Why Your Cell-Phone Bill Should Be Going Down But Isn't | Hacker News Reader

Why Your Cell-Phone Bill Should Be Going Down But Isn't

readwrite.com

138 pointsnkurz128 comments

Prices are coming down because the government prevented the ATT-Tmobile merger and Tmobile decided to start competing on price. Thank goodness for a little government regulation and that created competition.

> Thank goodness for a little government regulation and that created competition.

Heh. Funny comment, since it's the government who created in the first place the band of 3/4 major actors in very country, by restricting the entrance with a system of auctions to get access to mobile frequencies.

That's precisely because we lack competition that the prices are not coming down, and that's what the government has been preventing for ages in the mobile industry. With 3/4 key actors it's very easy to have everyone meet for dinner and agree on fixed prices that are not going to move.

Maybe I'm naive, but how else would you prevent interference -- or even a nefarious "competitor" from otherwise "using" the frequencies? I know very little about radio waves/signals, but AFAIK a HTTP-type approach wouldn't work without a lot of up-front collaboration and a free market wouldn't have went that direction either, correct?

Here's a libertarian argument against government ownership of the airwaves: http://mises.org/journals/jls/20_2/20_2_2.pdf.

Essentially the idea is that the frequencies would be treated as private property, with the exact nature of this being worked out in the courts. There's precedent for this from 1926, before the US government officially took up ownership of the airwaves, in Tribune Co. v. Oak Leaves Broadcasting Station.

"The Chicago Daily Tribune, calling itself WGN... broadcast entertainment as a means of marketing its publication. WGN filed a complaint in state court against another radio station, Oak Leaves, which had begun broadcasting in an adjacent wavelength. WGN claimed that it was necessary to maintain at least a 50-kilocycle separation between stations located within 100 miles of each other. They accused the Oak Leaves station of injuring their lawfully acquired business property.

Chancellor Francis S. Wilson decided the case in the tradition of property rights to common resources. His landmark decision, which established homesteading rights in “the ether,” found precedent in western water rights, among other established property traditions.

Wilson concluded the court was "compelled to recognize rights which have been acquired by reason of the outlay and expenditure of money and the investment of time. . . . We are of the further opinion that, under the circumstances in this case, priority of time creates a superiority in right."

The Congress responded to Oak Leaves instantly. After years of debate and delay on a radio law, both houses jumped to pass a December 1926 resolution stating that no private rights to ether would be recognized as valid, mandating that broadcasters immediately sign waivers relinquishing all rights, and disclaiming any vested interests."

*Edit: This ownership wouldn't necessarily extend to _entire spectrums_ as it does in the case of the government selling spectrum. In Tribune Co. v. Oak Leaves Broadcasting Station, for instance, WGN was granted the right to broadcast on a specific frequency within a certain area, which differs immensely from a company being able to 'buy' a series of frequencies from the government such that no other entity in the country may uses this spectrum.

The government has sold the cellphone bands to private companies making your point moot.

Firstly, it's not 'my point'; I was presenting that argument as an answer to the parent post's question of how else interference could be prevented. Secondly, the ownership of entire cellphone bands isn't the only kind of property right over spectrum that could exist.

Ownership of entire cellphone bands would actually be somewhat unlikely. For a court following 'Tribune Co. v. Oak Leaves Broadcasting Station' to consider an entire spectrum across the entire nation as property in court, the transmitting organisation would have to show that it was and had been using that entire spectrum to transmit content across every part of the nation. Such a situation would be unlikely, as the organisation would likely have faced competition for spectrum use before it could expand its own spectrum use over the entire country.

The problem with the "treat spectrum as property" argument is that it's effectively what we have now. Make it even more of a free market and you would expect to see much as what you see with agricultural land in the US: A small handful of companies own the large majority. Put spectrum on the market and AT&T will buy it, and they have more money than you (for any value of "you" that is not a similarly sized corporation). You end up with the status quo.

The idea is it would have only come onto the market slowly as companies were able to progressively utilise more and more of the spectrum in various areas. Ownership hence would have been more diversified from the beginning, preventing the development of a continental monopoly capable of buying it all, similar to how no monopoly has been able to buy up all the country's land.

Agriculture is the US is incredibly subsidised, and as such things tend to benefit large incumbents the situation could be quite different if no subsidies took place.

> The idea is it would have only come onto the market slowly as companies were able to progressively utilise more and more of the spectrum in various areas. Ownership hence would have been more diversified from the beginning, preventing the development of a continental monopoly capable of buying it all, similar to how no monopoly has been able to buy up all the country's land.

But they have bought up all of the land. I mean not "all" as in 100% obviously (AT&T and Verizon don't own 100% of the spectrum either), but a small handful of large agribusiness corporations own the large majority of the farmland in the US.

And it doesn't really matter how it starts. A hundred years ago there were quite a lot of family farms that have since been subsumed into ConAgra et al.

Large corporations have economies of scale. All else equal, the rich get richer and large corporations get larger. Eventually they tend to become mismanaged and die, but not all at the same time, so their carcasses are eaten by their other large competitors, which makes them even larger.

Competition doesn't come from rational self-interest. Rational self-interest for powerful people and corporations is to get together and collude with each other. Competition comes from changing circumstances that modify who has power and create uncertainty as to who existing interests should collude with. People bet on different horses when they're not sure which is going to win. Nobody bets on the locally owned hardware store against Lowes and Home Depot.

I'm not American and not particularly familiar with the economics of the agriculture industry there, so I don't really feel qualified to comment. I'll just point out that government actions often favour such large corporations, preventing for instance the collapse of certain banks and auto manufacturers during the GFC. I do know that farm subsidies in the US are quite extensive.

If the government had decided 70 years ago that it owned all the nations farmland, and had auctioned it off to four large corporations, making it law that nobody else could own farmland, would the farmland situation would be better or worse than it is now? I'd argue for worse.

First off, by bringing up / presenting "a libertarian argument against government ownership of the airwaves" it becomes your point even if you don't agree with it. Also, so saying their not owned by the government does make your point moot.

At a more basic level, before technology improved the high frequency bands where directly worthless but that would not have stopped people squatting them just like with DNS. Secondly, unlike radio cellphones need to broadcast so you need a setup that does not break down as people move around the country thus at least the protocol needs to be national for a given frequency.* Simply giving a free for all to the first people to use a given spectrum is horribly inefficient and effectively just a first come first served auction.

*Radio and TV are examples of this where the protocol does not change even if different people are using the same frequency in different areas, but if it's unregulated they could in theory do whatever they want.

