Some of NZ’s biggest names in tech investment have announced a preliminary plan to develop a new high speed international broadband service, in competition with the current incumbent, Southern Cross.  Pacific Fibre has been set up by Stephen Tindall, Sam Morgan and Rod Drury, together with John Humphrey, Mark Rushworth (ex-Vodafone) and Lance Wiggs.

I’ve long argued that what NZ lacked wasn’t fast broadband to the home, farm or small town.   If those communities want it, let them pay for it themselves (perhaps through a locally subsidised utility since it isn’t economic outside the CBDs and central suburbs of NZ’s significant cities). What we do need as a nation is superfast, attractively-priced telecommunications pipes between our major centres and the rest of the world.

It’s that need Pacific Fibre plans to address. At this stage, all they really have is an idea and an intention, so activity is focused on planning, partnering and funding:

The group is looking to secure funding and build a 5.12 Terabits/sec capacity fibre cable to be ready in 2013 connecting Australia, New Zealand and the USA – the initial proposal is a cable which will deliver five times the capacity of the existing Southern Cross system.

I’m told by chums in the broadband business that Southern Cross has plenty of capacity available now and in future, through relatively straightforward upgrades, but maximises its profits through high pricing, which discourages heavy broadband users such as film or online services being based in NZ. Pacific Fibre may simply be a PR pressure play against a de facto monopoly, but so far it looks real enough.  I do hope that a viable alternative can get traction.  Competition is always better than an unregulated monopoly - not just for price but also for quality, service and, when things go badly wrong, for back-up.

So, Rod & Co., I’m keen. Call me.

Disclosure: My family trust has been a long-time shareholder in Telecom, part-owner of Southern Cross.  However, the trust’s investment manager has sold out now.

Update: Lance Wiggs has posted some technical details on the PF website.

Like any Telecom XT Mobile customer, I’m unhappy with the recent spate of outages, but I’m sympathetic to the harassed engineers at Telecom and Alcatel-Lucent striving to find and cure the problems.  Anyone who’s ever built a large new system dreads a spate of apparently unrelated problems which act like pouring acid into an open wound.  The shrilling and wailing of customers, competitors and often-ill-informed commentators does nothing to help.  However, I’m struck by one curious feature of the New Zealand environment which should have alleviated the problem - roaming. One of the advantages of GSM-based systems is that roaming is very easy for customers and operators - it’s a built-in aspect of the technology and the business model. If I’m overseas, my phone selects a network from all those with coverage for my location (in my preset preference order), which I can easily override if I so chose.  But back home, even though I can see other networks on my phone, my SIM card bars me from choosing them.  That’s anti-competitive, as well as being a damned nuisance.I’m normally anti-regulation but telecommunication seems to need it. Operators naturally want to maximise revenue spent with them and not their competitors, but that should be achieved through pricing, service, quality and loyalty.  Local roaming won’t diminish payment commitments under pricing plans and phone purchase agreements.   If I was the regulator, I’d definitely be taking a hard look at mobile network operators barring their customers from using other networks (and while s/he’s at it, local and international roaming charges relative to own charges).

In a new twist, Wired magazine reports that driving distracts cellphone users

Routine driving impedes a person’s ability to relay information from a cellphone call accurately to a conversation partner and to remember key elements of that information, say psychologist Gary Dell of the University of Illinois at Urbana-Champaign and his colleagues. Although many drivers regard talking while cruising a straightaway as no harder than walking while chewing gum, “that intuition is incorrect,” Dell says.

The road safety legislators have it all the wrong way round!  Same result though; just a 180 degree different spin.

Users of mobile devices enabled with data capability: answer a Victoria University survey and be in to win $200:

Eusebio Scornavacca from the School of Information Management,  Victoria University of Wellington  is conducting a research project on New Zealanders’ use of mobile information systems in the workplace.

