Microsoft co-founder Paul Allen is out with his autobiography and Vanity Fair has an excerpt available online. As the Nth richest man in the world, Allen isn’t doing this for the money.  Maybe it’s for posterity. Maybe to settle old grudges, and he certainly does that in Vanity Fair.

The part of that excerpt everyone will be talking about this week is Allen’s story of overhearing Bill Gates and Steve Ballmer plotting to recover Allen’s Microsoft shares or dilute him into insignificance, this at a time with Allen was dying of non-Hodgkins lymphoma.  It’s a great story, that’s for sure.  But if you are a longtime follower of this column or its predecessor you’ve read it before.

Tell your friends, “Oh that. Cringely covered it back in 2006.”

Podcast: Play in new window | Download

I have been doing business in Japan for 20 years, consulting for big and small companies, speaking at conferences, writing for Japanese publications, and helping both American and Japanese companies do business with each other. For years I flew to Tokyo once a month, generally in my role as giver of bad news, which I could get away with as an American. Throughout those 20 years I have been astounded by the energy and discipline of Japanese industry, and by its turgid impenetrability. For a country known for advanced technology, Japan is astoundingly resistant to outside ideas, as the current earthquake and nuclear crisis show yet again.

You’d think they’d want our help, and they do to a certain degree. But it’s like those people I meet on airplanes who find out what I do for a living and tell me they would really like to write a book: what they mean is that they would like to have written a book. The Japanese would like for us to have helped them — to gain the benefits of our assistance without the embarrassment of admitting they need help or the complications of arranging to accept it.

This lesson was learned to some extent during the Kobe earthquake of 1995 when Japan waited several days before even responding to international offers of assistance — days during which Japanese citizens were still dying under rubble. In Kobe what was on offer were mainly trained dog teams to sniff-out survivors. That part of the lesson was learned: when this earthquake happened, Japan was quick to accept such assistance and thousands of lives were probably saved as a result. But the Fukushima nuclear accident is a very different story.

In this nuclear accident the situation is complicated by an extra party — Tokyo Electric Power Company — with its corporate personality and internal agendas. TEPCO is embarrassed by this accident. Embarrassment, either corporate or personal, is a huge deal in Japan. It’s not like they can just give up their corporate face for a few weeks or months while necessary things get done. I saw a similar unwillingness to squarely face reality at General Public Utilities back at Three Mile Island in 1979. In both corporate cultures there was too much emphasis on political damage control — emphasis that often comes at the expense of good engineering.

If a nuclear plant manager is worried too much about his job he isn’t worried enough about his reactor.

TEPCO just this morning announced that four of the six Daiichi reactors can never be repaired. I wrote that right here less than 24 hours after the earthquake and tsunami before the emergency batteries had even run out. It was instantly obvious to even a moderately informed observer like me, yet why did TEPCO take two weeks to come to the same conclusion? Internal politics, which can only increase public danger.

But wait, there’s more! Now we have reports of water contaminated with plutonium at the plant and possible plutonium ground water contamination. Radioactive cesium and iodine are bad enough, though that water can be stored in pools for a few months while the radiation decays then carefully diluted for disposal. But plutonium contamination is forever — at least 10,000 years.

There are right now two plutonium remediation technologies on offer to the Japanese government and TEPCO that I know about — one from Russia and one from the USA. One approach uses nanotech and the other uses biotech but both are novel and unique. Both have been offered to the Japanese through government channels and in both cases the Japanese government or TEPCO have yet to respond.

I know about these technologies because the Russian one is represented by a friend of mine and the American one comes from a Startup America company so I took it straight to the White House myself.

I think it would be smart for TEPCO to adopt both technologies in case one works better than the other. But my sense is that if an answer ever comes from Japan it will be months from now and will probably be “no thanks.”

Think about this as you read about that plutonium-contaminated water, because it is going to be in the news for years to come. If only there had been a technology available to clean up that stuff early in the crisis, the pundits will say, lives could have been saved. There was such a technology available — two of them in fact.

