Ask HN: What will be Facebook's weakness?(twitter.com)
twitter.com
Ask HN: What will be Facebook's weakness?
http://twitter.com/yegg/status/24471492383
37 comments
Location. I thought they'd be perfect for it, but I think their implementation shows that they just don't quite get it. It doesn't matter to me that someone was at Wawa 15 hours ago.
My hunch is that they've just scratched the surface of their plans. If not, I am similarly disappointed.
What are Zynga's weaknesses? Every one of them hits Facebook, because of the huge proportion of revenue FB gets from Zynga (and businesses which are Zynga-with-worse-execution).
How about 'Reliability'. Their API is constantly changing in what seems like random ways, and after updates there are always a couple weeks where everything is incredibly flaky.
Thanks for posting this. I'm having all sorts of problems right now (today) with code that worked yesterday suddenly not. And nobody answers their bug posts.
Agree. I'm doing a lot of Facebook app development lately and the API is really a mess at the moment. The documentation is terrible (and they shut down their only hope: the wiki), the API is confusing and inconsistently implemented, and they have a lot of downtime due to various bugs.
It has intermittent latency too. The new "like" button doesn't work right on FireFox either.
Off-topic: Are you the guy who made the IAMA on Reddit about your successful website? I was just wondering how you're doing and if you're still working on startups? :)
Probably a small gap, just big enough to fly a fighter space ship through. At least that's what I find...
Erosion of trust due to lack of user control is the only real issue I see outside of guesses like "people will get tired of it".
The way Facebook connects data and people is very clever. There is lot going on behind the scenes that you never see as a user. One of the results is that you just don't know who will see what. This results in a lack of perceived control.
Here's an example. At a friends wedding, I got introduced to a friend of a friend that I hadn't met before. She knew all about my child and has lots of pictures of him. How? My wife sends picture on her cell phone to her sister. She, in turn, uploads those to Facebook and tags them with my name. That opens up those pictures to her complete circle of friends. One of her friends comments on a picture, and another group of people gets exposed. The result is that I can only assume that hundreds or thousands of random people know everything about my child.
The result is also that Facebook is building up a profile of me without my knowledge or participation.
That's why I prefer Twitter. They don't try to outsmart me. I know who will see what. It's obvious that everything is public.
The way Facebook connects data and people is very clever. There is lot going on behind the scenes that you never see as a user. One of the results is that you just don't know who will see what. This results in a lack of perceived control.
Here's an example. At a friends wedding, I got introduced to a friend of a friend that I hadn't met before. She knew all about my child and has lots of pictures of him. How? My wife sends picture on her cell phone to her sister. She, in turn, uploads those to Facebook and tags them with my name. That opens up those pictures to her complete circle of friends. One of her friends comments on a picture, and another group of people gets exposed. The result is that I can only assume that hundreds or thousands of random people know everything about my child.
The result is also that Facebook is building up a profile of me without my knowledge or participation.
That's why I prefer Twitter. They don't try to outsmart me. I know who will see what. It's obvious that everything is public.
Being a mile wide and an inch deep in too many functional areas.
Facebook's "Open Graph" is in reality a "closed" Semantic Web implementation. It is named so to precisely obscure this fact. Logically, Facebook's downfall will be an open Semantic Web.
That can only be a downfall if the FB "Open Graph" becomes the key to Facebook's long-term success, - it isn't, yet.
If the connections isn't the key, what is?
But haven't we been hearing that about the Semantic Web before even Facebook came on the scene? The number of interesting semantic tools coming to market seems to be decreasing rather than increasing.
Contextual Sharing (and the new social network that figures it out) will kill Facebook.
I'm convinced that contextual sharing is the next evolution of social networks, and I don't think FB is in a position to pivot towards it.
Contextual sharing is the idea that in real life you have multiple, often separate social circles (work pals, college roommates, siblings, friends, grandparents, frat brothers, etc.), and that social networking 1.0 (i.e. facebook/twitter) doesn't account for this very well. You really want to share stuff only with specific groups, but this is hard to do on existing sites.
This leads to all sorts of real pain points, many of which are becoming unbearable for some users: i.e. you don't want your boss seeing photos of you drinking, or you don't want your grandmother to see that racy link you just posted, so you suspend your FB account.