There's a difference between making an argument and presenting an argument; I was doing the latter, but an argument as a 'point' generally refers to the former. The 'point' is analogous to the point of a weapon; the most effective part. But that only makes sense in the context of an offense of some sort. (For reference, I do not in fact support that position, I just find it an interesting argument.)

If the high frequency bands were 'directly worthless' then people could not have 'squatted' them under the kind of judicial reasoning behind 'Tribune Co. v. Oak Leaves Broadcasting Station', as property was only granted for use. If someone was 'squatting' an airwave, that implies they're not using it for any significant business activities, and hence they couldn't be granted ownership.

There is no international regulation binding all countries in terms cellphone frequency standardisation, but that doesn't stop a phone bought in one country from working in others.

A 'first come first served' auction (or at least, 'first use first served') is how much of the property in the US originally came into ownership (excluding that acquired by violent conquest). The first person to strike oil owns it. The first person to discover gold owns it. The first person to patent an idea owns it. While this may not necessarily be efficient, in the current social context it's quite accepted, so if it's okay to acquire ownership of such resources this way then the same could apply to spectrum.

'There is no international regulation binding all countries in terms cellphone frequency standardisation, but that doesn't stop a phone bought in one country from working in others.' Countries have not standardized so cellphone in country A often does not work in country B. Often there is a choice where some of the cellphones sold in country A work in country B.

More importantly people are far less mobile between countries than they are between city's so the interference caused by non standard cellphones is minimal but still exists.

As to squatting, all it takes is setting up a radio station which is vary cheap. The issue fig leaf economic activity is inefficient, more importantly the having courts decide ownership is vary expensive for everyone involved.

PS: "point" references pointing such as with a finger. Bringing up an idea is often referred to as pointing out something even if no finger is involved. Swapping that which does the pointing to that which was pointed to is how you get from a 'something that is pricked' to a zero dimensional mathematical idea aka location.

Ah, I'd always assumed point was meant in the sense of 'pointy'.

I've never encountered trouble with cellphones traveling internationally, but I've only traveled in Asia/Oceania so I suppose that's not necessarily representative of the global situation.

Having the courts decide ownership may be expensive, but those costs should be weighed against the costs from having only 4 organisations owning pretty much the entire spectrum across the entire country, as is currently the case. If a court-based system resulted in for instance 20 companies now owning various parts of the spectrum and competing vigorously, the benefits of that might make up for any extra legal costs.

The economist Ronald Coase wrote a very interesting (and brief) history on the FCC, and makes a compelling argument for the use of auctions to grant rights to spectrums. While long, I think the article is well worth reading.

http://thesefragments.files.wordpress.com/2007/07/coase_fcc....

A) the government should have picked GSM or CDMA and mandated that.

B) they should have then enacted regulation to permit roaming for "reasonable" access fees.

This would have the net effect of not forcing 4 companies to build essentially the same network across the US. Showing to Europe is an eye opening experience. Land somewhere, get a local SIM, pop it in and you are good to go.

That would have precluded the development and deployment of LTE.

How so?

One way would be to have the infrastructure financed by both the government and the actors, and all actors using the same infrastructure, and therefore competing on prices and services instead of each of them having a separate monopoly on their band. That's exactly how it works for phone lines, and that's where competition worked the best to bring the prices down.

Without getting into the whole rabbit-hole of monopolies/government good vs evil... Do you see any realistic alternative to an auction allocation system for mobile frequencies?

Rather than have a tiny band for each company you could select a protocol for each band and let companies freely compete based on which has the lowest cost and a clear signal to each cellphone. Thus making it reasonable for 3rd party's to for example add a few towers near a convention center.

Unlicensed spectrum works extremely well for low power devices that don't cover long distances. If your roommate is using too much bandwidth you can hash it out together. The problem with using it for high power long distance frequencies is you get tragedy of the commons pretty much immediately.

It would be interesting to try for the midrange though. Provide unlicensed spectrum at power levels that could cover e.g. a street rather than a city. Then you could get very local ISPs: One person on your street bites to bullet to buy a fast leased line for $$$ and then sells internet access to all the neighbors over wireless.

Maybe lottery? Though unless you had some restricitions on entry that might just result in the winners reselling their allocations to the established players for a profit.

Auctions are a superior solution to lotteries from an economic theoretical point of view. No point giving spectrum to a company that can't use it properly.

Lotteries have been tried, and yes, it did lead to all kinds of work being done to be able to enter the lottery multiple times, for people to enter the lottery just to resell, and other similar shenanigans. More here: http://www.nytimes.com/2013/06/05/business/in-wireless-spect...

So a lottery quickly becomes a special kind of auction where lower bidders still have some chance to win.

The mobile sector is a great example of a natural monopoly:

http://en.wikipedia.org/wiki/Natural_monopoly

In this case, competition is the problem not the solution. It makes no sense to build out massive cellular networks in triplicate.

Connectivity could be provided much more efficiently at the municipal level, like water. Certainly, my experience with city water has been orders of magnitude more pleasant that with mobile providers, many times cheaper and more reliable, and I really don't see one as being harder to provide than the other.

> In this case, competition is the problem not the solution. It makes no sense to build out massive cellular networks in triplicate.

I agree with you, but I also think that it makes no sense to grant access to the airwaves at the same time as granting a monopoly to all services that might be performed over those airwaves. I think that this effectively extends the monopoly into other areas that are not natural.

In the UK, the incumbent telecoms company (BT) is required by the regulator to provide access to its lines to companies that compete with them. The (regulatory) agreed pricing is relatively close to BT's own cost structure (95 percentile billing, etc). This is just for the part that is the natural monopoly - primarily the last mile. The competition provide their own Internet backhaul, and sometimes the links to the exchanges too. I think this works remarkably well - the market has real competition in providing all of the other parts of Internet service (customer service, peering quality, how different ISPs end up managing their bandwidth, etc).

With no history, I think it might be even better to limit companies granted rights to the natural monopoly area to only providing services to other businesses for access only. For example: a company granted access to cellphone frequencies might only be permitted to provide transit to other companies, rather than to retail customers directly. I think that by replacing the irrational retail customers with more rational businesses, we'd get better efficiency, real competition and prices closer to real cost.

Why does it make sense to build out massive supermarket chains in triplicate?

The capital costs of supermarkets are much, much lower relative to their operating costs.

However, using that context: contrast supermarkets with the roads that are used by both their suppliers and their consumers, and you have another good example of a capital-intesive infrastructure where redundancy is inefficient and therefore is better supplied by government.

Completely agree. I stopped using contract wireless services since a few years ago and have been able to consistently pay less and get more than before.

