Fibre 3 - it's a cost centre / discretionary expense

Mon Sep 15 03:53:00 -0700 2008
manage

Thomas may have come late to the party, but he finally rung the bell that nobody else touched.

This says that we aren't all getting fibred up the wazoo because quite frankly the internet is just a discretionary expense, and while it is quite common for a household to "make a return" of having a telephone, a car, paying rent etc, that isn't the case for the intarweb.

Thomas goes on to say that he basically has enough bandwidth now for the Internet.

Thomas is right, this is the reason it isn't seen as a priority by anyone, and why calling the cable company to report loss of the internet is always going to be a low priority compared to reporting loss of the TV sports channels.

It really isn't that important, which is why it is quite easy to dismiss the whole idea due to the vast size of the USA and low population density.

We have just created the self perpetuating myth.

I will issue a challenge, and defy anyone to find anyone for whom "the internet" was actually just a cost centre.

I will issue a challenge like this because I've issued it before, and won it before, and always on the same terms, with the subject saying "well, you know, I've never really looked at it like that before."

E-mail alone ensures that home internet is exactly like business internet, and not a cost centre.

"Ah, but if I didn't have the internet I wouldn't be emailing so much!" yeah, and if you didn't have a car you wouldn't be driving everywhere, specious argument.

I am currently converting my mill and lathe to CNC control, and make no mistake, none of that would be possible at anywhere near the actual cost if it weren't for the internet connection at home, not just sourcing parts or info, or buying software and having it electronically delivered.

If the only way to find out about and buy Linux (insert distro flavour of choice here) was to buy a dead tree magazine that the distro paid for advertising space in, then send off a cheque, possibly internationally, then wait for the postman to deliver a CD, well folks, Linux would cost just about exactly the same as Windows.

Linux in particular and Open Source software in general is a child of the internet, before the internet (and I remember it well) the closest you could come was your local shareware library vendor, sneakernet on floppy.

No internet doesn't just mean no Linux as we know it, it means no alternative to the 200 dollar operating system and 200 dollar per suite of application software as we know it.

Bill Gates spectacularly under-estimated the impact of the internet in the Windows Chicago era, really, this was an error of judgement on a par with IBM granting MicroSoft that infamous licence.

Today, to say that phat pipes to the home as a Universal Service is not necessary is quite frankly an error of judgement on just as grand a scale.

Just look at us all posting and replying here, before the internet there was the BBS, and before that there was putting a stamp on an envelope and writing a letter to the editor, and then buying the paper again to see if your letter had been published.

To make the claim, as Thomas has, that the pipes we have are fat enough (even Zog on dial up) for "The Internet" is to get the picture hung upside down, this is a clear case of making an assumption that the pipes were built sufficient to carry the content out there, when what actually happened was the content always lead the pipe capacity, and all that has happened NOW is that generally speaking (500 kilobyte home pages aside) one specific form of content, namely http web page text publishing, is mature enough that most pipes, even Zog's, can carry it adequately.

To declare that this is a watershed, and "sufficient" is like declaring that the weekly Cunard sailing from Liverpool to the USA is enough and sufficient. Stop the world, I want to get off.

Back when I was paying a 5 pence connection charge and a penny a minute to connect at 9,600 baud to fido and we were talking about this new fangled internet thing over the horizon then yeah, Thomas was right, it was indeed a money pit, which is why only the self selected few were online, and boy I can remember the first time I met an actual woman who felt the same way, weird.

By the time xDLS always on connections came along however the net had ceased to become a money pit, at that point on balance the net saved you more money that it cost you to maintain a certain standard of living.

Every year since then as the inflation adjusted cost per X-abyte transferred has fallen that particular bit of mathematics has made the net an ever better bargain investment, speaking personally the net has both earned and saved me more money than everything else combined.

But Thomas is right, in that he is indeed stating a common perception of the truth, that the net is an expense, and not a vital one at that, and it is from that common conception that we can form opinions such as the one that states that due to the vast size of the USA it is simply unrealistic to expect everyone to get fibred up.

When I state quite plainly that this is the exact same size country that has utility electric running everywhere and metalled roads running everywhere it gets dismissed, because everyone thinks the net is just an expense, not an essential part of civilisation like roads or power.

No Thomas, I do not pretend to know what you would do with a gigabit up/down for 50 bucks a month, and more than a buggy whip maker would have known what use an interstate would be, or a candlemaker would have known what use a 3 phase supply would have been, or the pony express riders would know what use direct dialled telephones would be.