Even though Facebook has a crude groups feature, this doesn't come close to addressing the fundamentally flawed orientation of the network: one, big pot of 100 to 1000 casual acquaintances, as opposed to dozens of partitioned ~2 to 10-person real social circles.
This idea has been exploding, recently:
- Google Me (Google's soon-to-be-released last shot at social) is almost certainly going to be based around this concept of contextual sharing (as Google UX engineer @padday's recent slide show indicated)
- Diaspora (the OSS project from NYU students that raised $200k this spring) announced on their blog that contextual sharing is their #1 interface priority
- Frid.ge (YC '10) is built around this idea
- College Only (funded by Peter Thiel) is a solution to the problem just for one context (college friends)
- Groupme.com (betaworks) is an SMS-only approach
And ALL of this projects came to life within the past 3 to 6 months!
Next-generation social networks will mirror our offline social experience more closely by shifting the focus from a giant network of "friends" to private, micro-networks that mimic real friend groups.
I think this change is significant enough that Facebook will be unable to pivot without alienating their huge user base.
Also, one of Facebook's main competitive advantages has been the fact that it is a network good with high switching costs (e.g. in order for a new site to be as good as facebook, you need to convince all of your 800 friends to sign up for the new site too).
If the future of social networking really is contextual sharing, though, this barrier becomes much lower (if it continues to exist at all), because if you really just want to share with your 'real' friends, then convincing these small groups whom you know really well (your best buds, or your roommates, or your sibilings, or whatever) to try out a new site is actually pretty trivial.
Thoughts?
Full-disclosure: I'm thinking of building a contextual sharing tool of my own, because I think it's so damn exciting. Would love your feedback.
I'm convinced that contextual sharing is the next evolution of social networks, and I don't think FB is in a position to pivot towards it.
Contextual sharing is the idea that in real life you have multiple, often separate social circles (work pals, college roommates, siblings, friends, grandparents, frat brothers, etc.), and that social networking 1.0 (i.e. facebook/twitter) doesn't account for this very well. You really want to share stuff only with specific groups, but this is hard to do on existing sites.
This leads to all sorts of real pain points, many of which are becoming unbearable for some users: i.e. you don't want your boss seeing photos of you drinking, or you don't want your grandmother to see that racy link you just posted, so you suspend your FB account.
Even though Facebook has a crude groups feature, this doesn't come close to addressing the fundamentally flawed orientation of the network: one, big pot of 100 to 1000 casual acquaintances, as opposed to dozens of partitioned ~2 to 10-person real social circles.
This idea has been exploding, recently:
- Google Me (Google's soon-to-be-released last shot at social) is almost certainly going to be based around this concept of contextual sharing (as Google UX engineer @padday's recent slide show indicated)
- Diaspora (the OSS project from NYU students that raised $200k this spring) announced on their blog that contextual sharing is their #1 interface priority
- Frid.ge (YC '10) is built around this idea
- College Only (funded by Peter Thiel) is a solution to the problem just for one context (college friends)
- Groupme.com (betaworks) is an SMS-only approach
And ALL of this projects came to life within the past 3 to 6 months!
Next-generation social networks will mirror our offline social experience more closely by shifting the focus from a giant network of "friends" to private, micro-networks that mimic real friend groups.
I think this change is significant enough that Facebook will be unable to pivot without alienating their huge user base.
Also, one of Facebook's main competitive advantages has been the fact that it is a network good with high switching costs (e.g. in order for a new site to be as good as facebook, you need to convince all of your 800 friends to sign up for the new site too).
If the future of social networking really is contextual sharing, though, this barrier becomes much lower (if it continues to exist at all), because if you really just want to share with your 'real' friends, then convincing these small groups whom you know really well (your best buds, or your roommates, or your sibilings, or whatever) to try out a new site is actually pretty trivial.
Thoughts?
Full-disclosure: I'm thinking of building a contextual sharing tool of my own, because I think it's so damn exciting. Would love your feedback.