Currently using AT&T GoPhone's new $45/30days "Walmart" plan with unlimited talk/text and 1GB of data. Its much less than on contract. Prices are falling, you just have to look beyond a contract.

Oh wow, that is expensive. Here in the UK on prepay Three offer 25GB/30 days (marketed as "unlimited") for $25 (£15). Bonus: You get LTE where available for no extra charge.

Government regulation doesn't create competition.

Edit: It's remarkable that any statement regarding regulation that isn't overtly Marxist immediately gets half a dozen down votes on HN.

Except when it does? There are like, plenty of examples of government regulation creating very real competition? For example, there is fairly decent competition in the ISP market of the United Kingdom thanks to the Enterprise Act 2002 that requires British Telecom to share its local infrastructure with its competitors. Otherwise, it would be like in many places in America: you can choose Comcast, or if you feel really adventurous and you really thought it through, you can choose Comcast too.

It's government that is granting Comcast the monopoly privilege to be the only ISP on the market. That is not the natural state of a market. It's government intervention in the first place that creates the monopoly.

> It's government that is granting Comcast the monopoly privilege to be the only ISP on the market.

Not always. Comcast may have got in early and laid substantial infrastructure (copper/fibre/HFC) - something that can be substantially cheaper if you leverage "new" (cable reserves and/or piggy-back off the utilities when they are expanding their networks.

New players, however, need a substantial amount of capital to get started. Comcast (or anyone like them) can leverage their position in the market to drive up prices or push smaller players out. By using regulation to effectively turn the infrastructure into a public utility, it allows smaller players to compete.

The argument against this regulation (which is in place in the UK and Australia, amongst other countries) is that the monopoly (BT/Telstra) spent money building out their network and therefore should not be forced to share it.

The argument for this regulation is that often the telco built out this infrastructure in a particularly favorable environment that cannot be replicated today due to the huge costs of installing physical cable. As a further example, Telstra in Australia (previously Telecom) was government owned and therefore had a directive to supply the population with telecommunications. Many parts of their copper network (cable, conduit, pits, exchanges) were built during this phase. When they were privatised (sold by the Government) this was no longer their key directive. Any new player attempting to get into the market-even a small part (i.e. inner Sydney)-had a lot of catching up to do. Huge capital costs, a significant amount of regulatory hurdles and cable reserves to negotiate access to, etc.

This is of course a simplification of the issue, and you could argue that in Telstra's case the government did effectively grant them a monopoly on the market by privatising them. But had they started out as a private company originally, I believe we would have still ended up in a similar place - getting in early when rolling out disruptive infrastructure (as trenching cable tends to be!) projects is a huge advantage in and of itself.

Hey. Great example.

I think you're correct to point out that "in Telstra's case the government did effectively grant them a monopoly" in the first place.

The specifics I'm talking about in the Comcast case is that government is continually granting the monopoly privilege by preventing other companies from creating infrastructure (laying wire), even if a company wishes to do so.

This is precisely what allows Comcast to charge whatever it wants, begging more regulation down the road to force Comcast to charge 'fair' prices ('fair' in the minds of the legislators).

If any competitor was allowed to enter what is an enormous market it would provide continual downward pressure on consumer prices. The monopolistic privilege granted to Comcast leads to higher prices, as consumers can only pay the high price or forego service.

How so? It certainly seems like it's a combination of high cost of entry and Comcast's ability to take a regional loss to drive out other competitors that's granting them a defacto monopoly. Also known as market forces.

Those cable lines that run down the streets into all the houses. By law nobody can use them but Comcast and by law nobody else can string up another set of lines to run to all the houses. That's the monopoly privilege.

Adam Smith described how free markets can deliver socially beneficial results when individuals and firms compete vigorously with each other.

Since when does ensuring vigorous competition equate to Marxism? Seems pretty capitalist to me.

Regulatory intervention into a marketplace is not "ensuring vigorous competition."

Markets are vigorously competitive in their natural state. This is the nature of a market, where firms compete for customers' business. This can only be hindered with regulatory interference where a legislator is arbitrarily granting privileges to certain firms and not others. That's crony capitalism or mercantilism. It's not a free market.

Free markets don't exist without governments. They don't have a natural state. A government is a monopoly on force that enforces property rights and contracts. If there was no government then you would have no market incentive to do anything because you would have no protection from theft or breach of contract. The contours of what you can own as property and what you can enforce in a contract are necessarily defined by the government.

The problem with idealized free markets is that "free" is not the natural state of a market. You can act as a de facto government by having a monopoly on something other than force (like land or infrastructure), and de facto government actions have all the same defects as de jure government actions, generally with none of the public accountability.

That is the definition of antitrust regulation.

No it's not. There is no definition. It's all up to the interpretation of the politicians who wish to pursue it and the opinion and action of the judges on each case. Remember the Microsoft case?

Can you name one monopoly that exists anywhere that is not the creation of a government in the first place?

Do some digging into the history of antitrust legislation. The Sherman Antitrust legislation was an act of mercantilism designed to protect incompetent businesses who were losing market share to superior competition, most notably Standard Oil, the poster boy monopoly which gained an 85 percent market share through lower prices and economies of scale, delivering a benefit to consumers but harming their comparably inefficient competitors. It was those same competitors who spearheaded the antitrust legislation, not the poor consumers 'suffering' from lower prices and superior service at the hands of the 'monopolistic' Standard Oil.

"Can you name one monopoly that exists anywhere that is not the creation of a government in the first place?"

Microsoft.

I'm a libertarian, but one of the reasons I can't even consider being an anarchist is that I consider anti-trust regulation to be a good and proper function of a government. I'd break up a lot more, personally. It's one thing that I find the otherwise-excitable class warriors to be bizarrely reticent to do.

I'd also observe that antitrust regulations do not forbid monopolies... they forbid the abuse of monopolies, according to a certain definition of abuse. Microsoft was not punished for having a dominant desktop OS, they were punished for using that dominance to engage in shenanigans to try to "abusively" (in accordance with the legal definition) extend that dominance into a another sphere. Of course that is the theory, and what happens in the courtroom in fact may be arbitrarily related to the theory, but that is the theory.

I certainly suffered under that monopoly (as did the entire industry) but couldn't the argument be made that market forces took care of it without government intervention?

I'm aware that there was an anti-trust judgement against them, but as far as I can tell it didn't deter them in any significant way.

It seems to me that they simply collapsed under their own hubris and the blowback from their decades of abusive practices, and as soon as they weakened the competition ate their lunch.

"I certainly suffered under that monopoly (as did the entire industry) but couldn't the argument be made that market forces took care of it without government intervention?"

