Hacker News new | comments | show | ask | jobs | submit login

Maybe there's a trend here, but maybe its just me being short-sighted. I'll type out what I'm thinking.

I think that the uncomfortable future for a lot of the e-properties that people enjoy happens to hinge on the fact that the way they are the most useful clashes with the ways they can make money. Big time.

Facebook will sponsor stories, alienating users a little for eyeball time.

Twitter will make sure promoted things get their eyeball time by crushing 3rd party (alienating users).

Digg will- well. You know.

All of these moves hurt the utility of those companies. All of the companies know that.

But something happened with Digg, didn't it? The models of the above have to very carefully balance utility/pleasure of use with a certain amount of intrusiveness to make money because they have found few other good ways, though some get credit for experimenting.

And every time a major e-property does this they make a chink in their dam, inviting their sweet sweet reservoir of users to look beyond into a river of startups. Maybe there are more useful options out there for little fish. Cleaner spaces, open air. Let's be honest I'm bad at metaphors. Moving on.

Myspace became ridiculous and thousands (millions) fled to Facebook.

Digg became ridiculous and many fled to Reddit.

Twitter and Facebook are trying to monetize more and will become more ridiculous and less useful in the process. They might survive, but they are inviting others to take their place in the meantime.

And someday those new companies will build dams around users and either try to get them to gawk at more ads and maybe fall in the same way. This model seems to work for a little while, but they still need to solve the large clash of interest brought about by advertising. Search engines still do it best: ads displayed that try to match queries with relevant content aren't necessarily offensive to taste.

Almost everything else is still trading user abuse for irrelevant product. All users wanna do is talk about the kind of sandwich they ate. Nobody wants to "Learn how YOU can help Hyundai make August National Fuel-Efficiency month!" (real ad).

Yet another reason things systems like Twitter and Facebook shouldn't be silos with centralized control, but distributed, decentralized, self-hosted systems built on open standards.

I really can't see decentralized social networks being successful. If all hosts are going to be synced the traffic would be huge. If they aren't synced search for both messages and users would be painful. Patching together a stream when the people you follow are scattered across a hundred different servers painful. Account creation for average tech illiterate users painful. If they have to rely on some third party to provide a host, how is that any better than relying on Twitter or Facebook, as soon as the hoster decides it isn't worth the expense or to boot you for some random reason what happens?

Wouldn't it be better to build a centralized message bus and identity system under some kind of foundation like Wikipedia. You would have to get benefactors to fund it is the big problem. Twitter has proved itself to be a priceless service for communication and news streams, its like the real-time Wikipedia.

Its just major regrettable the people who are running it are completely trashing it in the predictable pursuit of the beloved buck and a fat IPO.

I'd love nothing more than foundation based centralized message bus and identity service offering API services to any and all responsible developers and apps, short text only messages and links to avoid all the issues with attachments little or no censorship, private communications between users encrypted so even the foundation can't read them or offer them up to prying governments, some mechanism to control and kill spam.

Because Twitter and Facebook's search features are so great? Twitter doesn't even allow you to search your own tweet history past a certain point.

Patching together a stream when the people you follow are scattered across a hundred different servers painful. Account creation for average tech illiterate users painful.

This is a matter of execution. I think BrowserID is making this a lot easier. Also, we have an example of a decentralized communication system: email.

If they have to rely on some third party to provide a host, how is that any better than relying on Twitter or Facebook, as soon as the hoster decides it isn't worth the expense or to boot you for some random reason what happens?

If it's decentralized, I have more control. I use gmail, but I can keep a backup of my email and change my email provider whenever I want. Not so FB or Twitter.

I think Wikipedia's great, but I think its governance system is pretty whack, and I wouldn't want my communications to be subject to anything that capricious, undemocratic, or otherwise not in my own control.

Not sure about Facebook since I never use it but Twitter's search feature is crippled because Twitter wants it that way. I think they are planning to monetize access to full history, though I dont remember exactly what was being said on this a week or two ago. When it comes to your posts and your stream I think you should be able to search it any way you want and cache it any where you want.

I cite Wikipedia only for the foundation model, funded by contributions, to free us from corporate control, being bombed with ads, and having our personal information harvested for profit.

With a message bus foundation you wouldn't have most of the politics associated with group editing of articles which is where Wikipedia has its worst problems. The only places there would be a risk of politics at the Message Bus Foundation is in terms of use, abuse mitigation, spam filtering, censorship issues and interacting with intrusive government requests. When it comes to privacy and freedom of speech it should be designed to be as private and free as possible.