Who’s embarrassed now?

This is my son Fallon’s shoe, a Skecher’s Hot Lights Chopper, U.S. size 12.  The toe, filled with LED lights, is glowing in the picture. It has been blinking at me from across the room all evening now. It reminds me of the heart of a shark I once saw beating on a dock in Boothbay Harbor, Maine, hours after it had been cut from the fish. Unlike that shark, Fallon’s shoe isn’t dead, but it is getting close, the flashes coming further and further apart.

Light-up shoes are a fact of life for American parents.  Fallon, who is four, is in the golden age of light-up shoes and he loves them.  But I gave little thought, until tonight, to how such flashing shoes work.  Kids ask, of course, so I always gave them my best guess.  Strontium-90 energy packs were out of the question since the 1950s, I explained, so the only mechanism that made sense to me was a piezoelectric crystal generating a small voltage with every step.

But this single Skecher, mocking me from across the room, showed I was wrong.

My theory of shoe operation required walking to generate power to light the lights, yet here was a sneaker lighting-up all by itself for hours and hours.

A little research on the Internet and I had it and you can too if you are willing to buy 5000 pieces at a time — the LED light controller and battery from a child’s light-up shoe (below).  I guess piezoelectric crystals cost more and batteries cost less than I thought.

And for the geekier parents reading this, yes, that’s a mercury cell in your kid’s shoes.  Two of them, actually.

Microsoft last week bought just over 600,000 IP addresses (a /10 block and a /11 block if you are counting) for $7.5 million from bankrupt Nortel. For a moment there it was everywhere on the web, a mild reminder of what happens during famine when gluttons hoard food. But what is really going-on here, and what does it mean in the near and longer terms? Well first let’s settle something: it is immaterial to Microsoft. Had the price been $7.5 billion or better yet $75 billion, I’d say that Redmond viewed as central to its survival having that block of addresses. But $7.5 million is pocket change and probably represents to Microsoft just a cheaper way than some other of doing the same thing. What it means in the long haul to the rest of us is yet more chipping-away of our facade of IT empire as we find increasingly complex ways to preserve IPV4 while China, for example, mandates IPV6.

If you aren’t up on this broader story it is simple — there are only around 4.3 billion IPV4 addresses yet a lot more than 4.3 billion people and digital Internet nodes already in the world. There are two ways to deal with this population problem: 1) move on to a new system with more addresses (IPV6 which has more than we think we’ll ever need but didn’t we think that the last time, too?), or; 2) hold the system together with a mixture of internal and external and static and dynamic IP addresses through that happy kludge called Network Address Translation (NAT). IPV6 puts your light switch or, for that matter, every individual light from your Christmas tree on the Internet and NAT can do that, too, but with a lot more effort and a lot less fun.

It’s not that we’re actually out of IPV4 addresses, either. The simple analogy here is to money. The economy is crap right now and yet all the economists talk about trillions of dollars being “on the sidelines” and “waiting” — though it’s never quite clear to me waiting for what. Same, too, with IPV4 addresses, which are all assigned, we’re told, but not all are being used, like those 600-odd thousand snapped-up by Microsoft.

I have no doubt those addresses are for Redmond’s cloud strategy, by the way. If they want to virtualize hundreds of thousands of customer servers they’ll need hundreds of thousands of IP addresses, simple as that. Microsoft actually thinks about stuff like this unlike, say, me.

There are plenty of IPV4 addresses either not in use or improperly in use today. I’m told that Verizon, for example, has two /16 blocks (128K addresses in all) that are external addresses assigned to internal nodes like printers. Those could all be recovered since they are being misused, or Verizon could sell them for close to $2 million, following Microsoft’s act of price discovery.

I’m sure there are millions and millions of IPV4 addresses to be regained just as I am sure that most of them won’t be because the rich don’t see themselves remaining that way by giving their stuff away for free.