Saying we cannot afford to run gigabit to everyone is like saying we can't afford to run roads to everyone or electric to everyone IF YOU LOOK AT THEM FROM THE SAME ECONOMIC PERSPECTIVE, it would be utterly insane to build a nationwide electrical grid if there were no domestic electrical goods or industrial electrical equipment available for purchase, and utterly insane to build a nationwide road grid if there were no vehicles.

SURE, we can let the markets work it out, growth and sales moving hand in hand, slowly, slowly, slowly.

OR, you can adopt a "build it and they will come" mentality, also known as an industrial revolution, so the only question really is, "where do you want your kids to grow up?" Some third world technological backwater where everything you can shake a stick at is expensive unobtanium, or a modern country where pretty much anything is available at reasonable cost?

BT offered to roll out nationwide fibre 20+ years ago.

STOP AND THINK!

In 1998 when technology companies were springing up all over the planet, think of an alternate world where you could, at great expense, connect computers at 9600 baud, or at serious expense a staggering 32 kilobit leased line, or you could go to this island nation in europe where they were rolling out fibre and man, the speeds were just insane, and getting insaner all the time as laser development progressed, you know? that file that you just took four days to transfer would take, like, 2 minutes or something insane like that.

Every financial institution on the planet would be there for a start, because not being there meant an active disadvantage compared to those who were there.

High tech industry, that's a given, Microsoft would just move, so would Cisco, IBM and Intel, sure, might leave the fabs in "third world technology" countries to save money..

The whole CNC industry, that's a given.

The entire mobile phone industry, that's a given.

Those asian "tiger" economies... yeah.

Thatcher's decision to stop BT from rolling out fibre in exchange for being able to shove telly down it too was by far the costliest economic mistake ever made in this country, and quite frankly it is not like it wasn't obviously a stupid mistake at the time too.

Now WE are sitting here with our thumbs up our asses nodding sagely that it is just too expensive and not needed anyway, meanwhile other nations have leapfrogged us and the inevitable results of that are incubating in their 100 megabit and up networks, and we thing the new hybrid hummer and KFC recipe and reality tv show is going to power us economically into the 2010's and 2020's.....

you want to know where that kind of "well the US is just too big to fibre up" thinking gets you?

nearly FORTY years ago you could put a 4x4 on the moon, now you're reduced to hitch-hiking lifts off the russkies, and hoping they don't think you smell to bad to catch a ride because you've been playing in georgia.

snatch and miss

Mon Sep 15 10:33:28 -0700 2008
manage

I must have crowded your draw. Too hasty. Snatch and miss. That one whistled by my ear. I guess I get another shot now. I'll try to be clearer making my point this time. Not more direct, mind you -- just clearer ;-)

Back around 1987 I was part of a "test user" group around the campus of Carnegie Mellon University. We were testing an early stab at what would later become DSL. In particular, as the test group, we were given some very fancy new fangled modems that could talk out of band (i.e., with concurrent voice traffic) on twisted-pair copper at 9600 or maybe 19.2 KBaud. This was quite amazing, for its time: I could be chatting on the phone with someone, fire up my PC, turn on the modem, run a terminal emulator program, and instantly get a log-in prompt from a computer on campus (all while still talking on the phone).

A few years later the speeds went up and traffic went from dial-up to IP traffic -- the rest is history. I hear it said that the cable co's also learned to carry IP on all that coax they laid, so there's that, too.

Pretty soon we reached the stage of "bandwidth-enough" for basic web apps. And, indeed: our household spending patterns, the transaction costs for paying utilities, our capacity to advertise and sell stuff: all went through the roof, relatively speaking. As a younger person, finding work meant literally spending some shoe leather, literally knocking on doors. These days I can save dollars on shoe leather by spending pennies on bandwidth. On and on: in lots of little ways, you're right -- the Internet is not just a "cost center" to most households.

We aren't just talking about "the Internet" though. We're talking specifically about a marginal increase in bandwidth (real broadband vs. what we have). The question is: where is the value in that?

So, yeah, the country is wired for electricity and making that ubiquitous was indeed a political cause -- a social policy decision -- at one point. At the point that occurred, light bulbs and electric refrigeration were already on the market -- already in production for mass consumption. And light and refrigeration are huge profit-making tools for households. They are profound in their impact on lifestyle.