YMMV vary, but my conclusion is this: The government successfully prevented Microsoft from crushing the rest of the browsers. They really were trying, really did damage Netscape, and really were well on their way to more dominance in that field than they had a "right" too. And they were on their way, too; Netscape was faltering and IE was a better browser for a good long time. As it is they still had enough dominance to end up causing web tech to stagnate badly for a few years.

However, recall why Microsoft was so scared of the browser, which is that it threatened to become its own OS and make it so you could run a computer without Windows that could do everything you needed. Broadly speaking (bear with me here), this still has not happened. Here in 2014, you could run a small business out of a browser, and you can do a lot with a Chromebook, but Windows is still around, still powerful, and still pretty full of legacy software that doesn't exist in a browser and businesses are not having an easy time disposing of... witness the continuing lurching life of Windows XP, which Microsoft wants to be rid of and can't be.

I think it was a good decision and still correct, even if ultimately it really only served as a very large shot across Microsoft's bow (even if it was aimed at the hull) due to the fact that the browsers of the time weren't going to be able to manifest the promise of displacing Windows. We'll never know, but the secondary effects of chastising Microsoft may have been worth more than the primary effects ever could be. And I do think the secondary effects were more important; the direct impact pales in comparison to the fact that they'd just been notified they couldn't proceed down that path any farther than they'd already gone without the costs exceeding the benefits.

Of course, what truly displaced Microsoft turned out to be more Apple, via the MacOSX and mobile, which Microsoft continues to struggle with. But so may years ago, who'da thunk it? And it would have been silly for the government to count on that.

Not the way I remember it. Netscape owned the browser market. Microsoft didn't even have a dog in the fight (remember they were "blindsided" by the internet). You signed up for internet access with Earthlink or Juno or some other company like that and they sent you a CD with a free copy of Netscape.

Microsoft licensed the Mosaic browser from Spyglass and that became IE. As you pointed out, they developed IE until it was a better browser. The Netscape stagnation happened when they had a long delay in releases due to their undertaking a "thing you should never do [1]," a from-scratch rewrite. Around that time the 2000/2001 dot-com collapse happened which certainly didn't help, but Microsoft didn't cause any of that.

Fast-forward a decade, and now Microsoft is stagnant. Apple developed WebKit and Safari, and Google came along with Chrome, a better browser. By the time Microsoft worked its way through IE 6-7-8 to a browser that was actually competitive again, they had lost a lot of their browser share. Apple, with Safari, and Firefox, rising from the ashes of Netscape, also offered compelling alternatives.

"Who'da thunk it." Well I'm not sure. I don't see any evidence that the government has any better thinkers than the companies in the tech sector though.

1: http://www.joelonsoftware.com/articles/fog0000000069.html

IE 4 was a better browser than Netscape. I think we may finally be distant enough that we can drop our allegiances and just admit that. Netscape did not randomly decide to drop everything and rewrite a browser, they did it because they were trapped in a terrible, terrible code base that could not easily be extended to grow into the future. Netscape "layers", their answer to what we at the time called DHTML and today don't even have a name for it because it's just how the web works, were atrocious, and it was directly a result of their code base not being able to do anything else.

Netscape was boned either way... not doing a from-scratch rewrite would still have doomed them to being stuck behind Microsoft for a long time. It would have allowed them to keep making releases instead of just going silent, but they still would not have been able to be as good as IE. And they'd have still be stuck hard by the fact that Microsoft monopoly'd the price of a browser down to 0... and Microsoft would probably have monolopy'd the price of a server down to 0 too even faster than it did, given the chance.

Microsoft was "stagnant" because they could be stagnant because they had won. The judgment at least prevented them from pushing home the advantage, because they knew they'd be slapped down. This is what I'm saying was probably the most important thing about it, and it's easy not to see what "didn't happen", but I suspect that Microsoft would have done yet more "evil". Perhaps it would have gone poorly, but it would have been foolishness for anybody to count on that. (What we know now about Microsoft and Ballmer's leadership makes that a safer bet, probably, but we didn't know that at the time!)

I think people look back at the world in which Microsoft was slapped, and see a world where bad things didn't happen, and don't realize that, due to second-order effects, there's probably more relationship between those two things than they realize, even if obvious first-order effects are missing (like, Bill Gates never rent his clothes in twain on national TV going "Woe are us, for we are injunction'ed!"). In the end Microsoft's dominance would have been cracked sooner or later, sure, but in the long run we're all dead; no policies can be written based on that theory.

Microsoft exists due to intellectual monopoly (copyright) granted by government. Copyright wouldn't exist in free market because it conflicts with property rights.

Hard to prove, hard to disprove. Even without IP the sheer staggering output of such a large and at the time, effective corporation would not have been easy to copy in any reasonable manner. Enterprises certainly would not have trusted their support to people without the source code. Lack of IP, after all, wouldn't mandate that Microsoft make their source available externally, and without that, the knockoffs aren't going to have a lot of luck doing anything but blind aping. Consumers might have had some copy-based knock-offs available, but frankly, they probably would have always had reliability problems and Microsoft would have made efforts to make those problems worse, many of which would have worked well enough to probably make the problem mostly go away.

I can't disprove that Microsoft depended entirely on IP, but I'd suggest that it's at least a reasonable theory that it really didn't.

I've read multiple biographies of John Rockefeller and Standard Oil engaged in a variety of anticompetitive practices such as operating at a loss when competitors emerged, or withholding lubricant shipments to railroads during rate negotiations. That's just what I can remember off the top of my head.

That Standard Oil succeeded just through scale is a complete fabrication.

Government intervention is practically required to allow competition. Any market left to its own devices will devolve into a monopoly. Governments are required to regulate to avoid monopolies thereby preserving a level playing field on which competition can occur.

When Adam Smith was talking about a "Free market", the freedom he was talking about was freedom to enter and freedom to exit a market. That is, freedom from barriers to market. These freedoms are guaranteed by governments. Businesses left to themselves will raise the barriers to market thereby making the market non-free in Smith's terms.

Oh, and pointing this out is not "Marxist".

The Marxist term was misused in this case, you're right. I was just making a general comment about a perceived anti-capitalist bias in these forums.

"Any market left to its own devices will devolve into a monopoly." This is not an observable phenomenon. A market left to its own devices will continually deliver better, cheaper products to its customers over time. This is the effect competition has on a marketplace.

"Businesses left to themselves will raise the barriers to market thereby making the market non-free in Smith's terms." If a business is not using physical force or fraud then the only barriers they can erect to keep out competitors are superior offerings in the marketplace.

