Hacker News Reader

Comment by olau | Hacker News Reader

Actually, I think behind the nationalisation idea, which doesn't seem workable given the circumstances, there's an interesting thought: these companies are sitting on basic infrastructure so instead of treating them like other companies, perhaps we should treat them like utilities.

Note that this piece appears in The Guardian. I don't know how it works in the US, but I'm of the impression that in many European countries, basic infrastructure like energy or transportation or communication links is either very tightly regulated or owned by a non-profit organisation that in turn is owned by either the local consumers, the local authorities, or by the state.

DNS operates in this way, as far as I'm aware.

Can we move in this direction with things like search (Google) and personal publishing (Facebook)?

Perhaps if the tech community delivers the software, and the societies themselves start a serious discussion about it and delivered the users?

View on HN

A simple solution is to treat these companies as telcos.

We could learn something from history. See e.g. the Mann-Elkins Act (1910), [1]. And the telecommunications acts, [2], [3].

[1] https://en.wikipedia.org/wiki/Mann%E2%80%93Elkins_Act

[2] https://en.wikipedia.org/wiki/Communications_Act_of_1934

[3] https://en.wikipedia.org/wiki/Telecommunications_Act_of_1996

What we should learn from history is that ICC regulation was crippling (nearly destroying the rail industry in the US: http://www.theamericanconservative.com/articles/a-nation-der...) and the county has been a lot better off since railroads, air freight, airlines, and yes telcos were deregulated.

That is not to say that we can't regulate things we care about (data privacy, competition, etc). But looking with rose-colored lenses at ICC regulation is foolish. It's the worst sort of regulatory regime: where government bureaucrats are in charge of approving/denying the rail ties that go in the ground and how much can be charged for a ride.

Definitely no bias in the take you linked...

Whatever you wanna say about how regulation worked out for trains in the US, the situation in the early era of US rail was not a good one. It was one of the worst monopolies we've seen and was terrible for the workers.

What bias? As far as I can tell, the author has no connection to the rail industry.[1] As to your other point, the fact that early rail monopolies were bad doesn't mean that ICC-style regulation was good. You can have labor protections, antitrust enforcement, and safety regulations without having the government be involved in deciding how much a trip from Chicago to Milwaukee should cost. There is a reason there was a bipartisan effort to get rid of those regulatory regimes in the 1960s-1990s, and why Europe copied those reforms in the 1980s-2000s.

[1] If you're referring to the fact that the article appears in the American Conservative, merely being conservative is not a "bias." In any event, left-leaning organizations like the Brookings Institute have featured similar coverage: https://www.brookings.edu/research/the-success-of-the-stagge....

I wouldn't call Brookings left-leaning. Centre-right or better neoliberal.

I agree with your points though, the rail situation in the US is awful. Good regulation is hard to do, especially in a place where regulators typically do not share the interests of the people they are in the service of.

I don't think Europe copying the reforms you mention is necessarily a good endorsement of them. The rail situation in (at least Western) Europe has also become quite bad.

> A simple solution is to treat these companies as telcos.

Woh woh woh. Back the truck up. We're not even treating ISPs as common carriers. I mean, we are, but the FCC is about to reverse that.

If that happens we'll all end up paying more for slower access to limited content. It'll be like going back to when you thought AOL or Prodigy was the internet. Even if Google and Facebook were put under greater government supervision, you wouldn't have many options outside of those anyway.

> we should treat them like utilities.

As these are large organizations involved in a variety of activities, it might be possible to separate the infrastructure parts into a strictly regulated utility. Google Search becoming a utility doesn't have to involve stuff like their hosting services or their hardware products.

It will be hard to get buy-in for an all-or-nothing plan like "Nationalize/Utilitize ${BUSINESS}", because it groups together too much into a single proposition. We "just" need to figure whatever minimum change is needed to protect the infrastructure. I realize this, like most important political problems, doesn't have a trivial solution. It will end up with imperfect compromises that everybody is annoyed with. I'm just suggesting there must be a better, more nuanced way to fix this problem than a Boolean "Convert to utility? (Y/N)".

> but I'm of the impression that in many European countries, basic infrastructure like energy or transportation or communication links is either very tightly regulated or owned by a non-profit organisation that in turn is owned by either the local consumers, the local authorities, or by the state.

"Europe" is a big place so it's hard to talk in generalities, but this is mostly not true. Europe followed the US's deregulation trend and in many cases European utilities are less regulated than American ones.

The EU electric market is similar to the U.S. As in the US, it is split up into distribution, transmission, and generation components. Distribution tends to be handled through regulated local monopolies or municipal entities, while generation happens in a competitive market: http://www.europarl.europa.eu/RegData/etudes/BRIE/2016/59351....

Telecom market structure is roughly similar too. Big telecom companies that were previously state-owned were privatized in the 1980s and 1990s (e.g. British Telecom and Deutsche Telecom). The biggest difference probably is that the EU has strong rules requiring wholesale access of "local loops" (the wires into peoples' houses). However, this is a smaller difference in practice than in theory. In the big EU countries, unbundling doesn't apply to cable, http://www.oecd.org/sti/broadband/2-7.pdf. In many EU countries (notably Sweden and Spain, which have among the highest fiber deployment), unbundling doesn't apply to fiber either.