Around that same time, phonographs and radio came into mass production. Notice that you saw no political cause around universalizing those services! Awfully nice technology there: but nothing vital.

Around the same time, phone service really took off. Some functionality of the telephone was seen as vital, like lights and refrigeration. Universal service / life-line service was implemented.

Now, what services are around, waiting in the wings to justify the bandwidth gain of going to "real" broadband?

Not only is nothing obviously vital there to use that bandwidth, what is there to use that bandwidth looks like snake oil! Oh, boy, movies and TV. Oh, boy, unreliable phone service. Oh, boy, unrestricted porn. Oh, boy, with all the hot properties operated by power movers and shakers who are already too big for their britches.

Suppose I have some "social policy" capital to spend on a quarter-million new workers, trained to be techs.

On the one hand, I could make it easier for Netflix or Comcast to send me the latest hollywood releases over the net.

On the other hand I can train a bunch of them to install solar panels or (homage to zogger) insulation and shiny roofing materials.

See where I'm going with this?

We've got quite enough bandwidth for the basic web. We're nowhere close to exhausting all the useful applications of the current bandwidth. The main applications waiting in the wings for more bandwidth are most certainly almost entirely discretionary expenses.

Record player manufacturers are not common carriers obligated to provide universal service. Right now, the "more bandwidth" folks look a lot like folks in the recording industry.

I don't even want to talk about a build-out of additional bandwidth until the capitalists waiting to use it start talking about serious, useful applications as the dominant use.

Also, and as a kind of footnote: you don't strengthen your case by trying to embarrass us with comparisons to those other nations that "leapfrogged" us. There are many other metrics to consider other than just average and median bandwidth. Robustness is a good one. I think we are quite far ahead -- quite! -- by that metric. And, gee, wasn't that one of the goals?

Slow and steady wins the race. It isn't that more bandwidth is bad. It's just not time quite yet. There's other stuff to clean up first. Cake will be served for dessert and not sooner!

Oh, and: we could put a 4x4 on the moon inside a year if it really mattered.

-t

snatch and miss
Mon Sep 15 12:12:33 -0700 2008
manage

There are a lot of jobs in the U.S. that amount to not much more than paper shuffling and talking on the telephone.  From a technical perspective, they could be done entirely at home with decent bandwidth.

The current amount of bandwidth that I have (8 Mbps down, 1 Mbps up) is adequate for most of what I do.  I could really stand to get 2 Mbps up, but that isn't my issue.  My biggest issue is that this bandwidth isn't considered critical infrastructure.

If my power goes out, ComEd will have someone out to my house within an hour or so, barring major disasters that stretch their resources.  This is any day of the week, any time of the day.

If my analog telephone line goes out, AT&T will be on it immediately.

If my cable service goes out, the last time Comcast told me "We can have someone out there Thursday, between 2:00 and 5:00 p.m.".  This was on a Monday.  I am no longer with Comcast.

Critical infrastructure means if things stop working, people hop.  Shop doesn't close on weekends, evenings or holidays.

Yes, I am aware that I can pay for a business line with an SLA, and I've considered it.  But, right now, they are too expensive.

What the U.S. needs isn't necessarily 100 MBps symmetrical connections for $50 per month.  What we need is 10 Mbps symmetrical connections that are considered critical infrastructure for $50 per month.  That would allow proper telecommuting, even if only part of the time.

Then you could do things like VPN into the "office" and have access to the corporate storage; your office phone would connect up via VoIP; all the benefits of being on the corporate LAN would extend out to the teleoffice.

snatch and miss
Wed Sep 17 08:14:33 -0700 2008
manage

My biggest issue is that this bandwidth isn't considered critical infrastructure.

...

Critical infrastructure means if things stop working, people hop. Shop doesn't close on weekends, evenings or holidays.

Wouldn't it also mean no mucking about with breaking services you don't like, or perhaps even real common carrier status?

snatch and miss
Wed Sep 17 08:43:30 -0700 2008
manage

Not really.  It depends on what is in your contract.  If they're selling you pure bandwidth, then yes.  If they have Terms of Service other than uptime and QoS, then no.  Common carrier status would be the wise move, and is something I look for.

SpeakEasy is starting to offer 5x5 and 10x10 symetrical connections (in Mbps) as what they call their Business Ethernet.  It is really bonded DSL, from what I can see, but their main sell is liberal ToS and rock-solid uptime.

Unfortunately the cost is $849 or $1,249 per month.

