Source

A lens in our eye focuses light onto our retina, our retina send signals to our brain, and our brain spends a huge percentage of its capacity, by far the largest of any brain function, just to decode those signals and recognise what’s in front of us.  We look for patterns, seeing lines and movement; and we learn over time from when we are babies how to interpret those things and figure out what we’re looking at. We also constantly scan; if our eye is rigidly fixed in one spot (through painful apparatus) the image disappears.

These basic biological mechanics are the reason that the cockpit of nearly all aircraft look something like this:

An early 747 cockpit.

Lots of analog dials, rather than digital readouts.  Dials work better than digits because our visual system lets us scan and notice when things are out of place; we instinctively zero in on things that are changing.  We’re also good at figuring out the direction and rate of change with analog dials. Experienced observers can often estimate where a dial will stop based on watching it while it’s still moving. Even in modern aircraft where digital displays are replacing analog dials, the key instruments are always present as analog representations.  The peripheral systems are monitored by computer software rather than the pilot.

Digital readouts, while undeniably more accurate, require much more mental energy to read and process. We just find it much harder to take a snap shot of what’s normal, and notice what’s different, when faced with a panel of 40 numbers vs 40 dials. This is also why we tend to skim reports, and zero in on the graphics, graphs and pie charts rather than tables of figures or words. We decide what to look at from the visual cues, and then check the relevant reference data only for the more interesting items to work out how and why.

So as we apply this understanding of our visual system to an Agile context we can see how Agile boards are very suited our our natural strengths. We’re good at noticing when things change, move or something looks out of place.

Boards also give us a good way to use our instinctive ‘gut’ feedback system – over time teams and even people outside the team get very good at knowing what a good ‘on track’ iteration or cycle looks like – and equally when things are off track.  It’s often not as simple as one or two cards or tasks, but noticing the natural weighting and flow of the system changing over time.  It’s a worrying thing then to hear from a team talking about their Agile board – ‘It works ok, but doesn’t change very much’.  It’s also why habits like putting fences and rows on cards that are stuck is important; it reminds us to check back later when we naturally gravitate to the shiny distraction of the things that are moving instead.

At the beginning of June 2012, the Luna Tractor network will farewell Nigel Dalton as a consultant, to take on the role of Chief Information Officer at REA-Group. As one of the founders of Luna Tractor, Nigel will continue to write here, sharing his insights and experiences as part of our growing network of executives, consultants, leaders and change agents.

Over the last twelve months Luna Tractor has worked with many different organisations, from banking, superannuation and health insurance, to handmade bicycles and professional photographic services. We’ve helped them find new ways to work, with new approaches to innovation, transforming business by linking customer insight with strategy and agile execution. Co-founder James Pierce will continue to lead this mission hands-on with all of Luna Tractor’s partners.

Nigel is currently in the USA where he is speaking at the Lean Software and Systems Conference (LSSC) in Boston. Both Nigel and James with be back on stage along with Luna Tractor clients at the Agile Australia 2012 conference at end of May to continue sharing their stories. We look forward to seeing you there – Nigel and James.

Working software over comprehensive documentation – one of the tenets of the agile movement that was enshrined on the agile manifesto over 10 years ago. But it is often hard to convince grey-haired old men carrying sharpened stakes (aka the stakeholders) that when things go wrong, the safest path is NOT for someone to look up the very thick and comprehensive ringbinder your BA team created at release time, find the page with that code on it, and fix it.

Back in the day, a whole industry grew up around documentation and manuals for software development projects. You could even get a job as a ‘technical writer’! I’m glad to say we don’t hear much from those people today, but I have no doubt they still exist somewhere in the land of waterfalls. Plenty of people still make loser manuals, as Mary Poppendieck is wont to let slip. And between James and I, we must have tried ring binders, post-it notes, wikis, white boards, powerpoints, omnigraffle, photos, blogs, twikis, email, text files… just about everything except stone tablets.