If you want to turn these into utilities, you start with a standardization process that allows the public to utilize them as infrastructure.

Like there is an HTML standard, there would likely be a search standard and a social-network/photo-sharing/messaging-platform standard. People would run their own servers that host their own data that connects to these standards, with other companies creating apps that people can use to connect to them.

And make the standards UX friendly, so that it becomes obvious and intuitive enough to use.

The trouble is that nobody follows the standards. Well, very few do. Not even the browsers are standards compliant.

Do you think we should have regulated AltaVista and MySpace as utilities, way back when? If so, what would this have done to Facebook and Google's prospects? If not, what makes something a utility?

There's not a clear boundary, but it's basically the same as when electric power turned from something-only-a-few-rich-could-afford to a utility. At some point, electric power was important enough that it was deemed necessary to impose tight regulation that ensures its availability under fair prices to all citizens (which is what being a utility means to me).

That analogy makes sense. So the discussion should be around what makes Facebook as important as e.g. electric power.

Personally, I think Facebook is more like a lottery or a tabloid than it is like electric power. Citizens are fine, nay, better off without it.

Most arguments e.g. ITT fail to convince me on the critical-ness of Facebook/Google. Instead, the possible harm they cause is brought forward. That's a serious issue, but I don't think it has bearing on the utility analogy.

Google is undeniably important to people. And mainly because they've squeezed themselves into every part of our online lives. Facebook is less so, though it has become a platform that many businesses rely on.

It becomes less clearly undeniable once we separate Google into its constituent parts, though.

Is search a utility? How about email? Webmaster tools for SEM? WiFi thermostats?

It seems unfair to blanket classify Google's offerings as utilities, if many of their products and services are just also-rans.

Again, the concentration of power and the strong competitive position of Google can be problems in themselves, whether or not the products are utilities.

For certain pieces I certainly can make really good arguments.

For example, I can certainly argue that YouTube is a problem.

Google effectively uses search to subsidize YouTube. This chokes off any competitor to YouTube as they simply cannot afford the bandwidth costs (YouTube was hemorrhaging money when Google bought it).

If you forced YouTube to stand on its own, you might actually get some competition to it.

> Google is undeniably important to people. And mainly because they've squeezed themselves into every part of our online lives. Facebook is less so, though it has become a platform that many businesses rely on.

But both of these are free. I don't see how comparing these to the expense of early electric power makes sense. Title II classification for ISPs is currently under threat and surprisingly not the focus of this article.

In fact, the article doesn't mention net neutrality or title II classification once. It is odd that it discusses "platform" monopolies without mentioning "broadband" monopolies.

Facebook and Google only charge fees if you want to advertise through them. And, it is pretty cheap unless you're trying to control a nationwide or worldwide narrative.

I think we should move back to the model we had at the beginning of the internet era:

1. Universities and government agencies create the protocols and the software.

2. Companies create just the hardware.

3. Infrastructure is run and maintained by the government with the help of companies.

Universities and the government providing all the software doesn't sound like a great situation.

Good luck with the legal work required to patch up Google's physical infrastructure than spans, literally, worldwide. For example, if it owns a cable (or just wavelengths) that connect Asia to Europe, how would you handle that ? It certainly doesn't comply to US territory regulations. Unless of course I'm mistaken which is very likely.

An undersea cable will have to comply with the regulations of the connected countries. A regulator can just come and disconnect the part that runs through their jurisdiction if laws are broken.

You do realize a cable that connects Asia to Europe could be like 6cm long, right?

It's a stupid example I gave, also sometimes the connection has to take the not shortest route due to legal restrictions. So no, not always.

The Guardian is based in the UK where energy, water, transport and communication are all handled by the private sector.

Sort of.

The underlying infrastructure is quasi-nationalised in many cases and the private companies sit on top.

For an example look at the big six (largest electricity providers) they all buy electricity from the same place at the effectively the same price, they don't even maintain large parts of their own infrastructure.

The way it was privatised was the worst of all options.

Same with the railways. The companies Joe public interact with don't own the track etc.

It's a truly British solution, what's the worst way? Let's do that.

This would also make me a lot happier about sharing info with said companies. Right now I avoid their products (the non-open source ones) because they just have too much market share and power, but if a site like Facebook turned non-profit I'd be much happier to use it.

It just sounds like a great idea all around.

Facebook goes "non-profit" -> becomes regulated in terms of privacy/earnings -> competing social network springs up -> is able to advertise and boast features that the "non-profit public" Facebook cannot have due to cost or regulations -> new social network is now the de-facto Facebook, and you're right where you started.

and this is assuming that the whole "lets beg Mr. Zuck to pretty pretty please stop doing that whole business thing and start serving us for free, we'd appreciate it" thing works out.

