This partially explains Reddit's staying power. A popular subreddit that hits the point of degeneracy spawns new subreddits that aim to return to it's roots. No-one leaves the platform and the cycle continues.
I still say that the smartest thing we ever did at reddit was allow users to create their own subreddits and then decorate them how they choose. The second smartest thing was making text posts official.
Both of those features differentiated us from the competitors, and as you aptly point out, it has a nice side effect of keeping people on the platform who might otherwise leave (although that wasn't an intended consequence).
I always thought the smartest thing was keeping the frontpage unappealing enough to scare off anyone who hadn't already been told by a good friend it was worth checking out. It is sort of this pseudo invite-only filter.
I would love to see a "defluffing" filter for reddit. Just remove low effort comments, memes and low effort content such as images and short videos under 3 minutes. I'd like it to be easier to locate the comments and articles that are worth reading.
Personally I'd browse/read reddit 10x more if it had the interface of http://www.scrolldit.com In fact scrolldit.com went down for a week or so and I basically stopped reading reddit. I tried visiting the front page a few times but it was so hard to find anything sticky enough to keep my interest I basically stopped visiting.
Just checked and fortunately scrolldit is back up so my reddit reading will probably go back up too :D
The writer (Sangeet Paul Choudary) doesn't mention Clay Shirky but as far as I can tell, Clay has written the most essays about web communities and scaling problems:
"LinkedIn creates friction by preventing users from communicating with distant connections. This ensures that users do not receive unsolicited messages. This also allows LinkedIn to offer frictionless access (OpenMail) as a premium value proposition."
I'd say my inmail is 95% unsolicited messages from people who have bought access.
I read that phrase as meaning the degrees of connection feature is a way to stop users receiving unsolicited messages, something which contributes to the reverse network effects the author talks about.
Having the paid, 'message indiscriminately' feature undermines that. So if the intention is to reduce unsolicited messages to reduce spam then it doesn't work.
I think the intention actually is to reduce spam from those who don't pay for it. So yes, working as intended.
The real social network is the one you create on your own, with people you wish to network with. And it has to be small by design, otherwise it doesn't work.
Now, it doesn't matter what you use to create such a network, be that email list, facebook group, whatsapp, or fill in the blank.
No company or platform have been able to "create" such a social network for you. The best they can do, is provide appropriate tools for you to create your network.
And as far as I can tell, email lists still serve that purpose, irrespective of the hype the platforms like Facebook, Slack etc may generate.
As I mentioned in my first comment, you create network with whom you wish to create network with. So it doesn't matter where your friends are (be that Facebook, or at home, just a phone call away); it is you who is going to reach out to them, to create a social network.
> provide appropriate tools for your friends to create your network
Each individual creates his or her own network. So if someone reaches you first, seeking your company to get you into their network, it is as same as you reaching them, asking them to join yours.
My ultimate point is, that the "platform" does not create a social network (for humans). The best they can do, is create bells, whistles and tools. Humans create their own social network.
YouTube long ago reached the failure point for me in delivering recommended content. Their recommendations frequently include things I've already watched, and the tangentially related items they recommend miss the salient point more often than not. I still use it, but only when searching for specific things.
I don't think YouTube was a "network effect" success story. It was mostly a "free content" success story.
I think that's why it's been able to get away with a poor recommendation system and a broken comment system.
But that's also a reason that fixing these things would be very hard - there's no network consensus concerning what is the users want.
Edit: this relates "nicely" with the post concerning using deep learning in the youtube recommendation system. The sophistication of math deployed is impressive yet the responses uniformly deplore the result. I am suspecting now a big part is that the system broadly never knew "what it wanted" (what is a successful recommendation? Something people click on? etc).
I recently started "watching" youtube. That is, use the front page and the recommendations, and it is a very good alternative to channel surfing on TV. I have discovered a few very funny panel shows and it is fairly good at giving me new episodes of shows.
There are so many use cases to Youtube that recommendations can never be good.
