back to article Building the actual real internet simply doesn't pay

It isn’t unicorns that keep the internets running, but eye-watering amounts of capital investment by infrastructure operators. However a substantial chunk of that investment never pays for itself and most of it barely repays the cost of the capital, reckons consultancy PwC. The telco industry as a whole spends $320bn on …

COMMENTS

This topic is closed for new posts.
Anonymous Coward

I often wonder how much more effecient the world would be if it were run by accountants who seem to know the most effecient and cost effective to do everything... mainly I wonder if anything would actually work.

20
0
Silver badge

The world is run by accountants.

The problem is that the "most efficient" / "most cost effective" way of doing things for them is not the same as for their company's customers. That's why it's cheaper for me to talk to Outer Mongolia on my phone for an hour than it is to load up a webpage while travelling in France.

5
0
Silver badge

It'd work.

Think of "brazil"

6
0
Anonymous Coward

Real world, real money

Seems to miss the point that its real money that is spent and if the Telcos don't get income to cover the cost they will stop doing so. Would you spend money to do your job if you weren't going to get it back?

1
0
Silver badge

Re: Real world, real money

I would if my contract allowing me to use these frequencies required that I provide equal service to the entire area.

That doesn't quite answer the question so I'll also say, that they don't charge enough (or they pay their director and his staff and his pilot and his valet and his driver too much).

3
0
Anonymous Coward

Ah, accountants. They know the price of everything, and the value of nothing.

7
0
Silver badge
FAIL

NEVER let an accountant control your business. They don't understand what entropy is.

4
0
Silver badge

It's problems like this where an end-to-end monopoly provider would theoretically be able to provide a better solution than a bunch of independent operators. A monopoly would be able to leverage income from profitable parts of the network to subsidise necessary infrastructure elsewhere in order to improve services.

In theory.

Unfortunate that real monopolies, whether state or private, rarely have any incentive to improve their service in this way.

3
1

Pioneers go bust

The people who built the canals went bust. The people who built the railways went bust. The first wave of dotcom'ers went bust.

Pioneers die so the rest of us can learn how to live in the new world safely.

Why should the internet infrastructure be any different?

6
1
Anonymous Coward

Re: Pioneers go bust

Because the time for the pioneers to die has already happened and the Internet has been around for donkey's years?

1
2
Silver badge

Re: Pioneers go bust

> Because the time for the pioneers to die has already happened and the Internet has been around for donkey's years?

The first public railway opened in 1830, and the railway bubble peaked in 1846. 16 years

The WWW consortium was founded in 1994. 16 years on was 2010. Not so far out...

2
0
Facepalm

Re: Pioneers go bust

Internet != WWW.

0
0
Silver badge

Re: Pioneers go bust

Internet != WWW.

Certainly this is true. But the quantitative changes that happened with Internet use since the rise of the WWW - in nodes, users, traffic, bandwidth available and consumed - are so enormous that they constitute a qualitative change as well. Along with the other qualitative changes that have occurred since, and largely because of, the rise of the WWW and the concomitant rise of commercial and consumer use of the Internet - new protocols, new usage patterns (eg mass video streaming), widespread use of NAT, introduction and (extremely) gradual adoption of IPv6, widespread use of VPNs, and so on - the Internet+WWW can fairly be considered, while not sui generis, categorically different from the Internet of the 1980s and early 1990s (which in turn was significantly different from pre-IP ARPANet and other predecessors).

So Phil's 1994 starting date and 19-year interval are defensible in this context. The buildout of Internet+WWW infrastructure is what's comparable (if anything here is) to what happened with early railroads - not the history of the Internet since CIX (1991), or the Big Switch (1983), or since ARPANet (c 1970), or whatever other origin you're inclined to pick.

2
0
Anonymous Coward

I wonder how much cash Google is hemorrhaging in Kansas

I wonder how future proof 1 Gbit/s symmetric is

I wonder how often the other ISPs will upgrade their networks before they reach 1 Gbit/s symmetric

0
0
Anonymous Coward

I love the sound....

"I wonder how much cash Google is hemorrhaging in Kansas"

I love the sound of jealousy in the morning - it sounds like the wind through the Osage Orange trees.

0
1
Thumb Down

Re: I love the sound....