If all hosts are going to be synced the traffic would be huge. If they aren't synced search for both messages and users would be painful.

Not really - I could see that if you decentralize then you can setup something like a peering scheme between social networks. And if a person wants to find another person on another social network, then perhaps a peering relationship should be setup with that other network, or perhaps that person can migrate to the other network.

It could be practical, but the infrastructure at the moment isn't really there.

If someone is following thousands of people in their stream, and they are scattered across a hundred servers, how do you go about assembling their stream for them in real-time. Is the host server going to have to send user id's for everyone every user based on that server is following, to every server where the people being followed are based, and have every one of those server send all the messages of interest so the host server can build streams for all its users?

Twitter didn't stop sucking reliability wise until they switched to a single persistent user stream connection and abandoned REST most of the time. You really want to have that one persistent connection for a user's stream.

That's a reasonable objection. However, could you get an aggregating server that collects the updates between networks? You might need to batch it, but for networks with better network connectivity it could be done in real time, or close to it.

There is no real need for instantaneous updates, unless the networks so deem it necessary.

I couldn't agree more with your statement. User adaptation will always increase costs, and increased costs will always increase ads, and increased ads will always increase alienation. It's a lose-lose business model. It's very interesting to watch these companies win the hearts of millions of users then change, and lose those millions of users, in the matter of months or sometimes less. Crazy.

The takeaway lesson for me is that user-generated content is incompatible with advertising.

Now, as a caveat, I'm the sort of person who thinks that modern advertising is 99% toxic sludge in the mental environment: I am firmly in the Kalle Lasn camp. However, here's how I see it working: users come to the site because of the UGC, but advertising, by nature, has to grab attention at some level. You can't design for both "make the UGC, stuff users actually care about, primary" and "make sure that you get ad clickthroughs." Since the latter pays the bills, it tends to win - and users tend to leave because on one level or another, it's obvious that the site isn't about them (and if your site depends on UGC, your users ain't wrong to think that the site should be about them!)

The secondary takeaway is that advertising is not a long-term sustainable business model.

Advertising has a strong tendency to push startups into this insular turtles-all-the-way-down bullshit: social media with ads for analytics, analytics for ad-based startups, startups about how to reach demographics, and so on. They're as familiar an archetype around here as Punch & Judy. But it's all the tertiary economy, it's all flash: the fundamentals of advertising as a business model only get weaker over time. The effectiveness of advertising at all, in any context is debateable, and participating it is participating in an endless arms race. I argue that this should tell us that advertising is not the way to go if you want to build a lasting company: you need to take money in exchange for goods and services (real goods and services: advertising doesn't count, as I'm in the process of arguing).

That's hard! That's really difficult. PayPal, Square, and WePay should show us how ridiculously, gratuitously difficult the "take money for" part is, and taking people's money once you've convinced them that you've got something worth their money, is the easy part! Before you even get there, you have to build something good, and none of us should have any illusions that that's easy. Then between those two parts, there's "persuade users that your cool thing actually is cool and worth paying money for," and sweet leaping Buddhas that's a lifetime of work in itself (the lifetime of work, in fact, that advertising is doing in such a toxic and commons-destroying way right now).

But: you are a hacker. You are a hustler. You are an engineer.

Solving hard problems is your job.

Advertising is not a business model: advertising is a problem to be solved. Advertising is a bullshit legacy of past business models, here to be disrupted. The reason that I hang out on Hacker News is that I believe that its hackers, hustlers, and engineers are the people who can look at the way we do something now, think "that suck! I can do better than that!" and then do the difficult, frustrating, painful work to actually make something better, show people why it's better, and accept the monetary rewards that come from having made something better and proved it.

We can do this.

Your arguments about why UGC is incompatible with advertising seem like they could apply equally to all content.

i think a better argument is that unstructured UGC is incompatible with advertising. Pinterest, for example, seems to be doing all right, because their site is basically a repository of things people want to buy. StackOverflow doesn't seem to be having any issues using ads to monetize, because their niche makes their ad-space valuable to advertisers and relevant to users. Where facebook and twitter fail is that they have tons of data about users, but not about what users want at that particular instant. An ad for windows server hosting on ServerFault is not competing with the content, it's complementing the content. it's a natural progression that doesn't require me to refocus to be effective. Facebook knows what i like, but not what i'm thinking about. When i'm looking at pictures of friends, any advertisement requires me to refocus onto a different topic.