Microsoft is such a great example here. Their competitors were crying monopoly but Microsoft had created its marketshare through innovation and entrepreneurship. If they abuse this marketshare by offering inferior products at high prices, it creates a market opportunity for another competitor to create better products at more attractive prices. In this sense it's the free consumer choice that's providing the "regulation" in the market. Businesses have to follow the consumer's wants to gain marketshare.

The only government "regulatory" role in a market is to settle disputes about property or fraud or criminal behavior. With that basic structure in place, Adam Smiths's "invisible hand" would take care of the rest.

If you think using regulation to keep markets functioning is a Marxist, don't complain about down votes until you learn some basic economics and history from sources other than industry propagandists pretending to be libertarians.

You're right. Regulation is not explicitly Marxist. My edit was born of a frustration with the general bias in this community in which advocating for capitalism or free markets is guaranteed to earn you down votes.

You are making the false assumption that capitalism = free market.

One might argue that after taking into account historical context, modern capitalism is completely opposite to free markets. For example, in a free market, labour unions would be as powerful as the owners of capital - in the modern world, capital owners have a disproportionate influence on law-making to subdue this power. A very good example in the recent past is the wage collusion by Apple, Google and many other companies (working to preserve the interests of the capital owners) to suppress the power of the free market.

> You are making the false assumption that capitalism = free market.

anarco-capitalism = free market

crony capitalism != free market

I use the term capitalism to describe the anarco flavor. Capitalism is defined as the private ownership of production, and when this idea is adhered to rigidly you end up with something exactly like or closely approximating the anarco-capitalist flavor.

> in the modern world, capital owners have a disproportionate influence on law-making to subdue this power.

You're right. It's abhorrent but this is not capitalism, this is a mixed economy (state capitalism or crony capitalism) in which firms use government to buy influence over the marketplace. I am not advocating for that.

Again, you're being down voted because you're dishonestly redefining those terms – HN crowd leans heavily capitalist, as you'd expect in a community started by people going to invest or found successful companies.

You weren't advocating for free markets, you are pretending monopolies don't exist except where created by the government. That's simply ahistorical nonsense. Calling people Marxist for quoting Adam Smith would be an amusing troll if you didn't sound like you actually believe it.

> Calling people Marxist for quoting Adam Smith would be an amusing troll if you didn't sound like you actually believe it.

It's not as big of a stretch as it might sound as first; there a number of ways in which Smith was more like Marx than like a modern capitalist (and plenty of ways in which he was far from either, particularly his identification of the feudal landed aristocracy -- explicitly and particularly as opposed to the mercantile/capitalist class -- as the class whose interests were most naturally aligned with the common interest. (Though one who shared Smith's concerns about the mercantile class might see Marx's investment in the proletariat as the only solution given the demonstrated failure of the landed aristocracy, as such, as a viable class in the face of capitalism, so even that view could be seen as less incompatible with Marxism than with modern capitalism.)

To be clear, its ridiculous to call some one a Marxist just because they quote Smith, but not as ridiculous as it might seem from the naive association of Smith with capitalism and Marx with its opposition.

For clarity: all of the comments citing Adam Smith actually showed up later, after I had made a comment about Marx. So the Marx comment was not in response to Adam Smith quotes.

> you are pretending monopolies don't exist except where created by the government

There is a strong case for this view, and it is shared by many well-regarded economists throughout history.

And I haven't redefined the terms: http://wiki.mises.org/wiki/Monopoly

> There is a strong case for this view, and it is shared by many well-regarded economists throughout history.

Citation needed

> And I haven't redefined the terms: http://wiki.mises.org/wiki/Monopoly

Even ignoring the question using a heavily-biased resource, that page would not need to mention “government-sanctioned monopolies” if there were in fact no other kind.

The definition of "monopoly" in terms of government sanction is very common (its one of two common legal definitions); of course, the other common legal definition (and the more common economic definition) is " the ownership or control of so large a part of the market- supply or output of a given commodity as to stitle competition, restrict the freedom of commerce, and give the monopolist control over prices." [1]

There's obviously a pretty fundamental problem with any discussion of economics which limits itself to the former kind of monopoly and fails to at least discuss the latter kind (even if it uses a different name for it for some reason). And there's equally obviously a problem with any discussion of monopoly in economics which ignores that the use of the term in economics usually means the latter, not the former (while in law, "monopoly" without other qualifications in certain contexts does mean the former, though in other contexts -- e.g., anti-trust, it means the latter.)

[1] Both definitions illustrated here: http://thelawdictionary.org/monopoly/

Here's Hayek and Friedman on monopoly:

http://principlesnotmen.wordpress.com/2013/02/09/does-the-fr...

Rothbard, Sowell, Mises and others share this view. Google their names followed by the word "monopoly" for an abundance of material.

http://wiki.mises.org/wiki/Monopoly

"A monopoly is a grant of special privilege by the State, reserving a certain area of production to one particular individual or group." in bold in the middle of the page.

If I can't quote a university that advocates for the Austrian view without you labeling it bias, there's no point to this discussion.

You accused me of redefining monopoly. I point to an entire school of economics that defines monopoly as I defined it. You dismiss it as bias.

carriers dont deliver "10mbits" which becomes "15mbits" with LTE.

LTE, GSM, etc. use "radio bandwidth". Thats waves in the air if you will. The faster the wave oscillate, the higher the frequency and the shorter the range at which it will keep oscillating.

Radio bandwidth is a finite element and thus frequency ranges are assigned to operators, to wifi, to your microwave (yes it emits on 2.4ghz and interfere with wifi in fact), to your TV, to handled radios, to your drone remote control, etc.

To give an example, an HSDPA channel is 5mhz. In these 5mhz you may be able to send up to X mbits depending on how the signal is encoded (how the signal is encoded is actually what provides the spectral efficiency).

Recent advances, in particular OFDM allowed us to transmit more data for the same radio bandwidth (aka frequency range). Variants of OFDM are used for xDSL, Wifi (not all standards) and LTE (and many other things, including that HD video feed from sport cars or the ISS).

So, since the milkshake's straw does not and CANNOT change size (again: radio bandwidth - spectrum - frequency range) are we are bound to the laws of physic. We're more efficient in the way we organize the data (and mainly, how we're not losing the data - thats the main advantage of OFDM - if you're interested in how it exactly works, wikipedia)

I really despise articles with sensationalistic headlines, that attempt to carry a point through misleading or even incorrect statements - even if often the result of genuine ignorance.

Also, i'm in the USA and I pay 30/mo for 5GB LTE, unlimited DC-HSDPA, unlimited text, 100m calls. It's not great (calls should be unlimited) but thats a good enough deal. If you're paying more you should consider dropping verizon/att - not blaming milkshakes.