If you are currently using a mobile device enabled with data capability (e.g. mobile e-mail, mobile Internet, mobile business applications etc) for work purposes, he invites you to complete a survey questionnaire. It should take you approximately 12 minutes to complete the survey, which is anonymous and will be available until the 25th of December 2009.  By completing the questionnaire you will be entered in draw to win a prize of $200. Please use the following link to access the survey: http://vuw.qualtrics.com/SE?SID=SV_cBko9uZySNCz9Zi&SVID=Prod

Your participation is important and will help to improve the understanding of the adoption and use of mobile technology.  Should you require any further information about this project, please contact Eusebio Scornavacca at +64 4 463 6697 or e-mail eusebio.scornavacca@vuw.ac.nz  or Professor Sid Huff at +64 4 463 5819 or e-mail sid.huff@vuw.ac.nz.

Thank you.

Paul Quickenden sent me a link to a fascinating analysis of Google’s “cheaper than free” strategy, sharing its advertising revenue with companies (including telcos) building products using Google’s various tools and services. (This approach may also resolve the content vs search revenue debate, as well).  In effect, through Google search, maps, the Android phone system and Chrome operating systems, not to mention the growing richness of Google’s mail, calendar, office applications. Google sites, etc., etc., etc., why would anyone build their products and services using Microsoft et al, for which they have to pay?

Naysayers of these assertions will likely have the same retort – quality is key. They will argue that Google’s turn-by-turn apps are inferior to their well honed market leading products. With regard to Android, Google will lack the user interface or embedded software expertise necessary and will deliver a subpar product. Plus, because the Android OS will be so splintered, QA testing will be difficult and incompatibility issues will abound. In the short run, these issues will exist.

Despite these challenges, it would be a dangerous strategy for any of the many threatened players in these markets to hang on to this “quality” rationalization for very long. First, Google’s products will get better over time. The sheer volume of the Android phones in the market will give them new data feeds to complement their own mapping effort. Also, they can create UGC hooks for users to embellish their own maps (like in Google Earth), offering themselves further differentiation. With regard to Android, version 3 will be better than version 2 will be better than version 1. Microsoft knows this game well.

Another perhaps even more important factor is that when a product is completely free, consumer expectations are low and consumer patience is high. Customers seem to really like free as a price point. I suspect they will love “less than free.”

I suggest you read the whole article before bursting into comment mode. This isn’t so much about the changes for the end-user. This is more about the impact on makers and purveyors of products and services, and the platforms upon which those products and services are built. Google’s growing dominance will make Microsoft’s near-monopoly look trivial. I can hear the anti-trust lawyers sharpening their knives already.

PS. MS still has very powerful offerings in other areas (eg. .Net), not to mention huge familiarity and comfort within its installed customer base, so don’t write it off yet.

Disclosure: My family trust owns Google and Microsoft shares.

How do you solicit customer feedback, how do you track it and manage it, and how do you respond to complaints? Probably, like most companies, it’s a Cinderella business process, conducted out of the spotlight. Some online trading sites (eg. Trade Me) have public buyer and seller ratings.  But what if you’re a “real” business, dealing with hundreds or thousands of customers? I’m quite taken with how London shirt maker Charles Tyrwhitt does it.  I normally buy shirts in-store when I’m in London, but I’ve started using their online shopping site.  Once my order was shipped, I received an email linking me to a online customer feedback service run independently of CT by Feefo.  I could rate each product and service, as well as post a comment.  I could browse all the other (anonymous) customer feedback (overwhelmingly positive by the way), and see CT’s responses to complaints, which within 24 hours typically apologised for any dissatisfaction, explained what had happened, offered no question refunds if required, promised an immediate followup by email/phone to understand more, and so on.

Showing complaints and replies in public, balanced with all the positive feedback, credibly portrays Charles Tyrwhitt  as a concerned and responsive company wanting to look after its customers, and making no bones about it.  You can also see the other companies using Feefo, and I expect this has a community endorsement effect.  I’ve certainly started looking at what those other Feefo clients have to offer.

UK IT industry analyst TechMarketView has published a review of the UK software products industry (payment required):

“There is a tendency to write-off the UK software industry because most of the familiar software companies are in the US. That would be a big mistake…  organic and inorganic growth has resulted in the top 50 UK headquartered software companies growing c20% in the last year. Perhaps even more surprising, to an ever sceptical British audience, is the c70% of revenues that the UK software industry earns abroad. Indeed, the UK is not that far off earning as much from overseas markets as we buy in. Currently £4.6b plays £5.6b with the gap narrowing each year.”

