Saturday, September 18, 2010

Dejected of Fulham


I ran the Treeathlon 5km race today in a new personal best of 21m 27s. That's a full 75 seconds faster than my previous best. I put it down to the fast, flat course in Battersea Park (and figured had I not had the week I'd just had, I could have run even faster).

After the race, I looked at my Garmin GPS watch more closely to find that it had recorded the distance as 4.7km. I've also been trying out the new Nike+ iPhone GPS app and had set that off at the start, and it too recorded 4.7km. Two different devices, same distance recorded, makes it likely that they're right, and the race distance was wrong. I ran home the long way and the devices agreed again, on a distance that I know having run it dozens of times.

Sadly, then, there's no new PB. Had the course covered the full distance, I think I would have needed all of those extra 75 seconds, and probably a few more, to make it home.

The question, though, is why wasn't it a 5km course? Did they measure it incorrectly? Did all of the runners in front of me take a wrong turn and miss out 300 metres of course? Was this an effort at a feel-good conspiracy? Will this be a new trend ahead of the Olympics designed to make all Brit runners feel like they're getting faster?

Crowd Sourcing Running Photos

The Internet is a wonderful thing. 18 months ago I wondered why there wasn't a service where you could upload photos that you'd taken whilst watching a marathon or some other race. Turns out there is, and it's 3 years old today, much predating my musings.

You can find them at All you need, if you're a runner, is your bib number - you can then search against all races they've covered.

Wednesday, September 15, 2010

When Upgrading Can Break Your Business Model

At a government department where I work, the drinks machine, known affectionately as R2 (it is quite the most wonderfully complicated drinks dispenser ever invented), has received an upgrade, of sorts. The old R2 charged £1 for a bottle of diet coke (or any other soft drink). In a dozen simple steps it found the right bottle, took it from the shelf, arrives back at the dispensing slot and lets go.

If you put a £2 coin into R2, you'd have thought it would dispense a single £1 coin as change. No, it returned: 50p, 20p, 20p, 10p. With everything priced at a pound and most people paying £1 (I imagine, R2 rarely ran out of change.

Presumably as part of the cost saves, or as part of a revenue driving initiative, the price has now gone up to £1.10. Now what happens with change? Few people will likely carry the extra 10p so I can imagine routine payments being £1.20, £1.50 or £2. R2 will have to dispense more change and so will run out of change - and thus cease to function at all more often than it has to date. Only Dan could work out the likely frequency of this so I will leave to him.

That would all be fine, to some degree at least, but all the fiddling around to change the prices on each drink seems to have disturbed R2. Since the price upgrade, R2 has been in all of a dither. You put your money in, he goes to the right drink slot, grabs the drink with his robot arm, levers the drink out of the slot ... and then drops it. He's no fool so he knows he's dropped it ... or maybe he's not sure that he ever had it, so he asks you again what you want. This goes on - fetch, drop, fetch drop - until you get bored and ask for your money back.

For the sake of 10p a drink, the daily average revenue from R2 has dropped to zero.

Sunday, September 12, 2010

On barrels

The difference between french-style oak and Piedmont-style oak ... 225 litres versus 10,600 litres. The former are usually new every year, the latter are 50 years old in this case. Both are used to make Barolo.

That's not a barrel:


This is a barrel:


By the by, for French oak, the trees have to be around 150 years old. That was the legal minimum until a few years ago, when the age was lowered slightly, sometimes to around 120 or 130 years old. And for one majestic 150 year old oak tree, you will be lucky to get three oak barrels - more usually two. Of course the off-cuts go into making furniture and other wood products, but that still means a lot of tree for a little barrel.

Question 4 - Trends in Digital Delivery

I missed the cut-off for responses on the directgov review website so am posting my thoughts here and will email them separately to the directgov team. The comments on this question share many similarities, suggesting only open source be used (others say mostly open source), most support open standards, one says "give all unemployed people a smartphone and some free credit for accessing certain gov websites", one more says essentially that not every service can be digital so don't try for 100% and those that you do put online you should put in the right place. One more says:

Government must understand, study and perhaps consider how it might support the trend towards an active, contributing citizenry online, especially in relation to improving public services. It remains an open question, however, whether and in what settings government itself should aim to be the author of such online initiatives.

A real spread of responses then, from the thought-provoking don't force everything into a digital model, to the consultant-like blockquote pulled out above, to the banal mantra of open source software roolz. There was though, in the 26 comments left, two common thoughts: mobile, open standards / APIs.

Here, for what they're worth, are my thoughts on the trends that directgov should latch on to:

- Increasingly content is being rebadged and reused (not only through retweets and other social links but through white label websites and RSS pulling as evidence by the recent launch of the directgov content API). I've said before (probably 1,001 times) that I see no point in government authoring the same content hundreds, or thousands of times and hence am an ardent supporter of a single central government website that holds the definitive citzen-facing version of any given content (recognising that super-specialist content to do with e.g. accountants is best left to HMRC). The 309,000 instances of "disability living allowance" that google throws up today shows many in e.g. surrey county council, edinburgh, renfrewshire, hounslow and others. Sadly the number of instances is up from just 9,000 in 2003 and 122.000 in 2007. One definitive source, then, who makes their content available to others if they need it - with the accent on "if they really need it"; just borrowing content so you can boost the pages on your .gov website is hardly worthwhile (I really have no desire to comparison shop government content - no one is going to look for the best version of "disability living allowance"

- Content will increasingly need to be made available in multiple formats and styles. The proliferation of screen sizes and browser styles will mean an ever increasing number of style sheets that ally with the content to ensure that whoever is looking at the content gets the best experience.