I'm not at all convinced that there are any substantial new features that would overcome the inertia of lock-in (via one's existing connections).

I mean, G+ couldn't do it, and Google tried damn hard. If they can't do it at their scale, who can?

Honestly I'd be more worried about nation states having access to the information held on Facebook and Google than I would private corporations.

It's a perfectly fine situation for nation-states to have access to such data, because nation-states already have access to everything inside Facebook and Google. In the US, you only need a warrant to access anything inside those companies.

Indeed but at least with a warrant all access is documented on public record. Which has got to be better than already having access to all that data in private. Plus private companies can (and sometimes do) challenge a warrant.

Nobody said anything about a nation state (in this subthread). Just non-profit and tightly regulated.

The topic is about nationalisation.

True but the topic is about nationalisation so I still felt my point was relevant even if it was tangential to the GPs tangent

They can force Facebook etc. to hand over the data anyway, so it does not seem like much would change in terms of the threat posed by governments.

True, but I'd rather that process happen publically rather than in secret (as would be the default if the Facebook and Google were nationalised). Not that I'm implying there isn't already secret requests for data (if PRISM is to be believed) but at least with the aforementioned being privately owned there is some chance that unlawful / unjust requests for information can be challenged.

> I'd rather that process happen publically rather than in secret (as would be the default if the Facebook and Google were nationalised).

You're making it appear as through National Security Letters were public, which they very much are not.

It only appears I'm making that conclusion when you selectively quote small passages from my comment. The very next sentence that follows the portion you quoted starts off by saying:

> "Not that I'm implying there isn't already secret requests for data"

I appreciate that you consider it worth reiterating the point about national security since I had only glossed over it. But I don't appreciate you manipulating my comment to suggest that I was ignoring that point entirely when I quite clearly did raise it.

I doubt any forcing is involved. These companies are willing participants in government surveillance.

That's why we need universities and government agencies to work on federated protocols, just like they did with email in the old days.

The SMTP equivalent for Facebook and Google is HTTP which is a federated protocol. With email anyone can store the content on a centralised server in any format they want. Be that inside an open source RDBMS like MySQL or PostgreSQL, or a flat text file like Postfix defaults, or a proprietary format like Microsoft Exchange does. And the same is true for the services that Facebook and Google offer.

The issue here isn't federated protocols (though that is something we could benefit from in other domains) but rather the dependance on a centralised archives in the cloud. Most people simply don't care or even aren't aware of self hosting solutions. And those that are aware often choose cloud solutions because it's cheaper and/or less hassle than self-hosting. Federated protocols doesn't fix this (just look at how hard it is to set up a mail server compared to running Google Mail).

Plus in the case of Facebook you have network effects where a social network is only as good as the size of the member base.

Yes, I agree that content silos are a problem.

But with IPFS, [1], (which is a federated protocol) you can "host" a HTML file or video or any content without any difficulty yourself.

I'm not sure if it would be suitable for the functionality in Facebook though (sharing stuff with friends only), sending messages, so that's part of the reason why we need more research.

The network effect of Facebook can be mitigated by introducing laws that open up competition, like the telecommunications acts, which opened up the market for telephony.

[1] https://ipfs.io/

You could host HTML files et al without any difficulty long before IPFS came along. There are plenty of easy to use web hosts out there aimed at the layman for more than 2 decades. I mean that literally as I have none-IT friends that had sites up on Geocities (and it's equivalent) back in the 90s.

However as much as we could argue the differences between IPFS and the plethora of other hosted solutions all day and night, but ultimately it would be a moot argument because as cool as IPFS is - and I do think it's a genuinely interesting platform - it is still only a platform for serving static content. So it's definitely not suitable for Facebook-like functionality. In that regard IPFS is more akin to Amazon S3 than it is Facebook.

> The network effect of Facebook can be mitigated by introducing laws that open up competition, like the telecommunications acts, which opened up the market for telephony.

While I can't see that happening - or at least not inside my professional life time - it is an interesting point. We've seen monopolies split up before and while Facebook is a long way from being a monopoly it would be interesting to see which line they needed to cross before the government stepped in.

> it is still only a platform for serving static content

The name might be slightly misleading (InterPlanetary FileSystem) but IPFS has more functionality that just hosting files. Two that comes to mind is pubsub which enables you to have a distributed publish/subscribe network to send messages globally and "dialing directly to nodes" which enables you to send/receive messages directly to peers without having to use location-based addressing.

While the primitives in IPFS are simple, they enable you to build powerful architectures, such as CRDTs. So it wouldn't be impossible to re-implement something like a social network on top of IPFS.

Disclaimer: I'm one of the developers working on IPFS

Perhaps it's the online identity part that needs public infrastructure protected by legal constraints? Let the free market piggy-back on top of that platform then. Same thing goes for loyalty cards.

> DNS operates in this way, as far as I'm aware.

Your awareness is wrong. DNS is, by and large, operated by for profit enterprises. Y combinator, for example, uses Amazon Web Services as their DNS provider.

Reply on news.ycombinator.com