It is too costly for developers to chase the changes and the API itself is too restrictive. It's like AppleScript, a great idea, pity you cannot do much with it.

Facebook platform is only good for games. No other type of app can get traction there, because facebook is an entertainment medium , no matter how hard they 've tried to become more "serious". The thing is, it has gotten increasingly off-putting for indie game developers to follow facebook's labyrinthine platform roadmap. Their platform designs and updates are notoriously amateurish and short-sighted. I believe they have changed 3 platform leads, each of whom introduced a new API. Their weekly code pushes regularly introduce bugs along with the fixes.

For us, it's taking more time to adjust to facebook's everchanging APIs, policies, designs etc, than adding features to our games. The fact that facebook nonsensically banned Adsense advertising does not help either.

That said, facebook is still the most efficient amplifier that can provide exposure to web apps.

I would personally vouch for another provider that can offer a stable social gaming platform. There were rumors about a google game network, and i m hopeful about it

how can someone find only three?

Requests per minute!

Reply on news.ycombinator.com