- With content increasingly being made available, raw data has now followed and this trend shows no sign of slowing down. Plainly, people will have to use the data to warrant it continuing to be made available given that there is an inevitable cost of opening up the data and then keeping it available but so far, so good. This data will be made available in accordance with published standards, open ones whereever one exists.

- Content through apps. The debate over whether government should be in the app game or not continues elsewhere on the web (most eloquently by Stefan Cz at Public Strategist). My own take, as written before, is that government should be in the app game. At the same time, government isn't the only player in the app market and 3rd parties can often do better jobs (my current favourite is FIPlab's "London Cycle" - with something as good as this is, I see no point in anyone, let alone government, trying to do another version until or if FIPlab drop the ball).

- Transparency is an increasing part of the modus operandi of this government and there will be an increasing push for information to be made available that wasn't previously made available. The round robin PQ on the cost of government websites that has now turned into an annual review by the COI shows how this can happen. Government should get out ahead of this and publish what a site costs, what pages are being visited, how many people are visiting, where they're from, how did they get to the site (direct, via google/bing, etc) as well as stats on operating systems, browsers and mobile or desktop. Couple this with what the team looks like that runs each website, how many authors there are, how many editors and so on and quite soon the anomalies of government on the web will show up, helping drive an increasing focus on the cost of delivery and, specifically, where government is finding itself outflanked by other sites or providers who are doing what government should do only better.

- A long time ago, we ran open forums on We ran them first without moderation, then with post moderation, then pre moderation and then we closed them down. The problem was that Godwin's law, or variants of it, was universally followed. That is, within a few iterations of a post and its comments, everything would descend into increasing acrimony, and racist or other -ist comments would follow soon after. In restarting ukonline and then developing the model for directgov, we did a fair amount of research into bringing those forums back - because we thought that people who had been through complex processes (one of the models we looked at in detail was that for statementing a child - a 6 month process - as we had people on the team who had been through it) would be better able to guide others through it than government people would and that no amount of text on a website was a substitute for one to one guidance from skilled practitioners or people who cared. I'd like to see this thinking resurrected harnessing the social network trends. The forums need not be on a government website, indeed perhaps they should be anywhere but on such a site. They could be on facebook or on some or all of the more specialist social network sites; they'd be supported by government staff who would have the goal of getting people through the process and who would be measured based on feedback left by those they'd helped. But, in reality, it is likely to be the people who've been through it that prove the best at helping others.

I don't, though, believe that open source is an emerging trend or one that government needs to slavishly adopt. It should do what it's always done and use it where it makes sense, just as we did in building ukonline and directgov in 2001/2/3/4 and just as we didn't when building the government gateway in 2001 (for which we got no end of vitriole).

After Adoption 10k

The 2nd After Adoption 10k is in a few weeks, on 24/10/10. Sign up here. Feedback on the race last year was very positive so definitely worth a go this time.

Saturday, September 11, 2010

Two Months In Nokia's Life

Back in June I questioned whether Nokia might be a new entrant in the f*dcompany,com hall of fame. Until this week that looked a distinct possibility. Nokia shares are down 60% since the iPhone was introduced 3 years ago and until 7 days ago the downtrend looked to be continuing, as the graph of 2 months worth of daily stock prices shows (data from


Yesterday's announcement of a new chief executive (preceded, naturally, by the sacking of the old one) is seen as the catalyst the company needs to change. Stephen Elop, the new chief, is ex-Microsoft (but also ex-CEO of Macromedia, the original developer of Flash). His record is of turnaround and growth.

Symbian is not having a good time. App development is low (no users means no customers means no apps) and the o/s doesn't seem to have adapted to the touch screen / image rich / socially networked world. Perhaps the N8 due out soon will change that (my take on the N8, purely drawn from online reviews, is that they've thrown hardware at the problem so as to hide the lack of progress in the software; it has a great camera, great video etc but that's about it).

Developing a new o/s or even enhancing a current one will be a big job - and so there will be plenty of questions about whether to adopt Android or even Win Mobile 7 given Mr Elop's doubtless flawless contacts in that space. Or perhaps a multi-o/s strategy - why back one horse from the get go? Build on several o/s and upgrade Symbian in the background might work. It will take more than that to sort out Nokia though.

Mr Elop, despite Nokia's strong performance everywhere other than the USA and Europe, probably doesn't have the luxury of time. He'll need to describe and start executing on his plan very quickly. But the man at the top can make a difference - when Steve Jobs returned to Apple it was worth barely $2bn; it's now worth more than 100x that. Is Mr Elop going to do a Jobs on Nokia? His track record - the lack of experience in consumer product - suggests not. But he'll get the benefit of the doubt for now I think and rightly so.

It will be interesting to watch. I hope he returns Nokia to its former status as one of the innovators in the world, even if not to the kind of market share they enjoyed for so many years.

Wednesday, September 01, 2010

Boris Bike Robbery


Or did someone not get the supply/demand forecast right for each location and now they have to swing more bikes than expected between racks across town?

Took my first Boris bike ride today - two rides in fact - and came away hugely impressed. Easy to dock and undock, good quality bikes, no mechanical issues. Brilliant job.