My sub-optimal solution to this problem has been to use Facebook for personal stuff and Twitter for professional stuff. The problem you mention about your boss seeing drinking photos or your grandma seeing racy links is easy to avoid with Facebook today, but maybe not for users who don't know about the functionality. Put your boss in a friends list that can't access your photos. Click the lock button when you're posting the link and exclude your family members friends list.
My problem is that I want almost everything to be visible to everyone, but I don't want to publish everything to everyone's news feeds and pollute them. Everyone has hobbies, interests or professions they enjoy talking about, but they don't want to burden all of their friends with those posts. If you can fix that without making me drag all of my friends to another social network, I'll use your tool in a heartbeat. I don't think it's possible for anyone but Facebook to meet those restrictions, since only they could change the way posting on Facebook works.
My problem is that I want almost everything to be visible to everyone, but I don't want to publish everything to everyone's news feeds and pollute them. Everyone has hobbies, interests or professions they enjoy talking about, but they don't want to burden all of their friends with those posts. If you can fix that without making me drag all of my friends to another social network, I'll use your tool in a heartbeat. I don't think it's possible for anyone but Facebook to meet those restrictions, since only they could change the way posting on Facebook works.
I feel you.
I think the "drinking photos" example is illustrative because it's extreme, but I think the real, widespread pain points --- I know the ones I have, at least --- are more along the lines you mentioned.
Sure, you can put up certain walls relatively easy (i.e. only these friends can see my pictures).
But I think you hit on the essence of contextual sharing, which is that privacy aside, you simply want to share different things with different people.
I want to share the Paul Graham essay with only my start-up friends, I want to share the new NoSQL db github link with only my hacker friends, I want to talk smack about fantasy football with only those people, I want to post my family vacation pictures to only my family, etc.
I think the "drinking photos" example is illustrative because it's extreme, but I think the real, widespread pain points --- I know the ones I have, at least --- are more along the lines you mentioned.
Sure, you can put up certain walls relatively easy (i.e. only these friends can see my pictures).
But I think you hit on the essence of contextual sharing, which is that privacy aside, you simply want to share different things with different people.
I want to share the Paul Graham essay with only my start-up friends, I want to share the new NoSQL db github link with only my hacker friends, I want to talk smack about fantasy football with only those people, I want to post my family vacation pictures to only my family, etc.
I think Diaspora and others will fail because it is tackling the problem from the wrong angle.
I don't think people really want to micromanage things like that. Think about it. Every time you post a tweet, picture, blog post etc, do you want to meticulously decide who receives it, or would you rather click "all" and let everyone else decide whether they read/view it? Conceivably all facebook needs to do is tweak how you recieve messages a bit and these startups are toast.
You can't defeat Facebook by adding one little feature Facebook doesn't have. You need to come up with a completely paradigm-shifting approach.
I don't think people really want to micromanage things like that. Think about it. Every time you post a tweet, picture, blog post etc, do you want to meticulously decide who receives it, or would you rather click "all" and let everyone else decide whether they read/view it? Conceivably all facebook needs to do is tweak how you recieve messages a bit and these startups are toast.
You can't defeat Facebook by adding one little feature Facebook doesn't have. You need to come up with a completely paradigm-shifting approach.
I agree with you that people don't want to micromanage (the means) but I do think people want to segment their social sharing (the ends).
I think the trick will be to figure out an interface where you can share this way, without it feeling like a pain/micromanaging, etc., and I think this will require a paradigm shift.
Any solution that is, essentially, a "facebook tweak" (whether it's by facebook themselves or diaspora, or whoever) will always be cumbersome.
I think to solve this problem a new paradigm is required, one that replaces the ultimately inaccurate social metaphors of facebook (friend requests, wall posts, news feed, status updates, private inbox messages), with a totally new approach that comes at these problems in a different way (for instance, a unified feed with messages organized by group, as a replacement to all of the features listed above).
I think the trick will be to figure out an interface where you can share this way, without it feeling like a pain/micromanaging, etc., and I think this will require a paradigm shift.
Any solution that is, essentially, a "facebook tweak" (whether it's by facebook themselves or diaspora, or whoever) will always be cumbersome.