Luna hero Donald Knuth would tell you that to some extent the software should be its own manual (aka literate programming) – if fortune smiles upon you, it will be able to be picked up by a later developer and sorted out, based on the use of well-known programming techniques, sensible structure, and the occasional comment.

Architecture is another matter again – agile talks about it being a shared responsibility, about anyone being able to draw it on a whiteboard at any time. Good principle, but things can get quite complex at times – most people don’t want to know everything, just where the dragons be.

Alistair Cockburn made an interesting point at Agile Australia 2011 – that principle #1 of the agile manifesto was about the importance of face-to-face communication, and principle #2 was around minimising documentation. What if we combined these two ideas? In a world full of Youtube consuming, non-reading, video-savvy people, surely a short piece to camera was the next best thing to having the creator of a piece of code on the spot to consult?

So, from a small scribble in my To Do list, was born a video documentation project to take some risk out of moving the Lonely Planet website with a new team being started up in London – 18,000km away with all new people.

The idea was simple – 10 videos, less than 5 minutes each, hi-res with a Flip camera so the whiteboard drawings could be easily seen, and the focus being on pointing out the dragons. Explain the interfaces, where test coverage was weakest, some detail around the databases (recorded brilliantly with a screen grab tool by the DBA) and a great lesson in all the gotchas. The lowest tech thing we could do was label them well, stick them on a USB drive, and ship them to London.

Great idea for release notes, progressively recording improvements to code. A now out of production Flip is only going to cost you a couple of hundred dollars on eBay.

If nothing else, it might avoid the kind of developer comments we once found in some packaged software code -”what kind of f@#$tard wrote this? I have no idea what this is for!”

Give it a try with your iPhone and let us know how it goes.

1 April 2012

Luna Tractor has been leaked advance details of the hottest new internet property to emerge from The Y-Axis, Melbourne’s most successful  incubator and startup investor, proof positive of the benefits flowing from one of the nation’s first NBN hotspots in Brunswick.

“Airbnb were so conservative, they  never really showed signs of being brave enough to take their platform to its logical limits” say the Airbed.com founders in their embargoed press release. “It’s all very well letting out your front room, or your sister’s apartment while she is away for the

The bed as a platform.

weekend, and saying how cool and social it will be – but we all know what that really means for travellers! It’s about the bed as a social platform, stupid!”

It follows then that Airbed.com will be far more ‘mocoloso’ (mobile-cool-local-social) as a platform than any of its competitors. Mocoloso was the key term being used by the VC Community at this years SXSW trans-media conference somewhere in the USA, which none of the airbed.com founders could actually find, but they have been to America.

Slavishly following the Eric Ries ‘Lean Startup’ model, Airbed.com operate 100% in the cloud with virtually no servers, no software (“code is SO last Thursday” says the CEO), use an agile methodology, and focus heavily on the use of RESTful APIs. “Soap APIs were frightening off some of the bohemian clients from the Northside in our user trials” says the airbed.com CEO in the leaked release.

The startup team has ‘pivoted’ once already – their first website was called airbnd.com, a social travel guide that enabled travellers to locate and access B&D (bondage and discipline) services. “We ran into the iTunes store, and some silly issues over the logo, so we pivoted immediately.”

When asked about the obvious similarities to established Californian social travel business airbnb, the founders countered quickly: “this is the Y-Axis incubator’s innovation secret sauce – we’re fast followers. Mostly we like to copy people’s websites and use Powerpoint to cut and paste new logos and stuff over the top of old ones.”

37 Signals website fails to use the Y-Axis standard font of Helvetica. "It's a wonder they are still in business with that UX" say airbed.com founders.

“We saw the US dot com industry was now just duplicating 37Signal’s website as a standard business model. We tried that of course – but our Macs have had all the fonts deleted apart from Helvetica, which led us to airbnb’s site – they truly understand design in the Brunswick sense”.