“The problem for the UK software industry has never been the quality of its people or its innovation,” TechMarketView chairman Richard Holway said, instead blaming other factors holding the UK industry back:

• Lack of available financial backing: In comparison to the US, it is much more difficult for UK software developers to gain access to venture capital funding.
• Lack of marketing expertise: Many of the UK’s best developers simply fail in explaining how great their product is to investors and to the market.
• Local not Global: Many of the UK’s software companies focus mainly (if not solely) on products geared to the UK market.
• Lack of ambition: Many UK software companies are run as ‘lifestyle’ businesses. Very few UK software entrepreneurs seem prepared to risk the Merc for the seemingly scant possibility to become a global player.
• Lack of management skills: Growing from a small enterprise to medium-sized is hard enough - but not a fraction as hard as that required to grow to be large. Few founders are up to running large, global organisations; even fewer are prepared to step aside!
• Easily pleased: UK software companies have a long history of being other nations’ ‘acquisition fodder’. Founders seem to want to ‘take the money and run’ rather than take the risk of growing to something larger.

New Zealand (and Australian) industry observers will find the list very familiar, except perhaps for one thing.  When you’ve only got 4 million people in your home market, international ambitions are needed pretty much from the beginning, adding the complexities, risks and costs of heading overseas while still very small, unsophisticated, and fragile. Given that struggle, it’s no wonder that many are satisfied once they’ve got “the boat, the bach, and the beamer.”

PS. bach is Kiwi vernacular for a holiday cottage

It’s a rhetorical question.  The system owner has the ultimate responsibility.  Yesterday, Air New Zealand suffered a catastrophic failure of its core passenger handling systems.  I understand that loss of the external electricity supply was followed by failure of the data centre’s back-up power generator. Chaos ensued.  Air NZ was very quick to divert blame to its outsource IT service provider, IBM. I don’t know whether the criticism is fair, but I am not impressed with Air NZ’s actions.  Rapidly heaping public blame onto a key supplier doesn’t help solve the problem. When it’s IBM, one of the most respected and experienced  IT service businesses in the world and almost certainly more able at running data centres than a small (in global terms) airline, then I would be very cautious.  Hold your breath, count to ten, and wait for the analysis.

Irrespective of what went wrong, however, let’s be clear about one thing.  It is Air New Zealand’s responsibility to make sure its systems are robust and can cope with failure.  Airlines should have hot load redistribution via multiple distributed load-sharing  data centres.  That hot fail-over needs to be regularly tested, not by easy planned tests, but by kicking out the power cables on the server side of the UPS (uninterruptible power supply, basically a big battery that provides smooth power and controlled power-down if  external power fails for more than 15 minutes, say). IBM should have done whole-system failure testing, but more importantly so should have Air NZ - it is their system.

I’m back. Don’t worry; I won’t bore you with my travels and pictures (not even the trout I caught fly-fishing in the Yorkshire Dales).  I didn’t write anything here after the first few days away. Not that there wasn’t anything to read or write about, but I found the break too relaxing to want to break the mood.  After the first week when I did some business stuff, I made and received very few phone calls. I hardly touched the internet, apart from online booking sites and Google Maps.  I did check emails daily (via my phone), deleting most of them, filing some for attention later, and responding to a tiny few.  I only looked at my Google Reader news feed once or twice to keep up with people I knew, but ignored everything else.  It’s very liberating to delete thousands of items unread!  All in all, it was a pleasant reminder that being “always on”  isn’t a necessity.

OK - maybe just one picture.

Guns don’t kill people; the people who fire them do.  The same is true for Microsoft’s ubiquitous presentation software. PowerPoint is actually a very good tool; but lethal in the hands of bad presenters.  PowerPoint didn’t make them bad presenters.  They’d have found another weapon to kill their audience.  Do you remember those 35mm slide carousels, the flight bags full of overhead slides, and the stacks of flip-charts. I remember a university lecturer we called “Scribbler”; in a one hour lecture, he’d get through a roll of acetate on his overhead projector, writing constantly, reading out what he was writing, eyes glued to the foil and never once engaging with his audience.