I want it to recommend music I have heard, for when I use youtube as a radio, but I also want it to recommend me new music in the same genre, although sometimes I want to hear something completely different and want that recommended.
I want it to recommend funny videos I've seen when I'm you youtubing with a friend ("Oh, yeah! Did you see this one? You have to see this one!")
I don't want it to show me funny videos I've seen when I'm watching youtube alone.
Indeed, well you'll notice that on that thread, deep learning had been happening for a while yet every.single.post said recommendations were awful, worthless.
I think this is actually "cause for learning" rather than complaining. I think the OP gives a good analysis of the tradeoffs involved in creating content. I'm not sure if a deep learning system can make tradeoffs out of the box. Perhaps Google could consider non-zero sum game theory for this and use some of their lessons from AlphaGo.
A browser add-on to remove the right column from Facebook would be a win. Youtube apparently needs this as well. So do a lot of "social" sites. It's become a convention - that's where the crap links are.
The article makes a halfway decent attempt at the pitfalls of large-scale networks. However, the article doesn't come close (in my opinion) to making a case that this will lead them to "fail".
Facebook has become massively popular and is still growing. While I am not particularly active on it, I recognize this is atypical. Think about how many devs ise FB OAuth for example, as it is insanely prolific.
Networks can fail at any time, but very few have after reaching escape velocity. Probably the most notable was AOL/IM and more recently the significant degradation of reddit. Networks must seriously mess up accross the 3 main identified categories, and not keep any pace with their users changing needs.
I would say the biggest threat to networks at scale, which oddly wasnt really discussed, is a failure to monetize or convert their network into value. Many people find twitter useful, for example, but it is rumored to be shopping itself around. If anything, the danger of networks is expecting data and adrevenue to be a sustainable business at scale. Obviously, it is a good problem to have 500million plus active users, but it is expensive and difficult. It is harder, one might inagine, to have 50m paying users.
> Facebook has become massively popular and is still growing. While I am not particularly active on it, I recognize this is atypical. Think about how many devs ise FB OAuth for example, as it is insanely prolific.
But are they real users? There are lots of reports of many "users" being bots or sharecroppers on click farms.
There's this social network that all of us readers have been using since birth and will continue to use for life: the city. It also delivers connections, content and clout. It keeps growing the userbase eventhough friction is huge, content quality mostly sucks and clout for new users is a far-fetched dream.
My point: maybe Facebook is the first online-city and is here forever.
Ps: so far I am the only e-hermit I know that has left FB. Yet, once my kids grow and join FB I may need to re-join to go visit them.
> Ps: so far I am the only e-hermit I know that has left FB. Yet, once my kids grow and join FB I may need to re-join to go visit them.
I have also completely rejected their platform(s). A lot of people say this will destroy your social life, but the real city where I am always provides more than enough to enjoy.
I think your children will visit you in whatever place you like to live :)
I'm amazed at the failure of Facebook in this regard.
They were in a unique position to make it work, and then failed.
With make it work, I mean that everybody moves in different circles, family, work, friends, hobbies, etc. And you need to have easy control over who sees what.
And Facebook has most of it, except that last bit of UI to make it work.
That statement might lack nuance, but it's not wrong. Skype and Facebook only have value to me because my friends and family use them. The value to me increases as more friends use those platforms, and would decrease if my friends left.
Yes, but you can also get reverse network effects, e.g. young people not being active on Facebook because their parents are there (and denying a friend request would be awkward).
Skype seems like odd-one-out in these lists. Is there actually any sort of community?
I do use it, but we tend to arrange Skype "calls" ahead of time and out-of-band (e.g., by email). Unlike "real" social networks, I don't log in to Skype just to see what's going on in the world or with my friends.
Facebook is slowly becoming peoples new rolodex/contact list. For this, it needs everybody to be on. Phones are only good if there are somebody to call.
However, as a discussion forum / news site, it can be uncool to have to listen to all those people who aren't your real friends.
