Comment by panarky | Hacker News Reader

Comment by panarky | Hacker News Reader

I see this form of argument a lot:

Companies have an {obligation, fiduciary duty} to maximize value for shareholders, therefore they have no choice but to do horrible things to {users, customers, the environment, civic institutions}

Presumably, decentralization, federation and open source will eliminate the profit-maximizing corporation, thereby stopping all the horrible things.

The problem with this reasoning is that it only addresses one side of a two-sided transaction. The corporation doesn't transact with itself. The other side of the transaction is the user.

What if every user was willing to pay $5 per month for a "clean Facebook" without ads, without surveillance, without psychological profiling, without selling private data to malevolent third parties?

If every user paid $5 a month, Facebook could continue their current business with the best computer science talent and a global network of data centers.

If every user paid $5 a month, Facebook's shareholders would do fabulously well.

But of course, that's a ridiculous proposition. Only an infinitesimal number of Facebook's 2.2 billion users are willing to pay for the service.

Since a social network only has value to a user when all the user's friends and family are also on the network, a Facebook that only has a few people willing to pay won't be very useful.

Advocates of decentralized social networks need to explain (1) who will pay, and (2) how to get enough users to make the network valuable to a typical user.

Until we figure that out, we're stuck with social networks that are free for users, monetized with advertising, and regulated by the state.

We already have decentralization in three primary forms of communication: phone, mail, and e-mail. They worked great in 1998. Why can't they work great in 2018?

There are now phone plans that are paid for by selling your data and sending you spam SMS. There are e-mail providers that operate similarly, and most e-mail addresses are registered with one of them. Those things don't really work for snail mail, but few people use it anymore, because sending a letter costs more and they have come to expect communication not to cost money.

The formerly decentralized services are no longer so decentralized since "free of charge, paid for by data and ads" has shown itself to be a viable business model that users love. (Until they understand what's happening, at least.)

Your point is valid, but we still live in a world where we can call Verizon phones with AT&T phones; send e-mails to Gmail users with Yahoo; and send mail with UPS instead of USPS. With those means of communication, we still have control over our providers.

There is no control within Facebook's ecosystem. Similar with AOL, if you don't use Facebook, you're necessarily out of the loop, and I don't feel comfortable with that.

> and send mail with UPS instead of USPS

Not to detract from your point, just thought you might be interested to know that USPS actually has a legally-protected partial monopoly on the right to carry letters (18 U.S. Code § 1696).

An exemption was added in 1979 allowing private carriers to deliver "extremely urgent" mail. This is why UPS and FedEx express mail envelopes bear that text on their front.

Apart from this exemption, USPS, FedEx and the like are only allowed to deliver 'parcels', not 'letters', though I'm not sure where the distinction is legally defined.

More info: https://en.wikipedia.org/wiki/Private_Express_Statutes

Interesting, thanks.

I don't think decentralization just means > 1. That you can use 4 services is not very decentralized.

While true, I think it is important to emphasize the difference between 4 platforms and 4 carriers. With both postal mail and email, its not just that there are alternative systems, but rather that you as the sender can choose any of the offerings to reach anyone, even if they use a different carrier for their own needs, so there is a lot more competitive pressure on those 4 carriers than with 4 platforms where network effects keep people locked in.

>The formerly decentralized services are no longer so decentralized

People keep saying this but to my knowledge email still isn't centralized. While there are a lot of gmail users in the US but that's not the case in many other countries, many people also have a .edu email and a few years ago I ran my email server myself (I even handled signing papers when I switched jobs over it.) Google even contributed to and pushed for dkim and spf which solved a huge spam problem that threatened their ability to forward mail from smaller outside domains to their users. A lot of people might use one of the big services but as long as you don't have to in order to interact with them it's not "centralized."

You pay for all three.

The world was better when people paid for communications. It's why the internet was better in 1995. Most of the people who had access needed to afford a landline and likely had to own a home or rent an apartment. The only way kids got online is if a guardian had access.