The BBC has noted that PowerPoint is 25 years old. PowerPoint doesn’t deserve the opprobrium it receives. It has deserved its 25 years of successful sales.

Poor presenters? Just take them outside and shoot them.

Tom Forenski at ZDNet writes that “the internet devalues everything it touches“. He uses online business applications, outsourced call centres,  and online shopping as examples of how costs have come down for vendors and their customers.  To label that as “devaluation” is a strange way of describing what has happened. While in one economic meaning he may be right, ie. the unit price has come down, the value of a good or service to a customer (ie. its utility or benefits) doesn’t decrease when it is delivered via the internet.  On the contrary, its value may well have increased, through greater convenience, speed, and so on.  Or it may have become more affordable and accessible to new customers.  Likewise for a supplier.  I’ve never met someone in business who thought that increasing efficiency, effectiveness, customer utility or market reach for lower cost doesn’t enhance or at least maintain shareholder value. While unit revenue may come down, competitiveness may have been enhanced.

Lean manufacturing created much greater value for customers and shareholders (or at least those whose companies who adopted it).  The lean business processes made possible by the internet do the same.

And I’m sure Forenski knows that.

(Thanks to Paul Quickenden for alerting me to this).

Over at The Technium, Kevin Kelly asks “Was Moore’s Law Inevitable?” You’ll recall that Moore’s Law predicts a doubling of transistor numbers in an integrated circuit chip every two years, and it has held astonishing true for nearly 50 years.  Fulfilling this “law” has been a key driver in the astonishing growth of computer power accompanied by falls in the cost of that power.  Kelly explores whether Moore’s Law simply sets a target which engineers strive to achieve (2 years does match neatly with typical product development cycles), or if there is some other deeper factor, and extrapolates this to a variety of technologies.

I was struck by Kelly’s observation that most current new technologies have gained momentum from smallness:

The first thing to notice is that all these examples demonstrate the effects of scaling down, or working with the small. In this microcosmic realm energy is not very important. We don’t see exponential improvement in efforts to scale up, to keep getting bigger, skyscrapers and space stations. Airplanes aren’t getting bigger, flying faster, and more fuel efficient at an exponential rate. Gordon Moore jokes that if the technology of air travel experienced the same kind of progress as Intel chips, a modern day commercial aircraft would cost $500, circle the earth in 20 minutes, and only use five gallons of fuel for the trip. However, the plane would only be the size of a shoebox! We don’t see a Moore’s Law-type of progress at work while scaling up because energy needs scale up just as fast, and energy is a major limited constraint, unlike information. So our entire new economy is built around technologies that scale down well — photons, electrons, bits, pixels, frequencies, and genes. As these inventions miniaturize, they reach closer to bare atoms, raw bits, and the essence of matter and information. And so the fixed and inevitable path of their progress derives from this elemental essence.

At antenna manufacturer Deltec, we were frequently asked why our unit costs didn’t fall as fast as the electronic subsystems in mobile communications infrastructure.  While we did achieve cost savings from smarter design, leaner manufacturing and experience/scale effects, the laws of physics imposed limits on what could be achieved in a shaped piece of metal whose minimum size was determined by the frequency/wavelength and performance requirements of the network design and cellsite coverage. Smallness only came from shorter wavelengths/higher frequencies or lower range/coverage (increasing the number of cellsites).

Kelly also unearthed this old chart from the US Air Force.  Impulse power, Mr Sulu! You’re due about now.

People often think of contractual and business relationships lasting a few days or years, but rarely think about ones that might last decades and beyond.  I’ve seen several organisations bitten by seemingly unimportant or just plain forgotten undertakings given in the dim and distant past.

One  company received a letter from the city council announcing that the city would start charging the company for 10 up-til-now free CBD carparks that the city provided.  No big deal; after all, who expects to get a free CBD carpark these days? But one of the company old-timers remembered that the city had traded the free CBD carparks some 15 years earlier for some company-owned carparks where the city wanted to build a motorway.  Imagine the value of 10 car parks in the CBD close to the motorway exit - in perpetuity. A search of the city’s records (better than the company’s) produced the documentation. After some haggling, the city bought the car parks back - a very nice windfall for the company given CBD property inflation over that time. It was pure luck that someone remembered the original deal. Bad luck for the city; good luck for the company.