Fibre 3 - it's a cost centre / discretionary expense (cf Thomas)

Mon Sep 15 16:20:58 -0700 2008
manage

I will issue a challenge, and defy anyone to find anyone for whom "the internet" was actually just a cost centre.

I will issue a challenge like this because I've issued it before, and won it before, and always on the same terms, with the subject saying "well, you know, I've never really looked at it like that before."

Not that I really understand what a 'cost centre' is to begin with. I'm guessing that it means that the interweb is just a sunk cost with no benefit to the individual that pays for it.

Which, as always in my one trick pony kind of way, brings us to the basic axiom that humans act to better their lot in life. In this case it would mean that by observing that someone has an internet connection it would be safe to assume that they derive some benefit over not having an internet connection or else they would, in fact, not have an internet connection.

Where your logic breaks down is assuming that because a gigabit connection could benefit someone that this person must have a gigabit connection even if it isn't important enough for them to want it in the first place or want it enough to pay the full cost to acquire and maintain it.

I'm sure the financial institutions or Microsoft or even the people who sell you the CNC stuff all have exactly enough bandwidth to do what it is they do in the most efficient manner possible. How do they determine 'the most efficient manner possible' one may ask, by taking the cost of the service and determining the benefit they will receive through this measure that's called 'profit'.

If one were to take away this 'profit' yardstick and just provide some arbitrary level of universal service then there would be no way to determine 'the most efficient manner possible'. It's the old Economic Calculation Problem that y'all central planners have never been able to overcome or in all likelihood will never be able to solve without reprogramming the human brain to turn everyone into the New Socialist Man.

I, for one, do not welcome our dystopian future overlords...too much of an Aldus Huxley fan for that.

And, you know, the whole flawed public goods theory and proof by example argument you present in support of your basic assumption.

Fibre 3 - it's a cost centre / discretionary expense
Tue Sep 16 03:03:38 -0700 2008
manage

Replying to myself.

So basically, nobody can see a need for gigabit fibre to the home as a Universal Service, 640k is enough for anyone, and all that.

Others are saying large nodes (MS) feeding small nodes (users) have all the bandwidth they need, so where is the problem.

Charles sort of gets it, if his power goes out, it is a priority, if his internet goes out, we'll try and get a guy around next week.

Me, I'm just going to throw my hands up in hopeless despair, I'm getting the exact same comments we were getting back in the modem days, who needs an always on connection, who needs 512 kbit, bonded 64 kbit ISDN is enough (yes, really, no shit) for a company office with 75 employees...

Run a fibre between point A and point B and that is pretty much it, it is ducted wireless with no leakage, want more throughput, change the laser, running a thousand gigabit channels costs the same in throughput costs as running empty, to all intents and purposes.

The cost per megabit, and it is just ones and zeros don't forget, not real physical stuff, drops to essentially zero, watershed moment.

So yes, I can stream 1920 x 1080, or maybe even wall sized 3840 x 2160, and instead of wallpaper I can have a live view of the serengheti, or maybe niagara, or maybe earthrise.

Screw the phone, I can sit there on my sofa and sociaalise with other people sat on their sofa.

Teaching? anyone? The virtual classroom becomes a reality, just tune in to my channel.

Fibre 3 - it's a cost centre / discretionary expense
Tue Sep 16 14:31:45 -0700 2008
manage

So basically, nobody can see a need for gigabit fibre to the home as a Universal Service, 640k is enough for anyone, and all that.

No, what I've been saying across multiple articles is 'nice to have' doesn't seamlessly translate into 'critical to have' no matter how much you sugarcoat it.

Me, I'm just going to throw my hands up in hopeless despair, I'm getting the exact same comments we were getting back in the modem days, who needs an always on connection, who needs 512 kbit, bonded 64 kbit ISDN is enough (yes, really, no shit) for a company office with 75 employees...

And when it was discovered that this level of bandwidth was insufficient for the task at hand the Bandwidth Fairy waved her magic wand?

Or was it the interested parties who had a direct stake in the matter got together and agreed to contract for a higher level of service and ipso facto the needed bandwidth was produced out of the ether?

Ignoring Gore and his Magic Bandwidth Wand when he was in congress of course. Which turned out to be a giant subsidity for google when they bought up all that overproduced dark-fiber at bargain basement prices if I'm not mistaken on the funding method to lay all that cable.

Teaching? anyone? The virtual classroom becomes a reality, just tune in to my channel.

Yeah, the monopoly providers of 'education' are going to love that...