Hacker News Reader

Comment by nhangen | Hacker News Reader

I don't think any single service will kill Facebook. Hell, I don't think you can do anything but diminish them at this point.

My bet is that the micro-niche makes a comeback and is what catches on going forward. We dont' care about high school friends we never talked to, or family members that blab about their dinner. What we care about are the things that interest us...HN is a great example.

As more people move to avoid the noise, they'll find solace in mini-communities, and I think that's the way forward. Not just one service, but hundreds.

View on HN

Strangely enough, could it be time to attack Facebook the same way they attacked MySpace? Facebook bootstrapped a user base by initially restricting to people with harvard.edu email, which made them more comfortable sharing. Then they expanded to other universities, but only allowed visibility to people within the same .edu.

Now that everyone has been socially obligated to friend their Mom, maybe it's time to attack Facebook with .edu specific sites again? What other natural community boundaries would make sense to target with exclusive sites?

datemyschool seems to be killing it so far...

While I find the idea of many micro-communities compelling, I don't think that this is the direction things are going.

Here's why: the Internet is extremely diverse as it is and overall makes for a bad user experience where people have to cognitively deal with many different (and disagreeing) interfaces, rules, quirks, web addresses, and so on. At the beginning this heterogeneity was seen as an advantage because the amount of easily accessible information was unparalleled. However, in the past few years it seems like most successful platforms (be it YouTube, Facebook, Tumblr, Amazon, Wikipedia, iOS, etc.) have done well because there was a consistent and one-stop shop experience throughout.

There is still tons of room on many different verticals to improve and disruption to occur, but I wouldn't expect things to come full circle.

Well, there's Reddit and its collection of small communities under the same banner.

But I think as more and more CMSs start introducing semantic information in their HTML, we'll start to have more client side features that work across all such websites, and eventually we'll go full circle and reinvent USENET, but now with a dozen layers of abstraction to support it.

When you are shopping, you want a trusted second party. E.g., Amazon.

While Wikipedia is cool, it still doesn't beat the enthusiast's homepage.

And Facebook has never had the personality of GeoCities.

I'm in a mini community of about 10. We all know each other. We communicate through email. I've gotten more out of that than I ever got out of Facebook, and some of the people in my mini group were part of my network in Facebook.

Unless of course Facebook figures out a useful and effective way to provide the same focused view of a niche community within the larger Facebook expanse. At some point they may come up with a way to easily and intuitively allow users to expose different facets of their interests and persona to different groups; if they figure this out the niche sites (even HN) are toast.

They just changed all Facebook Groups to a more-simplified version just a month or two ago so I doubt they're going to get any more sophisticated on that level.

What do you imagine they would do otherwise? More to the point, what's the business case for doing what you describe? They could require FGroups to self-categorize in order to narrow and aid their advertising sales, but that's not foolproof and lots of groups would just as soon choose "Other."

Unless of course Facebook figures out a useful and effective way to provide the same focused view of a niche community within the larger Facebook expanse.

Facebook private groups

https://www.facebook.com/groups

are working quite well among my circle of friends to provide a focused view of one or another niche community that forms a natural subset of my friends. There is a particular Facebook secret group that I visit absolutely every time I'm on Facebook, which is several times a day. That's a practical place for me to ask focused informational questions, and also a place where I can share a lot of camaraderie with dear friends who have had similar--and yet very diverse--life experiences. Because of the stronger filtering abilities of Facebook, I can indeed see how it could supplant many niche sites. (Thus far, HN gets the second share of my attention after Facebook, among all websites, and it will be interesting to see what the consensus here is about the overall community experience in the next year or so.)

One problem with Facebook Groups I find is that you can't have wall posts from your groups appearing in your primary timeline/newsfeed. Because of this, you have to make a concious effort to visit the group to catch up.

Since none of the groups I joined have the same level of activity I'm used to on specialized forums for those subjects, I never bother to check up on them - which of course lends itself to a vicious circle of inactivity.

Posts from groups that I'm a part of do show up in my main feed. Maybe you need to tweak some settings here or there?

Do posts made by others to the group wall show up?

If they are really smart they'll make it even easier for niche communities to use their social graph, or to create custom data records within it. For a lot of site owners it makes sense to outsource account and user info management to someone else. FB has had trouble being reliable enough for that kind of service in the past, but it's not like they don't have the talent and money to do this.

If you look at the Like button, and their other endeavors, I don't think Facebook's ambitions are limited to just getting visits and clicks. They want to become a utility as indispensable as HTTP.

And Facebook are really smart. Keep in mind that if there is a drop, they've probably known about it for months and months, and we're just finding out it about now indirectly.

That said, I hope they don't succeed and a distributed system does succeed.

Not if they can't get us to join. Personally, I'll always be a little too paranoid about what they're doing with their data. Their past missteps gave us all an indication of what they really want to do.

I'm really hoping for decentalized peer to peer communities that are elastically created from locally computed interest profiles. Not a distributed facebook per se, but more like a user-centroid HN that aggregates similar people and interests... that would really make my academic life soar.

In the end, I don't think Facebook would fight for the long tail of user demographics. What they have is quite profitable enough. Why expend 90% more effort to convince all of the tinfoil hat wearers to join?

Technically speaking it takes to much "gardening" to run HN-like site. There are millions of niche forums there already. But it's the platform and the interface that makes HN different. I don't know if there is a technical expertise to similarly run other communities there.

Unless of course the HN platform is released as an open source project.

> Unless of course the HN platform is released as an open source project.

1. go to http://arclanguage.org/install

2. download, untar arc3.tar

3. see news.arc (hn source)

I concur with you.

Reply on news.ycombinator.com