Or the tornados ripping out the Osage orange trees.

0
0
Silver badge

Sorry, but I'd really rather have a telco that hadn't spent billions on 4G licences, but actually used 3G to the extent that it's possible (and that other countries do, e.g. 42Mbps in some countries using the same technology).

If they'd invested in larger back-ends and not new technology and licenses, maybe our 3G connections would be able to deliver closer to the achievable maximums (which are basically within the 4G everyday-average area), without contention, and limiting, and bandwidth limits. If they could deliver a "truly unlimited" connection for a price, and not still charge in units of Megabytes. Maybe if they didn't spend so much on blocking / hindering Skype but embracing it instead and letting customers go onto a "data-only" tariff that lets them use whatever they hell they want over it (not telling me things like Kindle's, GPS trackers, etc. don't already have similar data requirements and zero voice capability)?

There's a million and one things they could do, but apparently all they could think of was to spend insane amounts of money on equipment upgrades, mast upgrades, radio licenses and R&D over their already-stretched back-ends, and keep customers on ludicrous tariffs with pitiful data allowances to pay it back without everything just collapsing under the load.

The bottleneck isn't now, and never has been, the technology, the coverage, or the customers equipment. It's always been the back-end. And the same problem exists for home broadband, too. Hence why businesses with proper data needs would rather have a 10mbps leased-line or SDSL than even the best "business broadband" package on a contended line.

You're selling a six-lane motorway, with state-of-the-art cars, that comes down to a bumpy, single-lane (and even unidirectional with "give way" traffic management) unpaved, muddy road every time you want to get somewhere.

14
0
Silver badge
Unhappy

"If they'd invested in larger back-ends and not new technology and licenses, "

then they'd lose to those operators which did.

Because the cost of the larger back-ends comes out to about the same or more than the cost of the new technology, but the advertising value new shiny technology is much greater than the advertising value of large back-ends (except to those who are enamoured of such,and incapable of dishonesty.)

THAT is our problem. As a society, we tend to make decisions not based on what is efficient and gets us what we need and want in the long run, but based on what is cool and new.

5
0
Silver badge

Re: "If they'd invested in larger back-ends and not new technology and licenses, "

Except that this isn't quite true either, at least in my case.

Orange - AKA EE - was the 1st network with 4G but that would in no way tempt me away from 3UK. 3UK seem to have far better speeds than anything I ever got on Orange. As a result I will never go back to Orange/EE again.

(A top speed of 0.11Mbps is just not good enough IMO - especially when I've been on some rare occasions been getting ~10Mbps on 3UK)

1
0
Silver badge
WTF?

"...poor decision making..."

If PwC have a system for perfect decision making they should sell it. They could make a fortune.

(Would it be too low of me to suggest that accountants might be responsible for some of the "poor decision making"?)

4
0

Re: "...poor decision making..."

Other accountants maybe. I'm sure PwC's accountants are a better breed that will definitely help you make the right decisions.

Perhaps.

1
1

This post has been deleted by its author

Silver badge

Re: "...poor decision making..."

(Would it be too low of me to suggest that accountants might be responsible for some of the "poor decision making"?)

Two words: 'Andersen'. 'Enron'.

'nuff said... :)

2
0
Bronze badge
Devil

Agreed,... Kill all the Accountants then go for the Lawyers and Politicians

Seems the ban counters know everything these days. Unfortunately that goes for the whole Political sector as well.

Here's an idea. Only people who "Can Do" get to make the rules. Those who "Can't" or "Don't know how" have to live by them and get no vote.

The sad state of of the communication infrastructure is the result of lobbyists, graft, greed etc.

In NY State, the NYS Public "Disservice" Commission refuses to force Verizon to live up to it's agreement (Tarrif) to repair the existing telephone wiring that they have been getting a line maintenenace fee of roughly $60 USD per year from every subcriber in the state since they were Ma Bell.

My contention is that only the state and federal government should own all infrastructure and contractually be able to toss out any company that does not live up to the terms of the tarrif to operate, maintain and extend the networks.

2
1
Silver badge

Re: Agreed,... Kill all the Accountants then go for the Lawyers and Politicians

"Only people who "Can Do"

sadly you don't have to look too hard at history to see the effect of people who 'can do' and are 'certain' of things, and it ain't good. kings, popes and all manner of dictators were all very can-do and absolutely certain about their path.