I think to solve this problem a new paradigm is required, one that replaces the ultimately inaccurate social metaphors of facebook (friend requests, wall posts, news feed, status updates, private inbox messages), with a totally new approach that comes at these problems in a different way (for instance, a unified feed with messages organized by group, as a replacement to all of the features listed above).
Orkut has already that feature in recent roll-out
See this video: http://www.youtube.com/watch?v=R1xzJrPn-Bc
See this video: http://www.youtube.com/watch?v=R1xzJrPn-Bc
Previous good discussion on HN: http://news.ycombinator.com/item?id=1603374
What about Search/Discovery? Sure, Facebook's Open Graph is being cultivated by all of us as we like and share stuff, but currently, there's no way to search through my feeds, what I've liked, what my friends have liked, the most popular liked items, and any slicing/dicing of that data. I'm sure they're working on it, but it seems like a major gap right now.
The life cycle of any large company: they'll grow too large to innovate, and ambitious folks will leave for greener pastures.
I think the answer is "Business/Enterprise"
Facebook is a brilliant consumer web company, but they never looked at enterprise market. Enterprise market is bigger (but less sexy) than consumer market. So it is possible to build a new technology powerhouse from enterprise revenue and eventually beat Facebook for the next generation of talent.
Facebook is a brilliant consumer web company, but they never looked at enterprise market. Enterprise market is bigger (but less sexy) than consumer market. So it is possible to build a new technology powerhouse from enterprise revenue and eventually beat Facebook for the next generation of talent.
Fashion... It could be uncool, people could just become bored with it or something better comes up.
Look at what happened to myspace
Look at what happened to myspace
It could be a number of things.
Right now, there's a danger that people start thinking of Facebook as the company that's good for status updates and photos. That might prove to be a long term facet of our lives, but it could also prove to be fleeting. Another example of this could be users thinking of Google as that search company and ignoring the company's attempts at other products. There's lots of ways to take social software, but there's a potential that people will popularly think of Facebook as the status message and friend-tagged photo company.
Facebook's barrier to entry (its closed social graph) is both its strongest asset and its greatest weakness. A barrier to entry like this means that most of your competition is gone. You can get used to having very limited, if any, competition because if someone's 20% or 50% better than you are, the barrier to entry slaps them down and maybe you even buy them and incorporate them into you on the cheap. But then someone comes along who is a genius and changes things and shoots the moon. A bit of a comparison here could be Microsoft. Apple was weak for a long time. The Win32 API was a huge barrier to entry. Then a lot of forces conspired to really break Microsoft. The rise of the platform-independent web (with Google, Apple, Mozilla, open-source, and others), Apple's resurgence (where they took Microsoft by surprise on handhelds and might have even sent Windows Mobile to the annals of history), and just the changing landscape. I mean, I remember a time when I thought my Mac would always feel like an also-ran platform unless Microsoft was forced to open up the Win32 API. Nowadays, I don't think any of us Mac users feel like we're on an also-ran platform (at least not in the way it felt years ago).
People's tastes might change. Right now, people want a connected, low-privacy social graph. In 10 years, we might feel different. People's preferences change.
Facebook might become generationally limited. A lot of us got on Facebook in college. I remember when you could list your courses and find people in your classes. Some non-Facebook company might hit critical mass in a different generation of users who might see Facebook as lame because their parents are on it.
Those are a few that came to my mind. I'm not saying that Facebook can't rise to the challenges. They've shown themselves to be quite an adept company. However, it's hard to predict the future and social networking is something that everyone seems to be gunning for. Others are trying to out-maneuver Facebook and users can be fickle.
Right now, there's a danger that people start thinking of Facebook as the company that's good for status updates and photos. That might prove to be a long term facet of our lives, but it could also prove to be fleeting. Another example of this could be users thinking of Google as that search company and ignoring the company's attempts at other products. There's lots of ways to take social software, but there's a potential that people will popularly think of Facebook as the status message and friend-tagged photo company.