Routers from Cisco and Juniper have been ready for IPV6 for a decade or more. In one sense it’s just a matter of turning it on. And doing so would bring us advantages in network performance and security, too. But that isn’t going to happen anytime soon in the USA because of the complexity of all those NAT layers presently in operation, but even more so because of the threat it poses to entrenched network administrators and IT directors.

IPV6, you see, doesn’t differentiate between the workgroup and the galaxy, so workgroup sysadmins and net admins might disappear in droves. This won’t go down well in an industry based on keeping CEOs ignorant and in fear of the network and continually adding IT labor whether it is needed or not. Entrenched IT will fight tooth and nail against IPV6, telling all the appropriate lies to keep us from moving forward until we’re a decade or more behind.

They — maybe you — don’t really care.

This won’t happen — can’t happen — of course in China or India, both of which will shortly need at least a billion addresses each for smart phones alone. IPV6 is their only answer. So thanks to their late entry in this Internet thing and our gleeful willingness to self-destruct, they’ll shortly be ahead and we’ll shortly be behind.

But Microsoft, thinking ahead, will have IP addresses to spare.

Podcast: Play in new window | Download

If you were Verizon CEO Ivan Seidenberg, faced with suddenly becoming the number two mobile phone company in America following an AT&T/T-Mobile merger, what would you do?  You could try to buy Sprint, and for all I know Seidenberg will do just that.  You could make a counter-offer for T-Mobile, but that would just be too darned expensive. If I was Seidenberg, though, I would try to poach customers — millions of customers — from T-Mobile.

AT&T is paying $1300 per T-Mobile subscriber and by the time the deal is finished extra costs will probably raise that to $1400 or more.  Were I Seidenberg, then, I’d spend right up to that level to snag customers from T-Mobile.  Anything under $1300 is a bargain.

At a minimum, I’d match or beat whatever those T-Mobile customers are paying per month now, I’d cover their cancellation fees, and I’d replace all their phones for nothing.  Got a smart phone? Have an iPhone 4!  But why stop there?  Have a feature phone? Have an iPhone 4! Or Android or Blackberry or Windows Phone — whatever you like.

All smartphones all the time at Verizon!

Then watch five million or more T-Mobile customers defect to Verizon, raising AT&T’s per-subscriber cost by $300, pushing break-even on the deal to 2016.

But hey that’s just me.  Maybe Ivan’s a pussycat.

Podcast: Play in new window | Download

In the 36 hours or so since AT&T and Deutsche Telekom announced that the American carrier would be buying the U. S. subsidiary of the German phone company, there has been plenty of speculation (some of it right here) about what this will mean for customers and the wireless industry, but not very much, frankly, about why T-Mobile is worth $39 billion to AT&T. It’s about more subscribers, we’re told as though that is obvious, and back-office savings, plus extra spectrum with some special plans for 4G, but that’s not the biggest reason at all. The biggest reason why AT&T wants T-Mobile is because of WiFi.

Subscribers are nice, as are back-office and marketing savings, but unused spectrum — while it has value — also costs billions (and more importantly years and years) to build-out. But many of AT&T’s current broadband service problems could be solved almost immediately by more creative use of WiFi, which is definitely coming.

For any 3G or 4G wireless carrier voice and voice backhaul have become tiny parts of its bandwidth budget. Voice and texting are together pretty much ignored in that calculation they are so small. It’s web surfing and apps, apps, apps that cost the big bandwidth bucks, and the people who utilize that digital bandwidth in prodigious amounts are generally concentrated in major metro areas — just like each telco’s WiFi hotspots.

T-Mobile has 7,000 cell towers in the USA but 30,000 WiFi hotspots. AT&T has another 30,000 hotspots of its own. T-Mobile’s hotspots are conspicuously connected by fiber with major bandwidth — more so than AT&T’s hotspots, which aren’t so bad themselves.

If hotspots and cells have comparable backhaul capability and I’m told many of them do, then T-Mobile has more than four times the broadband capability through WiFi than it has through the cell network. And remember that an urban cell can easily cover a square mile (640 acres) or more while hotspot rarely covers more than an acre, making the effective data density many times higher.