3
2
Silver badge

Re: Agreed,... Kill all the Accountants then go for the Lawyers and Politicians

True... but the amount of Marie Antoinettes and Neros are even worse.

1
1
Silver badge
Devil

Re: "Only people who "Can Do" get to make the rules"

Great idea.

I'm sure the Mafia (Russian or other) and the Yakuza (not to mention the Triads) will wholeheartedly agree.

<sarcasm>Actually, one must wonder if placing authority in the hands of criminals would not be more beneficial than leaving it in the hands of professional liars, sorry, politicians.</sarcasm>

<cynism>One thing is sure, if your mayor is a hardened criminal, there will be much greater respect for his "laws" since everyone knows that the penalty for transgressing them is a lump of lead between the eyes.</cynism>

1
0
Silver badge

Re: Agreed,... Kill all the Accountants then go for the Lawyers and Politicians

the amount of Marie Antoinettes and Neros are even worse

And the number of people who substitute a passing familiarity with myth and anecdote for basic competence in history greater yet.

1
0
Bronze badge
Devil

Re: Agreed,... Kill all the Accountants then go for the Lawyers and Politicians

I guess I have to spell it out. The operative words "Can Do" was relative to the peoplewho actually work (technicians, IT people, etc) instead of the bean counters.

Who was taking a discussion about the state of internet infrastructure and turned it into a political argument?????

Wasn't me!

0
0
Silver badge

Re: "Only people who "Can Do" get to make the rules"

Of all tyrannies, a tyranny sincerely exercised for the good of its victims may be the most oppressive. It would be better to live under robber barons than under omnipotent moral busybodies. The robber baron's cruelty may sometimes sleep, his cupidity may at some point be satiated; but those who torment us for our own good will torment us without end for they do so with the approval of their own conscience. — C S Lewis

0
0
Silver badge

And yet even with all that investment that we're supposed to believe has been wasted the telecom companies still manage to make a profit.

As for the 'six fold increase' over the last thirty years this is roughly the same time that mobile phones were being increasingly used by the general public. Costs may have gone up but no doubt the amount of money they are making has done too.

http://www.guardian.co.uk/technology/2011/apr/11/mobile-phone-users-wasting-5bn

Remember that the article above refers to UK customers only. I wonder how much this is when global wastage is taken into account?

4
0

The poor decision making at big telco's has much to do with these businesses' scale and complexity. There's no mystery and not a lot that can be done whilst corporate ambitions are to grow in to wider and wider markets.

Once you've grown in to your natural addressable market as a bits and pipes company, to make your CEO's share options grow at the same rate the conventional wisdom is that you have to chase "value add", ploughing your company's cash in to app stores, virtualisation, security, sector-specific propositions etc. in a never ending cycle of reinvesting profits in to non-core (for a telco) activities.

In order to support this you need to divide your company's finite resources in to increasingly wider activities and dilute your company's previous effectiveness as a telco whilst expecting it to continue to be as successful as it used to be when it was a simple business with a clear focus.

But as long as you can keep moving the goalposts and persuading your investors that the ROI is just around the corner (and not really just about to be spent on yet another project), until your spell as CEO is over and you have your options vested, then all is good.

4
0
Anonymous Coward

"Interesting"

Interesting that PwC reckons the "cost of capital" is 9%.

Anyone think they can get 9% on their savings, right now, in the UK? (Without chucking their savings away by lending them to wide boys, that is).

We all have small amounts of capital, and in the old days the big chunks needed to build infrastructure were composed of millions of those small amounts. Yet today, small investors barely get any interest at all, while the huge banks charge 9% minimum - to lend OUR MONEY.

7
0
Anonymous Coward

Re: "Interesting"

"Yet today, small investors barely get any interest at all, while the huge banks charge 9% minimum - to lend OUR MONEYkite checks based upon treating other loans as being the same as money in the bank."

0
0
Anonymous Coward

What?!!!

If we want a better internet we're going to have to pay more for it? What sort of dumb-ass capitalist idea is that?

0
0
Silver badge
Facepalm

Re: What?!!!

Damn for those other couturiers with faster more reliable speeds for half the price for being so, so, damn successful!