It’s not just property deals:

• A letter offering a job to an overseas immigrant which promised repatriation if he was fired or made redundant - which he was, 20 years later.  He had no intention of returning, but it paid for a nice visit home.
• An enduring exclusivity clause in a services contract.  After several years, the work tailed off.  Some years later still, the contractor was invited in by another company - a possible rival  The defunct client invoked the clause to stymie its competitor.
• And so on.

Partly these problems are down to poor negotiating, poor contract drafting and poor record management.  That’s not my point. In these cases, at least one of the parties had forgotten all about the original deal. In each case, someone remembered and was able to take advantage of the situation. It also makes you wonder how many times such deals have been forgotten by all concerned, and whether they won or lost as a result.

So who or where is your institutional memory?

I held off saying anything about the 40th anniversary of the first moon-walk because I had nothing to add beyond the usual awestruck adolescent wonderment of the time (which hasn’t diminished, by the way).  Fortunately, UK IT commentator Richard Holway reminded me of something that is very relevant in my own career track:

The computer in Apollo 11 had just 64K of RAM. It seems amazing when we’ve now even stopped talking about megabytes in favour of gigabytes. In 1969 though, much of my programming was still in assembler code and I was expert in reading the holes in punched cards. I worried constantly about saving a bit here and a bit there. This was when dumping the ‘19’ bit from ddmmyy was common - and thus was born the Y2K bug!

My thoughts today are not about ‘Wow, look how far we have progressed in 40 years’ . It is more a wonderment that mankind could achieve a mission to the moon armed with such seemingly archaic tools. Truly incredible.

Indeed!

Just 3 years after the 1st moon landing, I wrote a complete business application - orders, inventory, billing, accounting, etc. - to run on an ICL 1901A with 8kb of RAM and 2×4Mb of exchangeable disk drive storage, written in NICOL (ICL’s version of RPG1, taught to me by David Nicholson, one of my oldest friends, who by happenstance ended up in Wellington NZ a few months before me and is still here). I was a dab hand with the old 3-finger card punch too!

I suppose this officially makes me an old fart!

Here’s a curious thing for you to discuss in the pub after work tonight. Has the Internet shrunk our horizons rather than expanded them?

Last year, in “The Myth of the Telecommuter“, I noted that, despite electronic communication supposedly turning the world into a village, most interactions are between people within walking distance, followed by people within the same city. Prompted by similar findings, a new study of US birth data has looked at the spread of baby names over time.

The Economist reports:

… Dr Goldenberg and Dr Levy speculated that when parents chose a name for a child, they were influenced by their interactions with other new parents, so the spread of the names the babies were given was a proxy for the pattern of those interactions.

… What Dr Goldenberg and Dr Levy found was that the proportion of babies given a certain name in a state where that name was already popular or in a neighbouring state was 20% higher than would otherwise have been expected. This was true from the 1970s to the early 1990s.

From 1995 to 2005, however, the effect became even more pronounced. The proportion of newborns with common names in any given state and its immediate neighbours became 30% higher than would have been expected if there were no geographic effect. Dr Goldenberg and Dr Levy ascribe this rise to the internet. It certainly correlates with the emergence of the web, though whether the correlation reflects causation is unproven. But whatever the reason, it is a curious result.

The report’s recommendations needed to be implemented, permanent senior management was needed to replace those in acting roles, the IT system needed an upgrade worth $117 million over four years and the entire process needed to be taken apart and looked at “from top to bottom”.

That snippet is from today’s Dominion Post, commenting on the need for a major overhaul of New Zealand’s Immigration Service. I’ve already heard much about the shambolic state of this government agency, so the need for a root and branch renewal is not surprising.  What caught my eye was the size of the proposed IT upgrade.