>So, since the milkshake's straw does not and CANNOT change size

The reason you get this impossibility is because you make the wrong assumptions. You assume that there is one channel (for a very specific definition of channel), which everyone has to share.

In reality this is just not true. By using directed antennas for instance we can create more channels. By using polarized waves we can create more channels.

you're not adding more spectrum - my point - its a finite resource. you can also emit lower power and add more cells since they wont interfere.

but theres only so many details you can put in a comment post without making it 20 pages.

But the details matter; you don't even need more finely grained directional antennas to get more bandwidth: send two different signals from two different antennas and receive them with two antennas; with the data you get and certain encoding schemes, you can work out the two underlying superpositioned waves on a DSP chip and nearly double your bandwidth while using the same spectrum, thus working around Nyquist and adding another "straw":

http://en.wikipedia.org/wiki/MIMO#Sampling_theory_in_MIMO_sy...

But it is very important! Because you could have two devices communicating using many channels on the same spectrum. This removes the theoretic Nyquist rate bound.

call me when the 2 waves are sent on the same frequency at the same moment (ie same period). it doesnt work. mimo doesnt double the spectrum. it sends on the same spectrum.

so "but blah!"

I'm not sure we'll be able to call you, it sounds like your phone works on different principles to ours!

or you just called 2 friends to downvote by 2 as its an old topic lol. sad that most discussion forums end up with such conversations.

you cannot send then at the same exact time because theyre the same particules of matter. you can send them at slightly different times, angles, etc. still doesnt make the straw bigger.

I pay $70/month for TMobile with unlimited LTE. My coverage isn't as good as ATT while traveling, but its fine where I live. Is this a sign of "gouging?" That it costs more for a service with good coverage in the less populated parts of the sprawling US?

Why are 4G prices higher in US? Demand is higher. Supply is more costly to build. Econ 101.

> That it costs more for a service with good coverage in the less populated parts of the sprawling US?

Did you read the part of the article where they compare to Canada which is paying less than us for more bandwidth? Compare population density there with here.

It's not as simple as supply and demand either. If your infrastructure is being utilized heavily (efficiently), therefore delivering your more customers for the same equipment compared to another region, then you come out way ahead.

This article was about how supply is actually also getting cheaper. LTE base stations get more out of fixed bandwidth allocations, and cost less as well.

Perhaps you can also tell us what Econ 101 has to say about oligopolies, high barrier to entry and natural monopolies.

Oligopolies are likely to have some price effect.

Although I'm not looking to switch anytime soon, may I ask who your carrier is? I use Republic Wireless right now, who was mentioned in the article, but that's a damn good price for 5GB LTE.

I'm always trying to convince my friends to look into MVNOs over the major brands.

t-mobile pay as you go - you have to order the SIM online.

there's also decent MVNOs but none offer LTE AFAIK. (well, maybe some do nowadays, but im not aware of them)

source: http://prepaid-phones.t-mobile.com/prepaid-plans

scroll down to "more plan options"

Also thats what ive been using for a while, i think its been there since a year or so.

My Verizon contract is up and I am really thinking about moving to Republic Wireless. The Moto G offers as much if not more than I want in a phone (I have a simple flip feature phone now) and I like that it uses wifi for voice which would make it feasible for me to use my phone at home (I live on the very fringe of cell coverage, and voice calls tend to have a lot of dropouts). Are there any other services I should consider?

When I decided to switch, I started here:

https://en.wikipedia.org/wiki/List_of_United_States_mobile_v...

and compared a couple dozen of them. My previous carrier was Virgin Mobile. The service is good, but they won't let you use a smartphone without a data plan, and their phone selection sucks. Republic has limited options too, but I wanted the moto x anyways.

With multiple sending and receiving antennas and spatial encoding (specialization of MIMO) you can receive and send more data over the same spectrum and basically just make use of a second (or third) "straw".

I've gone without phone or data service on my cell for a couple years, getting by entirely on Wi-Fi, and using Google Voice for the handful of occasions when I needed to call someone.

When I signed back up, I was happy to get unlimited data again, for less money than I'd been paying a few years ago. Back then, it was extra for a data plan, and extra for a "smartphone" data plan -- making my bill over $100. Now, I'm paying less than I was after corporate discounts were applied to my old plan.

And mobile data, even just 3G, is fast. And often I'm upshifted to HSPA or 4G, which is faster than the Wi-Fi in some bars. Plus apps and websites are built to use less data than a couple years ago, using AJAX and other methods to cache and reuse more data on the device, such as UI elements and layout. The app is just the mobile browser page, and that is a client that calls an API, minimizing repeated network calls.

ProTip: I just called AT&T today and asked when my contact ended, because I was considering going to T-Mobile. A few minutes later they offered me the approximately the service for half the cost.

Indeed. Prices _have_ come way down, but the carriers are engaging in price discrimination (quite rationally). You don't get the new prices automatically, you have to switch carriers/plans or persuade your current carrier that you are going to do so.

On the other hand, such price discrimination reminds me of things like this:

http://blogs.computerworld.com/node/3538

I'm certainly not saying discriminatory pricing is entirely fair :-) But the article's premise is that costs haven't come down, and that simply isn't true. (Admittedly the reasons likely have more to do with T-Mobile's strategy than technological shifts.)

The difference between wireless technology and computing technology is that wireless communication is limited by Shannon's Law which basically says that C ~= 1.44 * B(S/IN) where C is the information rate (i.e. bps) B is the spectral width (how much RF spectrum is being used to modulate the signal onto, and S/I*N is the signal to interference and noise ratio. Right now, LTE is very close to the Shannon limit, there is no Moore's law we are all familiar with. Most carriers face a choice of either obtaining more spectrum (which is limited resource and expensive) or increasing the SINR which is very expensive to do since it involves building more cell sites, if the tower is closer to the user the SINR goes up. I think often times the cost of maintaining the infrastructure around supporting tens of thousands of cell sites is under estimated. Yes the wireless companies are profitable, but no where near the kinds of profits the phone manufacturers are reaping where it costs them $50 to make a phone and charge the cell companies $500 for the phone, that huge mark up is passed along to the customer as part of their monthly bill because everyone expects their phone to be "free". It is no free, the cost is embedded in your monthly bill. The best way to drive down wireless costs in my view are to assert competitive pressures on the phone manufacturers.

The blogger is full of crap. 3G base stations don't automatically upgrade themselves to 4G. It costs a fortune to upgrade each tower. The backhaul also has to be upgraded to cope with the extra bandwidth that the tower can cope with due to the increased spectrum efficiency (more data in the same frequency space due to SCIENCE!).