Facebook's barrier to entry (its closed social graph) is both its strongest asset and its greatest weakness. A barrier to entry like this means that most of your competition is gone. You can get used to having very limited, if any, competition because if someone's 20% or 50% better than you are, the barrier to entry slaps them down and maybe you even buy them and incorporate them into you on the cheap. But then someone comes along who is a genius and changes things and shoots the moon. A bit of a comparison here could be Microsoft. Apple was weak for a long time. The Win32 API was a huge barrier to entry. Then a lot of forces conspired to really break Microsoft. The rise of the platform-independent web (with Google, Apple, Mozilla, open-source, and others), Apple's resurgence (where they took Microsoft by surprise on handhelds and might have even sent Windows Mobile to the annals of history), and just the changing landscape. I mean, I remember a time when I thought my Mac would always feel like an also-ran platform unless Microsoft was forced to open up the Win32 API. Nowadays, I don't think any of us Mac users feel like we're on an also-ran platform (at least not in the way it felt years ago).
People's tastes might change. Right now, people want a connected, low-privacy social graph. In 10 years, we might feel different. People's preferences change.
Facebook might become generationally limited. A lot of us got on Facebook in college. I remember when you could list your courses and find people in your classes. Some non-Facebook company might hit critical mass in a different generation of users who might see Facebook as lame because their parents are on it.
Those are a few that came to my mind. I'm not saying that Facebook can't rise to the challenges. They've shown themselves to be quite an adept company. However, it's hard to predict the future and social networking is something that everyone seems to be gunning for. Others are trying to out-maneuver Facebook and users can be fickle.
> Facebook's barrier to entry (its closed social graph)...
Huh? http://graph.facebook.com
Huh? http://graph.facebook.com
The data is squarely under Facebook's control. They are allowing other people to buy into their platform in the way Microsoft allowed third parties to develop for Windows. Microsoft gave developers the ability to make programs that ran on Windows with the Win32 APIs. They didn't open things up so that developers could write their own Windows-compatible operating system and compete with them.
It's still closed in a way that they hope manages and minimizes the competition. "Don't build your own social graph. Use our's which we'll give you a little access to." It is very closed in that you can't friend people across services and it all has to be Facebook mediated. It's a strategy to maintain their control over the barrier to entry and extend their barrier to many applications developed by third parties.
Being open means that you're on the same footing as everyone else.
It's still closed in a way that they hope manages and minimizes the competition. "Don't build your own social graph. Use our's which we'll give you a little access to." It is very closed in that you can't friend people across services and it all has to be Facebook mediated. It's a strategy to maintain their control over the barrier to entry and extend their barrier to many applications developed by third parties.
Being open means that you're on the same footing as everyone else.
Windows is a bad analogy. There's no data involved. Facebook's data is open. It isn't federated like XMPP, which seems to be what you're requesting. Federation would be a handicap for a product that changes as much and as quickly as Facebook has over the years. Instant messaging is simple in comparison.
Time. Eventually, someone will bring something to market which will reset all of the rules. Like those who came before Facebook, they will ignore it because they are "Facebook".
Privacy
To quote Yogi Berra (or was it Niels Bohr):
Prediction is very difficult, especially about the future.
Prediction is very difficult, especially about the future.
Over popularity...
Centralization
Having > 500 million users.
That's a lot of different kinds of people to make happy. That's a huge ship to turn around. That's a huge amount of surface area to be spread over.
Having to serve the lowest common denominator for that many users in a community oriented platform is going to leave a lot of needs unserved, or served poorly. It leaves plenty of holes for a disruptive company with nothing to lose to show up and do something interesting.
The converse strength they can leverage, though, is also their huge numbers: users, engineering staff, sales people, technology partners. If they realize the power of a disruptive idea just early enough, they have the resources and culture necessary to build and scale it in what may as well be an instant fashion, since they can drop it on millions of users at once.
That's a lot of different kinds of people to make happy. That's a huge ship to turn around. That's a huge amount of surface area to be spread over.
Having to serve the lowest common denominator for that many users in a community oriented platform is going to leave a lot of needs unserved, or served poorly. It leaves plenty of holes for a disruptive company with nothing to lose to show up and do something interesting.
The converse strength they can leverage, though, is also their huge numbers: users, engineering staff, sales people, technology partners. If they realize the power of a disruptive idea just early enough, they have the resources and culture necessary to build and scale it in what may as well be an instant fashion, since they can drop it on millions of users at once.
CONTEXT