Now add to this the WiFi capability in our homes, which T-Mobile already has software to leverage — software that you can bet will be shortly used by AT&T as well.  Clever use of other people’s bandwidth can add an order of magnitude to AT&T’s connectivity and backhaul for no marginal price at all.  Suddenly the network expands, coverage gaps go away, yet backhaul bandwidth actually drops.  Look for it.

What we’re likely to see, then, is more transparent use of WiFi on a combined AT&T/T-Mobile network. And I’ll bet a nickel that particular part of the network consolidation begins almost immediately because WiFi is WiFi and all the phones are ready to go. Why else would AT&T offered T-Mobile a $3 billion breakup fee if they didn’t want the Germans to start complying with certain consolidation terms even before the deal is approved?

So look for software updates that choose WiFi first and connect without asking, whether at home or McDonalds. Look, too, for new pricing plans that make WiFi connections not count against bandwidth limits, encouraging the cheaper among us to make a little more effort seeking-out that odd Starbucks or friendly neighbor.

With AT&T’s LTE 4G service rolling-out over the next couple years to settle the issues of dropped calls and lousy surfing speeds, AT&T still needs a quick fix to level the playing field with Verizon. Yes, all those other reasons for AT&T buying T-Mobile still apply, but the most pressing is WiFi integration because it can be turned-on almost immediately.



Podcast: Play in new window | Download

So AT&T is buying T-Mobile USA for $39 billion in a deal that makes perfect sense if you are an RF engineer or a fat-cat telco tycoon, but my question is what happens to all the jailbroken and unlocked iPhones?

T-Mobile and AT&T are the USA’s only GSM wireless network operators, so if you had an iPhone and wanted to dump AT&T to allow things like free tethering, the obvious (and frankly only) way for Americans to do so was by jumping from cranky old AT&T to the much friendlier T-Mobile. And so tens of thousands — maybe hundreds of thousands — of AT&T customer did just that, and were gratefully accepted by T-Mobile.

But now with the T-Mobile brand, back office, and customer service likely to go away, will AT&T turn all those iPhones into bricks?  It depends in part on Apple, on the Apple-AT&T contract, but mainly I think it depends on terms set by regulators in return for approving the deal.

AT&T has said it will grandfather T-Mobile customers, honoring their often lower monthly fees and continued use of T-Mobile phones, but AT&T has had nothing specific to say yet about T-Mobile iPhones.

Apple hates jailbroken and unlocked iPhones, of course, and would like to see them all die, but since Verizon began selling iPhones in the USA, Apple has lost some clout with AT&T.

So my guess is that AT&T will allow jailbroken and unlocked iPhones to run on their network if the Federal Communications Commission or Federal Trade Commission or Department of Justice demand it as a condition for approving the merger, which they will if we demand it. And if that happens, the even more important question becomes whether Apple will lose some control of its ecosystem?  Will jailbreaking and unlocking — enabling iPhones to add software features and do things beyond the ken of Cupertino — become the norm?

I hope so.

Update — According to a report this morning in Forbes: “AT&T said Monday that it in the year after the closing, it plans to rearrange how T-Mobile’s cell towers work. The spectrum they use for third-generation services, or 3G, will be repurposed for 4G, which is faster.  That would leave current T-Mobile phones without 3G. They would need to be replaced with phones that use AT&T’s 3G frequencies. AT&T said it had factored the cost of replacement phones into the total cost of the acquisition.”

This would seem to suggest that AT&T will give you a free replacement for your jailbroken, unlocked iPhone on the T-Mobile network.

But wait, there’s more! If AT&T is repurposing T-Mobile 3G service to 4G, doesn’t that strongly suggest that 3G is going away completely on AT&T?  It looks that way to me. So will AT&T be giving EVERYONE a free 4G phone upgrade or just the jailbroken unlocked iPhones?

Ironic, eh?