I know that government IT projects suffer from very bureaucratic (and often ineffective) environments: unengaged and unempowered users, long-winded decision-making, overly complex legislative and procedural requirements, etc, etc..  Even if that isn’t always so, government IT will still usually be more expensive than commercial IT.  Unique requirements (well, nationally idiosyncratic, anyway) tend to demand bespoke solutions, or at least customised implementations of standard case management and workflow systems.  Allowing for that, $117 million still seems way too much.  After all, this organisation only does a few core tasks:

• receive applications, process them, and issue visas for tourists, students, temporary workers, and permanent residents.
• weed out dodgy applicants (at least that’s the theory).
• provide information on the process to potential applicants and employers.

Let’s look at the IT investment per process worker (a useful metric for process/people-centric operations). Assuming that the NZIS still employs approximately 750 people (the last number I could find) and that 2 out of 3 staff are engaged in the process (as distinct from support functions and executive staff), that’s $234k per person.  That is ridiculously high. What complex business process does this organisation operate that requires such a high IT investment? If I was the Minister of Immigration or the Minister of Finance, I’d be demanding alternative proposals for the business process and supporting systems.

Curiously, that $117 million is a very specific precise number, given that “the entire process needed to be taken apart and looked at ‘from top to bottom’.”   I’m smelling the pungent scent of of desperation - let’s throw lots of money at a big IT project to give the appearance of decisive action.   More cynically, it buys 4 years of plausible excuses while the project is underway, and in 4 years time, everyone senior will have moved onto pastures new.   Sadly, you’ll find similar stories in every government in every country.

Microsoft has announced that, after 17 years, it will stop selling its Money personal finance software at the end of June.  I’ve always quite liked MS Money.  It has its flaws, but it did the job reasonably well for our home accounts and investments, and I never saw anything else worth changing to (unlike small-business accounting, where I strongly recommend Xero).

When I opened my first proper bank account on starting university, the bank manager himself sat down with me to talk about managing my money. I don’t think that happens these days! The bank manager gave me a book in which I could record my income, expenses and bank balance, so I wouldn’t lose track of how much money I had and what I spent it on.  It was a habit I never lost, and I still have all those accounts books.  In 1996, MS Money came bundled with a home PC package and so I went electronic.  Should someone ask what we spent at the supermarket in June 1998, or our expenditure in bars and restaurants last year, or the return on my investment in Telecom, I have the data!

Some day, someone will find these records and earn a PhD by analysing the minutiae of  our family’s lifestyle as evidenced by what we earned, saved and spent. So now I need Rod Drury to bring out the personal finance edition of Xero, with investment functionality, please. Posterity demands it!

Disclosure: Isambard Investments owns Xero shares.  My family trust owns Microsoft shares.

So your marketing people have told you that you should be blogging or twittering.  But you’ve heard that most blogs have an audience of one, and most twitterers have only ever posted one tweet. Why bother?

Blogs per se aren’t a cure-all, but they are very cheap and immediate communications channels to reach large and small audiences.  Technorati - a blog tracking and ranking service - reports that out of 133 million blogs it tracks, only 7.4 million have posted anything in the previous 4 months (<6%), and only 1.5 million in the previous week (~1%).  Hardly a huge activity rate, one might suppose.  But that probably represents a shift in who’s writing blogs and why.  When blogs first started, they were typically personal online diaries, often with a technical or hobbyist bent.  The top public blogs today look more like news feeds, or rather commentary on the news, usually with a particular focus such as politics, economics, technology, sport, hobbies or entertainment. 1.5 million very active blogs is still a very big number. And businesses are increasingly using blogs to inform and connect with target audiences, such as customers or staff (perhaps behind a security screen, and so hidden from Technorati).

Micro-blogging service Twitter is still evolving (not least, it has yet to find a way to pay for itself), but it clearly has several business applications, especially “creating a buzz” around an event, product launch or special promotion. No doubt other business system applications for Twitter will emerge, but I’d hesitate to build Twitter into mainstream business processes until I was certain the Twitter business model is sustainable.

The numbers for Twitter can be deceptive.  The BBC reports that a Harvard study of Twitter users found:

• There are an estimated 10 million Twitter users.
• 10% of Twitter users generate more than 90% of the content.
• More than half of all people using Twitter updated their page less than once every 74 days.
• Most people only ever “tweet” once during their lifetime.