0
0
Silver badge
Facepalm

Re: couturiers

Really, other people have faster and cheaper dressmakers than we do? Who knew?

0
0
Silver badge
Pirate

Cry me a river

Yet telcos somehow manage to show profits year after year.

Sounds to me like the oil supply chain. Everyone in the chain claims thin margins poverty, yet everyone shows million/billion dollar profits.

3
0
Gold badge

A sort of loss leader...

"It's problems like this where an end-to-end monopoly provider would theoretically be able to provide a better solution than a bunch of independent operators. A monopoly would be able to leverage income from profitable parts of the network to subsidise necessary infrastructure elsewhere in order to improve services."

A monopoly would also be able to leverage income to fatten their wallets, and (because they are a monopoly) do NOTHING elsewhere.

The thing the economists are missing here is this -- I think these rural upgrades in some cases are actually a sort of variant of a "loss leader" rather than just being an investment that was a big money loss and a waste. If you look at the big 4 cell providers in the US -- Sprint and VZW use the "CDMA path" (2G CDMA 1X -- which they still must keep running for voice until VoLTE.... 3G EVDO, and 4G LTE. Sprint started using Wimax for 4G but scrapped when it was clear LTE was going to "win" as 4G standard.) AT&T & T-Mobile are 2G GSM (EDGE/GPRS)->HSPA/HSPA+->LTE. (T-Mo uses 21mbps or 42mbps HSPA+, AT&T generally 14.4mbps with areas still on 3.6 or 7.2 last I heard.)

So, Sprint and T-Mobile cover mainly urbran coverage. T-Mobile has nice 21 or 42 HSPA+ urbanly, but EDGE and even GPRS along highways, with "no service" or roaming otherwise. Sprint has mostly 3G coverage, but again just cities and main highways generally. Verizon Wireless has nearly ubiquitous 3G coverage, 100% of their network is 3G with ~50% (by area) LTE now, and planning 100% LTE by year end. AT&T has less than Verizon Wireless (less overall coverage, a good ~33% or so 2G-only still, and less 4G) but way more than Sprint or T-Mo.

So, the economists making this report would see AT&T and VZW as wasting money on rural coverage, and I'm quite sure on it's own it doesn't pay back. But, AT&T and VZW charge FAR MORE for their service than Sprint or T-Mobile would dare to charge, specifically because they have more extensive coverage, more 3G, and more 4G, for people who travel a lot, or that value having their phone not pack up over price.

2
0
Silver badge

$320 billion: $46 each.

If I could buy my share of the internet, I would not be paying monthly.

3
0
Silver badge

Re: $320 billion: $46 each.

Less than $4/month each for a year. I'm on.

I wonder why the telcos aren't.

1
0
Silver badge

The total sum is irrelevant

After all the productive end of things is fairly cheap. Those are just routers (though 100k ones) and fibres. That's all very cheap.

There is some cost involved in the mobile world as you need towers and fairly expensive radio and backend equipment. Having decent quality links from moving vehicles isn't trivial and requires quite some effort.

However there's also the beancounters end of things. Beancounters want authentication, authorization and accounting. Which means you need a Radius server next to your BRAS, which means you need a BRAS instead of a simple router. Throttling down individual users causes even more cost.

Extreme cases are the mobile networks. Their whole design was made solely to provide AAA. It it wasn't for it, you'd just have the base stations connected to the Internet, perhaps with an intermediate layer in between to enable roaming. Instead of that you get elaborate messes of network infrastructure designed to make sure you are a paying customer and that you _will_ be billed for whatever you do. In fact many mobile ISPs even run large scale transparent proxy servers to mess with your traffic. All of that costs far more than just bare access to the Internet.

BTW if you want to see how cheap it can be, look at the Freifunk movement. They manage to do roaming and fast data with re-flashed 15 Euro routers. In the project I'm involved in we route the data out to Romania (and pay for the traffic) to get rid of the legal issues.

2
0
Unhappy

The problem with measuring efficiency is that

Often the math treats getting money for not actually doing anything as a 'good thing', and declares that everything is a zero-sum gain. With no concept of a win-win situation, is it any surprise that an 'optimal' strategy is to pull an Enron, then cash out right before things come to a head?

0
0
This topic is closed for new posts.

Forums