Podcast: Play in new window | Download

At this point in the Japanese nuclear emergency it is coming down to the simple proposition of how do you drop enough water on the stricken reactors, and especially the spent fuel ponds, to keep further damage from happening? The Japanese Self-Defense Force is experimenting with helicopter water drops which are, frankly, stupid. The choppers can’t drop enough water to make a difference and they are so slow that they potentially expose their crews to excessive amounts of radiation.  Much better to use the Evergreen 747 Supertanker.

The unique Evergreen Supertanker is the largest piece of firefighting gear on earth, capable of dropping more than 20,000 gallons of water or chemicals at a shot (40,000 gallons per hour!), then streaking back for more at over 600 miles-per-hour. The Supertanker, which works all over the world and can easily get to Japan at this time of year when most of the world isn’t on fire, could make a huge contribution to the nuclear effort. And the best part is that the addition of some lead shielding for the crew would make their time on-station virtually infinite while having no impact all on the Supertanker’s performance. What’s a little lead when you already weigh 800,000 pounds?

So why haven’t we seen this behemoth already at work in Japan, dropping boron-laced water?

UPDATE — An enterprising reader (far more enterprising than me) called Evergreen Aviation early this morning and learned that the company has been waiting for just such a call from Japan. But it turns out the Japanese authorities either didn’t know about the Supertanker or had not got around to inquiring. So this same reader contacted General Electric, the manufacturer of some of the reactors involved in this accident. So the wheels may be in motion (we’ll see) to send both the 747 and a smaller DC-10 air tanker to Japan. Both had been parked, just waiting for a call.

Podcast: Play in new window | Download

These boys need bikes!

New York Magazine wrote recently that YouTube was planning to throw large sums of money at celebrities who would then make short form (three minute) videos for the site. The numbers mentioned were staggering (up to $5 million per celebrity channel) but the business model is crazy. It’s the three minute thing that makes no sense. I’m sure if YouTube is planning something like this it is specifically for videos that are not three minutes long.
Youtube already owns the Internet market for three minute videos. While there are probably instances where YouTube might throw some significant money into getting the odd celebrity to do something in this space, it is traditional TV-length videos and movies where Youtube actually needs help.
Looking at total video views, Youtube is the clear winner, but when it comes to longer-form videos, both Netflix and Hulu have more viewers than does Youtube. And Youtube can’t really afford to lose this battle, hence the emerging strategy.
Now let me tell you exactly where this is going, because if you are a couch potato it is important.
The big risk (or big opportunity depending how you look at it) has always been that Apple would spend $1 billion optioning TV pilots and by doing so effectively grab control of television. I’ve written about that right here. But somehow Steve Jobs was too cheap or didn’t have the confidence to know which pilots to choose (I suggested buying online rights to ALL of them, solving that problem). Only now it’s YouTube, not Apple, and it’s Netflix and maybe Hulu because once one does it they’ll all have to do it — even including Apple.
And the one to dominate in this land grab will be the one that spends the most money, with the key being to grab control of longer formats. YouTube already controls the three-minute video. It’s Netflix- and Hulu-length videos they’ll want next.
New York Magazine says Youtube is putting $100 million into such productions, but I can’t believe it will be that little, especially if other players choose to compete. We’ll easily hit that $1 billion number.

If that happens, the TV industry in the United States will be thrown on its head, because producers will be selling online rights first, denying those to the traditional networks. That opens the possibility that TV series may succeed online while never even making it to TV. Or they could succeed online and only later make it to TV.
In one sense it is the beginning of the end for traditional broadcast and cable TV, though visionaries might see it more as the end of the beginning. That’s how I see it.
The result will be an even more fragmented video market that will see lots more hits of all sizes from little vertical shows aimed at specialty audiences right through to Glee-sized hits that will work well because they have global reach over the Internet and can aggregate huge audiences without having to be a hit everywhere.
Some see emerging ISP bandwidth caps working against this but I don’t. AT&T is the first to impose such a cap in the USA for hard-wired customers but I am sure we’ll see exceptions for AT&T-provided content. Just as Comcast has bought NBC-Universal, AT&T will get in the content distribution business, too, if only to better compete.
Netflix is already rumored to be commissioning a TV series from Kevin Spacey. I’m sure we’ll see a lot of this happening and I think it is all good. After all, more video outlets probably means more Cringely, and all three of my kids need new bikes.