Can you expand on why it costs so much to upgrade a basestation? the fiber is already rolled out, the contracts between land owners have already been signed, the surveys for radio propagation have already been done. I honestly don't think the upgrade from 3g to LTE was nearly as expensive as the upgrade from EDGE/2g networks to 3g was for carriers. Also it was much, much quicker for them to roll out LTE networks vs 3g networks.

I'd love to be a cynic, but my family's AT&T bill went down when we got on their new family plan back in February. And now we have way more data than we can use. I can only assume it was due to competition in the market in family plans.

My AT&T bill dropped by 25% just a few months ago.

This article seems like it was written by someone who has little if any knowledge about cellular tech.

This would be a better comment if, instead of just putting down the author, you taught us some of the relevant things you know.

Yes. Embarassing.

I would love to live in a world where increases in RAN efficiences doesn't result in increased backhaul demands.

The carriers are screwing us because the Internet Is MAGIC!

Completely irrelevant but I can't help myself given the milkshake example provided in the article.

Milkshake scene from the movie "There Will be Blood": http://youtu.be/s_hFTR6qyEo?t=1m20s

Actually in Japan 4G plans are cheaper, presumably because the 3G networks are so oversubscribed that carriers are trying to get people to swap (and buy new phones in the process).

You really need competition on your networks. In Denmark the carriers are forced to rent out their networks to competitors so that we don't get a monopoly.

Currently i pay ~37$ for unlimited speech, unlimited sms/mms and 10gb data a month. Oh yea and no roaming charges in sweden, italy, UK, ireland, hongkong and austria. The list of countries might grow since the european union is considering removing roaming charges completely within EU.

Meanwhile, in "socialist hell-hole" Sweden, I have an unlimited calls, texts and 4G data plan with these speeds http://i.imgur.com/62yT2ig.png (45Mbit down, 7.5Mbit up)

It costs me ~$48 per month. I average around 10-12gb per month, but don't have to worry about any charges when tethering my laptop to the 4G.

And for what it's worth in even more socialist hell hole Vietnam I got a prepaid SIM with voice and 2GB of data for about US$9. Only 3G sadly enough but fast enough.

There are two first would countries in the world comparable to the United States, and one of them is only an almost.

Australia - Widely dispersed population, universal coverage hard to build out, environmental factors. Unlimited Voice/SMS/MMS 2.5gig data allotment - 121.73 USD (Telstra)

Canada - Population in a small band near the border but many rural areas must be covered by mandate, universal coverage expensive to build out, environmental factors. Unlimited Voice/SMS/MMS 2 gig data allotment - 82.61 USD (Rogers Communications)

United States - Population widely Scattered, many rural areas without significant infrastructure improvement raising construction cost for bearer facilities hence huge construction costs in 4g rollout. Unlimited Voice/SMS/MMS 2 gig data allotment - 75 USD (Verizon Wireless)

Beyond the above - there has never been a time in the last 10 years where there was not a major network rollout in progress by the major carriers in the US, first they built the digital networks (96-2001) then the 2.5g rollout (EV-DO/EDGE (2001-2007) then the 3g Rollout (EV-DO Rev A and HSDPA) (2007 - 2009) now the 4g Rollout (WiMax then LTE) (2009 to Current). This has been the case world wide - but because of the geographical features of the US (higher construction costs and lower density) our telecom costs will always be higher than Europe and Asia.

Finland - Unlimited voice & SMS/MMS + uncapped data at 2MBit/s: 35 USD/month, but that's without a contract. If you're ok sticking with the same provider for two years, you can get the same with 50MBit/s uncapped data. On the other hand, if you're fine with 256 kbit/s uncapped data and paying 0.10 USD per minute or SMS, it's just 4 USD/month.

Our population density is on average half that of the USA (16 people/km2 rather than 32 people/km2), and we're just as dispersed. The coverage of the above-quoted provider you can see here: http://elisa.fi/kuuluvuus/

Explaining the high cost of US mobile comms in terms of the situation in Canada & Australia doesn't work, all you're really pointing out is that it's far too expensive there as well. In other words, your friends are also being fleeced.

Let me explain to you the difference between median and mean. Finland has an average population density that is half that of the USA but mostly because the population is heavily concentrated into six or eight islands around urban areas and the rest of the country has a very low population density. To serve 80% of the population of Finland I would need to provide service to less than 20% of its area. The US has many more urban centers, swaths of very low population density that are larger than most of western europe, and a high-density cluster along the eastern seaboard that would be the equivalent of providing Helsinki-level infrastructure over the entirety of Finland.

Not even close to similar.

Ummm, did you even look at the coverage map I linked to? For the sake of completeness, here are the other two national networks:

http://www.sonera.fi/etsi+apua+ja+tukea/verkkokartat/peittoa... http://kuuluvuus.dna.fi/Peittokartta/

My point here is that each of the three has nearly complete coverage of the whole of Finland. Sure, faster data than EDGE isn't available everywhere, but there's still service. And given that we're offering to these service providers only up to 5.5 million theoretical customers in total, they still manage to make a profit at the rates they charge.

And somehow the situation in the US is so radically different that even with your economies of scale customers are charged perhaps triple compared to here? I'd say explaining how it works out that way calls for extraordinary evidence, and I've yet to come across such evidence.

I did.

On both of those maps LTE coverage covers but a fraction of the country, Verizon has rolled out LTE to about 90% of its network footprint (http://media.idownloadblog.com/wp-content/uploads/2013/05/Ve...) and they are still in the midst of an active deployment - the sheer cost of this deployement is staggering - the figures I know, for the Samsung LTE hardware - just the hardware alone for a mid sized market, just 700 or so sites costs around 70 million dollars - thats ahead of installation/construction costs - which I'm estimating at somewhere between 100-150k a pop (assuming no structural mods are needed, and there are no construction charges to bring in new fiber backhaul) - figure the US has around 200,000 cell sites - I cant find out how many finland has - but just on the size 104k sq miles (about the size of colorado) I'd estimate no more than 2000 sites - probably less.

Like a year ago I wanted to see if there was a way to get an unlocked sim not tied to a carrier and just use data. Reason being, I wanted to see how much I was over paying for voice [I have 1000s of unused minutes]. I wasn't able to find a data only sim option.

But by a base line calculation I'm spending ~$300+/year over the cost of VOIP.

This article is seriously flawed. Yes, 4G transmission is more efficient for the allotted wireless spectrum, but users are guaranteed to be using more bandwidth, so the effective number of users per tower stays the same or decreases. And it costs more, not less for 4G rollouts.