Nowadays, any kid can get online. The range of intelligence and character has expanded dramatically.

> We already have decentralization in three primary forms of communication: phone, mail, and e-mail.

I disagree that we currently do. And only agree for email in 1998 when its decentralized nature costed more to the users than the benefit of being decentralized, hence the transition.

That's only true if you ignore the long-term externalities of centralization.

E-mail is largely centralized now. There's a handful of major companies who are able to run an e-mail service and communicate with the hordes of GMail's users who are living behind GMail's spam filters. (Spam filters, it should be noted, that make the service usable for most people.) So what happened is that decentralization worked for a moment in time, but as the Internet grew it worked less well, and we reinvented the pre-Internet solutions (BBSes/Prodigy/AOL/Compuserve) on top of the open Internet.

It is unfortunately true that e-mail is mostly centralized (as in even if I don't want to use GMail & Co., I will mostly communicate with people on those bign centralized platforms, thus giving them again my data).

But it is absolutely federated - I have the choice of running my own server and can communicate with every other e-mail address on this planet.

Unfortunately, it isn't easy to setup - average Joe has no chance to get his own server. On the other hand, for IT people it isn't extremely hard either. Once setup, it is nearly maintenance free...

I'm running my (and a company) e-mail server since 2001 and I don't want to miss it. I only once had a problem with being blocked by big sites, after moving to a new server where the IP range has gotten on some blacklists, because of the previous owner. Also never a problem with spam filters.

I hope e-mail will never be replaced - or at least there's some new way of communicating federated where every internet user is reachable.

They can work but people have largely moved from them because mobile/Internet based platforms are more convenient. Just as people can travel by horse and carriage but most prefer the convenience of car.

This is what happens as technology progresses. I don't think the solution here is to abandon imperfect technology in favor of old, but to improve upon it.

I hear what you're saying, but the analogy doesn't hold because we no longer need horses. In order to use Facebook, you have to sign up with an e-mail and typically a phone number (and in many cases, a photo ID, birth certificate, etc.)

If we need e-mails and phone numbers to sign up for social media, then we don't actually need the social media. We can go back to simply using the numbers and e-mails.

Of course we don't need it, but the fact that millions use it is the proof that people prefer it....for some reason or another. Added value of some kind.

Better to extract the value of Facebook and move forward by eliminating the negative, than revert to something many people find less convenient.

> If every user paid $5 a month, Facebook's shareholders would do fabulously well.

> But of course, that's a ridiculous proposition. Only an infinitesimal number of Facebook's 2.2 billion users are willing to pay for the service.

Well, at least some tens of millions of users seemed to be happy about the idea to pay USD 1 a year for WhatsApp.

They were running this operation with something like 50 engineers so it should have been really profitable.

Then came Facebook...

WhatsApp without any server-side storage nor content discovery? These are primary requirements towards a social network and they'll cost more than another dollar to fulfil.

You can rent a dedicated server for less than $10/month that has enough resources for an estimated 100+ users. Heavy users that upload a large amount of data (videos, pictures) may be willing to pay more than casual users.

Seems some people disagree but I think you are right.

All chats I've been in on Telegram and Whatsapp over the last few years could be stored in a few gigabytes of disk space.

How would I know? I've had a look at what Telegram and WhatsApp stores on my phone.

A sharing model where the content you create allows you to access content is viable (see Bittorrent ratios). Facebook created a side market where they made markets to trade user data and targeted ads. That should have never been part of the bargain. It's like getting an ad toolbar from installing Java.

What if every user was willing to pay $5 per month for a "clean Facebook"...

The problem is that, to a sufficient degree, they're not. Likely because the value proposition (or its assurance) is a hard sell. And that hard sell limits network scale (and hence, value).

Mind, even free alternatives have exceptional difficulty in competing with scale. Viz, G+.

Reply on news.ycombinator.com