This is sort of inspiring, but until I see more legitimate businesses making money off of a better model than advertising, it just doesn't mean much to say it. If people didn't have to pay for costs for things like bandwidth and marketing, we would probably see a lot less advertising too.

Yea, advertising isn't a long term business model... tell that to Google, and most other successful advertising firms.

As I start building a social sharing startup I know we might be crazy telling users, from day one, how we are making money on Referly (shared commissions on purchases they generate when they share product links - and matching social disincentives for spamming your friends and having a crappy signal to noise ratio of clicks to purchases) but I just don't see how companies doing the "bait and switch" of adding a business model later is going to keep working. People are getting smarter about this, and more aware that they are becoming the product when they use these services that are eventually ad-funded.

I'm not trying to say we are some genius with Referly, because the money from day one part does turn off some people and might be slowing our adoption a bit. But the upside is that we have to deliver a degree of utility that makes it worth it to the user to know we are making money (and they are making money too in our model).

Is this too much capitalist utopia - or could this really be a model for future businesses besides ours?

A very good point.

Ads are probably toxic to social media in most cases.

And worse than the ads, are the advertisers, and what they want.

Can we really say that we are individuals interacting with each other? When those that we interact with are only facets themselves?


link's broken...

Can we really say that we are individuals interacting with each other, when those that we interact with are only facets themselves?



What I wonder is what the effect could be if Facebook or some other "social" company were to cut users' in to some of the ad revenue, e.g., as in somehow getting compensated for having your photo appear next to an ad for a company you "liked". I can imagine benefits, as well as strained trust, between the company and users -- but it seems to me that so far the users themselves have only played one part of the process when there may be a lot more to that.

But I haven't thought about this very in-depth yet, it's only crossed my mind once or twice reading these headlines. Can users, advertisers and "social" service providers interests overlap somewhere?

No thanks.

Your compensation for being “used” in the advertisements is using the service, that’s how you pay for it.

No way to make it more lucrative?

Edit: I don't pretend to really know how to do it, or what the outcomes could be, but I guess I'm talking about socializing social networks (part of the business value itself), if that makes sense.

no, the minute you add financial incentives people start to wonder if they're fair, and get piqued that they aren't greater

What is the solution then? Ads are going to be integral part of the Internet.

Ads need native integration that is clean and unobtrusive. Nobody complains about Google AdWords. There is whole new industry of SEO/SEM around it.

I'm hopeful about "sponsored stories" or "sponsored tweets". I hope they will do well because they won't distract user like banners or interstitial. User can completely ignore sponsored stories.

MySpace was train-wreck due to its UI and it was never popular in masses (only teens loved it). Digg failed due to major UX upgrade and users fled to Reddit.

This makes a lot of sense.

Facebook + Twitter alienate their users by watering down the experience with ads, losing the 'cool'ness they need to keep eyeballs on them.

I think that's where Google+ can swoop in and make Twitter+Facebook look like Myspace. Thoughts?

Why wouldn't Google+ also throw ads at us, since that's their core business?

A couple of reasons.

First, G+ has three values for Google. 1) It gives them rich information about their visitors demographics, interests, hobbies, friends, etc. 2) It's a platform for displaying ads. 3) It's a way of screwing with Facebook, who is a potential threat.

But as Facebook has learned, social networking websites are very poor places to display ads. When someone is searching for a "web host" or "how to stop mysql from crashing", a targeted ad for a new web host or a new database engine is (relatively) likely to be clicked on. When someone is checking to see if their friend Hazel has posted any pictures from the party last weekend, there is no targeted ad which is likely to be clicked on.

G+ is simply not a good platform for displaying ads. And trying to do so too forcefully will push people away, which will undermine reason 3. It will also - crucially, from Google's point of view, since it's (probably) the entire reason they developed G+ - it will undermine reason 1. Google's core business isn't throwing ads at people, it's throwing carefully targeted ads at people, and G+ is a major part of their strategy for getting the data they need to improve their targeting.

Because their enterprise business is growing and they could take a loss to undermine Facebook and eventually funnel eyeballs towards docs, etc.

But they probably still will.

The Google+ API is not designed to empower third-party developers. Maybe they're still searching for a way to do it right, but currently it's too limited for an ecosystem to flourish. Why should a challenger come from a big company anyway? Both Facebook and Twitter were small startups during the fall of Myspace.

> "Learn how YOU can help Hyundai make August National Fuel-Efficiency month!"

Learn what YOU can do help OUR company! Dale Carnegie must be rolling in his grave.

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