Each tower supports more voice channels, and given than voice has higher priority than data and most people use mobile phones for calling, then each tower does support more subscribers. In Cambodia, prices have been dropping - 3.5GB on 3.75/4G networks costs $5/month. The competition between mobile companies is intense, and Cambodia seems to be leading

If users are on the same limited plans they're not using more bandwidth over any appreciable time scale.

I'm not talking about data use, I'm talking about instantaneous data rate. Base stations are designed to handle a particular maximum throughput for a given number of users relative to the available spectrum in that area. The maximum throughput needed for many users on a 4G tower will be much higher than it is for 3G, it isn't the same or less.

Again, this is only true on very short time intervals. If you're an AT&T user throttled after 2GB per month you simply aren't going to be using 4G speed for more than a single coffee break per month.

In the real world, of course, people rarely download data endlessly and so the main difference is that a 4G user will use data at a higher rate for a shorter period of time since they're downloading the same YouTube video as the 3G user.

Since 4G takes considerably less time to establish connections than 3G, this trend is even more favorable to network operators since more phones can go from passive to active and back in any given time interval.

> Cellular plans in the U.S. with 500MB of data cost about $85.

I highly doubt that. The piece links to [1] which in turn cites [2].

In [2] "Mobile-broadband prices are collected from the operator with the largest market share in the country", "prices applying to the largest city (in terms of population) or to the capital city are used", and a the 500MB plans are ones that include voice and text.

Moreover, the reported prices include taxes; while including taxes is necessary to accurately reflect the cost to consumers, it does not give insight into weather or not carriers are "gouging" consumers, which is what the RWW piece alleges.

AT&T charges $65 for unlimited talk and text and 2GB of data [3], Verizon $60 with 1GB [4], T-Mobile $40 with 500MB [5], Sprint $55 with 1GB [6].

I won't get into prepaid plans and/or MVNOs exhaustively, but for instance, Aio (AT&T's own MVNO) charges $35 for unlimited talk and text with 500MB of unthrottled data if you enable auto-pay [7].

According to a 2011 report [8], the average tax rate is 16.26%, which would still only bring the most expensive plan I mentioned (AT&T's) to $75.57.

Based on that, unless there is a very large number of consumers on old plans that have not been brought down to current prices (and I'm not sure if that even happens), I highly doubt that $85 figure.

[1] http://www.economist.com/blogs/graphicdetail/2013/10/daily-c... [2] http://www.itu.int/en/ITU-D/Statistics/Documents/publication... (page 112) [3] https://www.att.com/shop/wireless/data-plans.html [4] https://www.verizonwireless.com/wcms/consumer/shop/shop-data... [5] http://www.t-mobile.com/cell-phone-plans/individual.html [6] https://www.sprint.com/landings/framily/?INTNAV=ATG:HE:Frami... [7] https://www.aiowireless.com/shop/plans.html [8] http://taxfoundation.org/article/states-target-cell-phones-s...

I get 5GB of 4G data and unlimited 2G after that for $30 in the US

Through?

T-Mobile, I'm guessing, as I have the same plan. Look at http://prepaid-phones.t-mobile.com/prepaid-plans. It's near the bottom, under "Pay by the day" even though it's a monthly plan.

$30/mo. for 100 minutes, unlimited text, 5 GB LTE data, throttled to 2G but unlimited with no overages after reaching 5 GB. And their LTE is fast as hell, I have pulled 30/10 on it which is faster than my Qwest/CenturyLink DSL (at the fastest speed they'll sell me).

I've got T-mobile for $30 a month. I live in Knoxville. It is almost not terrible when it comes to data. I put up with it because I mostly just wanted it for the occasional call and lots of texting.

Actually didn't TMobile recently defect and then the others followed suit? AT&T lowered my prices recently!

My price went down massively thanks to the TMob kamikaze mission. 180->135 for 3 iPhones on ATT. Thank you TMob.

It seems like it is getting cheaper. I get unlimited minutes, texting, and 300 mb of data for the same $40 I was paying 5 years ago, but I was also paying extra for minutes, texting, data, etc.

Hey patio11, you're spending less time on BCC these days so even though the feature set is the same you SHOULD be charging less.

Hey colinp, your backend AWS costs have come down, so you SHOULD be charging less for tarsnap [1].

Hey you (yes, you), ignore everything you've read about surviving as a startup. You SHOULD be charging your customers based entirely on your cost base and ignoring any conversation about the value your customers receive.

[1] Is that even possible?

I might advise not phrasing this sort of advice sarcastically on HN, as I know there is a substantial contingent here which would actually believe that software being mostly done implies that the price should go down.

TarSnap dropped its prices five weeks ago.

A move not-well-received by proponents of value-based pricing - http://www.kalzumeus.com/2014/04/03/fantasy-tarsnap/

"Proponents of value-based pricing" is like "believers in anthropogenic global warming" or "supporters of the theory of evolution". Be careful not to "teach the controversy".

Tarsnap is tragically underpriced, and we have years of its business track record (compared to every other well-known backup provider) to see that the strategy of delighting the kinds of people who account for their storage costs to the picodollar does not result in runaway success.

Colin does fine for himself, but that's because he pours an avalanche of competence and expertise into what he's doing. If I were him, I'd be furious that I was getting lapped by people who do a much poorer job. But probably also hesitant to accept that my own low pricing had something to do with that.

I'd be furious that I was getting lapped by people who do a much poorer job

You'd only be getting lapped if you think life is a race to make money.

I'm doing Tarsnap because I wanted to build a good product, and by that metric I think I'm far ahead of the "competition".

You can treat your business as a labor of love, but be aware that you're making it less likely that the broad market will benefit from what you've built. Which is frustrating, because you have perhaps the only trustworthy cloud backup system I'm aware of.

Why frustrating? The guy's sitting right there telling you he's happy. The $tartup "culture" isn't for everyone, aight?

And I think this is a reasonable choice, primarily because you've made a choice (as opposed to many business owners I work with who are at effect of their circumstances).

Colin, I don't think I've had a chance to acknowledge you for supporting / allowing Patrick to write up the article I linked above. That was a brave call, and few business owners are bold enough to stand up while someone calls their baby ugly. (I think it was a masterpiece article on his behalf, and that Thomas makes a good point in his response here too AND I support your ongoing choices. I'm on the record as a long-standing fan of the picodollar.)

Tarsnap is, at least on HN, a great case study in the pricing and product choices startups have. I would make different choices regarding pricing (hence my grandparent post), but I would never have the capability to create a product remotely comparable - and the product is more important.

Reply on news.ycombinator.com