Podcast: Play in new window | Download

The gaping maw of Chernobyl.

Absent some terrible news from Japan this will be my second and last column about the nuclear accidents unfolding there. It turned out I was right last time about the sodium polyboride or boric acid or whatever neutron absorber the Japanese authorities dumped in first one and now two reactors along with a lot of sea water in an attempt to quench the reaction heat. I think it is pretty clear, too, that most of these reactors will not be coming back online… ever. This column looks at what that can mean for the nuclear industry in Japan and I also want to look at how these accidents are or aren’t like Chernobyl — a comparison I am seeing far too often in news reports.

To come up with a good comparison I turned to my friend Robert Bishop, one of the only Americans actually at Chernobyl back in 1986, helping fight that reactor.

These Japanese reactors are old and fairly well understood while Chernobyl was brand new. These Japanese reactors had already been in service for 16 years when Chernobyl melted down. In comparative terms there is no comparison — Chernobyl was vastly worse.

As just one example, when it came to poisoning the nuclear reaction in Japan they inserted the control rods, dumped-in the boric acid, then allowed sea water to enter the containment. In the course of all this the roof of the reactor building exploded revealing the containment vessel, inside of which lay the reactor core. But at Chernobyl it was all so much simpler — no reactor building, no containment vessel, just the out-of-control reactor, standing in the rain, emitting radiation.

Poisoning Chernobyl, according to Robert, came down eventually to dumping tons of powdered bismuth directly into the gaping maw of the reactor core from above. “It was almost instantaneous, ” he recalled. “Bismuth is a good neutron absorber, but more importantly it went through two complete phase changes from solid to liquid to gas, absorbing huge amounts of heat from the core, which was cooled by hundreds of degrees in just seconds. ”

It was a clever gambit, poisoning Chernobyl, but remember that somewhere between 30 and 60 workers died in the process leaving a further legacy of birth defects in the region.

These Japanese nuclear accidents come down to the simple fact that nobody back in the 1960s designed nuclear plants to run for 40 years then go through an 8.9 earthquake. Nor are today’s nuclear plants probably designed to that standard, which means Japan is facing what will by necessity be a significantly different nuclear future.

We’ll see rolling blackouts for months, maybe years, in Japan and the new nuclear plants that replace those old nuclear plants will be vastly different, too. If I were to predict a clear winner in Japan’s new nuclear future it would be Toshiba with its innovative 4S (Super Safe Small and Simple) reactors.

Japan needs increased generating capacity fast. They would like to replace nuclear with nuclear. But the new plants also have to show they can survive an 8.9 earthquake and reduce the number of critical failure points. Toshiba’s 4S reactors, which have been around for several years now, though not yet commercially successful, do all that quite easily.

4S reactor cores are like nuclear building blocks, built on a factory production line and transported by truck to be installed 30 meters under the ground. Each 4S puts out 10 megawatts of electricity or enough for 2000 Japanese homes. Following this path means the lost 1000 megawatt reactors will need 100 4S’s each to replace them or a total of 1200 4S reactors. 4S’s are fueled at the factory, put in place to run for 20 years then returned to the factory for refueling. They are sodium-cooled and pretty darned impossible to melt down. If the cooling system is compromised they automatically shut down and just sit there in a block of sodium.

The biggest problem facing the 4S has been regulatory approvals, which would normally take in aggregate 100 times as long (and cost 100 times as much) if done the same way as a larger nuclear plant. That’s where this earthquake will probably change everything, at least in Japan, where the process will be streamlined almost to nothing with a 4S soon stashed under every power substation giving Japan a smart grid in the process.