However, you need to bear in mind that to receive Twitter feeds, you sign up to Twitter, whereas most blogs are followed directly using your browser or a generic feed service.  That 10% active user rate is actually quite high, compared to blogs where the readership may be several thousands or tens of thousands every month (or even millions for the big-hitters).

So is blogging or twittering appropriate for your business?  They are only means to an end.  Like any communications, you have to define your audience, what objectives you’re trying to achieve, and for what duration, which may be long or short (for a specific project or event).  Blogs and Twitter are just some of the tools you might use.  On the question of duration, be aware that to be effective over an extended period requires organisational stamina.  Like all those personal blogs, many corporate blogs have withered away after an initial flush of enthusiasm, because the business really had very little new to say over time. Even so, it’s worth experimenting to explore what they can do for you.

So you’ve got all your business and family records on computer, you store all your email online, all your business documents are in a wiki, and your family photos are on Flickr.  You’ll never lose anything again, right? Wrong.

As any archivist will tell you, the half life of information storage is shrinking all the time, and they despair at the increasing pace of information loss. Messages written in stone lasted for millenia, documents written in natural inks on linen papers lasted for many hundreds of years if stored well; synthetic inks and industrial papers last a century, celluloid films several decades. But it’s the move to electronic media that has really sped up the losses.  Not only do these media only last a few years, but also the technology to read them (both electronic and software) rapidly becomes obsolete, unobtainable and unmaintainable. Anyone got any files on floppy discs? Can you do any analysis of your survey records from 20 years ago? Can you even find them, let alone read them? Take optical discs (CDs, DVDs, etc) - the discs get scratched, the surface film (in which the digital pattern is etched) breaks down, the plastic substrate breaks down, and the recording standards change.

For the time being, there is only one working solution - keep your data moving.  If you transfer your data to a new medium from time to time, you keep it accessible. That’s easier to do now cloud computing storage is so cheap. Create a document archive for everything online, copy everything into it, and back it up on a second storage service elsewhere regularly (ideally automatically). And from time to time, move everything to the latest current medium. But if you aren’t there to keep your data moving, who will look after it after you’ve gone, in 10 years time, 100, or 1000? Make sure others know where your information is stored (you probably want to share much of it with them anyway) and how to get to it if you’re run over by a bus.

For the archivists, a long-term solution may be on the horizon.  Wired magazine reports on a nascent carbon nanotube technology with the potential for data to be stored accurately for a billion years “from now until long after the Earth has been overrun by superintelligent, fusion-powered cyborg ants“.

Carbon nanotubes are molecular-scale tubes usually made of a carbon allotrope. For data storage, a small electrical signal is applied across the nanotube causing the iron nanoparticle shuttle to move back and forth. The movement of the nanoparticles from one end to the other of the tube creates the binary ‘1′ or ‘0′ state.

The position of the shuttle can be read out directly, explain the researchers in a paper published in the current issue of the Nano Letters journal. …The technique has significant potential for archival storage, say the researchers, because the nanoparticle-based bits show significant persistence. It’s also possible to store a lot of data in a small space: With information density predicted to be as high as 10¹² bits per square inch, you could store data from nearly 25 DVDs in the space of a postage stamp.

One of the weakest excuses you’ll ever hear is “Sorry, it was a computer error.”  My initial reaction is almost always derision.  Computers very rarely make errors, whether in reading or writing to a file, or moving data around (internally and over communication links), or in processing.  Computers have inbuilt self-checking so that errors are detected and overcome or at least alerted. None of these should be sufficient to actually cause the incorrect recording and processing of an organisation’s business with you. So let’s look at what really went wrong:

1. An error in understanding what systems and processes that the organisation needed to conduct its business with you.
2. An error in specifying that need.
3. An error in designing a computer system to meet that specification.
4. An error in building and implementing the system.
5. An error in testing the system.
6. An error in the business processes around the system.
7. An error in training people how to use the business processes and the system.
8. An error in entering information into the system.
9. An error in understanding what was said in your encounters with the organisation.
10. An error in understanding what was heard in your encounters with the organisation.

Errors 6-10 are almost always the cause of the problem, followed by error 1. All outside the computer system. And did you notice something else?  Every single one of these 10 errors is a HUMAN error.