There are already mature interfaces to travel social dating services like Grindr, and strong rumours of a 4Square API in the summer. “Essentially, airbed.com have totally out-localled the specialists with this offer” says internet commentator Buzz Reilly. “A room? That is yesterday’s business model – nothing could be more local than a bed. With a local in it, preferably”.

The iPhone app is planned to be in store in the next few weeks, with ‘epic’ new interfaces to accommodate its use when intoxicated. There are also strong rumours of a corporate platform being developed, though the rumoured partnership with Blackberry have been scotched by commentators. The likelihood of any of the airbed.com services catching virally is considered extremely high according to their PR Agency.

Airbed.com are looking to raise around $20m for Series B, having begun their business using only the Burger King sponsored free wifi on the 86 Tram and two iPads, occasionally having to get off to avoid MYKI inspectors and to source long macs.

“Actually, we were hoping for a big mezzanine, but some douche put it on airbnb.com, some Japanese college students booked it, and there was no room for us” notes the disappointed airbed.com CFO and founder.

Click through to see the next 21 billion kilometers (or just under 20 light hours…)

A human institution designed to create a new product or service under conditions of extreme uncertainty.

So what does ‘extreme uncertainty’ look like?

Po Bronson coined the phrase ‘radical uncertainty’ 13 years ago in The Nudist on the Late Shift, the book that got me so excited about San Francisco’s dot-com era I left a perfectly good career in Australian corporate IT to chase the dream of startups in America. And those conditions of uncertainty are what led me directly to the agile way of working (thanks Kent Beck), as the only way to cope with business models, shareholder requirements, and customer needs changing on a weekly basis.

Here’s some of the consumer expectations that are driving that kind of uncertainty today, in a brilliant infographic from the team at onlinegraduate programs.com. Grab your iPad, a stop-watch, and possibly a glass of whisky, and head to your own website immediately after reading. You have 4 seconds or less…

Created by: Online Graduate Programs

They love feedback down at onlinegraduateprograms.com so drop Tony Shin a note with any feedback using Twitter dm @ohtinytony

It boggles us! How can this be? Learning is the hot topic du jour!

Eric Ries just reiterated his 10c worth that ‘validated learning‘ is the only true measure of progress when you are building a product or company under conditions of uncertainty.

This has caused a bit of handbag swinging in agile circles, as ‘learning’ isn’t anywhere near as easy to measure as a team’s velocity (the throughput of development stories); or a burndown chart (points per iteration); or cycle time (time for a story to turn into cash). His point is that while you may have made good progress towards achieving a plan, that plan may actually be leading nowhere useful for customers.

“Validated learning is not after-the-fact rationalisation or a good story designed to hide failure (btw I hate those smart-ass CEO put-down quotes about ‘experience is what you get when you didn’t get what you wanted’). It is a rigorous method of demonstrating progress when one is embedded in the soil of extreme uncertainty in which startups grow.”

Revenue, no matter how small (in one of Eric’s startups $300 a week), is one good way to validate you have learned something useful.

It’s therefore obvious we all want learning – it makes us smarter and richer (and more like Eric Ries) for a start!

The science of learning is well documented – there are several theories, which if you happen to be an educationalist, we hope you are familiar with. One or more will likely have been inflicted on you by 10 years at school and perhaps more at a tertiary institution. We’ll side-step the science here for a minute.

Instead, we’ll talk about guitars.

As it happens we are both musicians – James an accomplished trumpet player, having played for years, and me, a struggling noob rock guitarist. It’s a skill I chose to pick up in my 40s, and whilst reading about strategy, lean and agile is a painless habit for me, guitar playing has been a start-from-scratch nightmare.

But what the guitar has been for me, I realise picking up a book or learning about agile is for others. Far out – no wonder people avoid learning!

The first thing about learning the guitar is the pain. Pain in the note-holding finger-tips (and some fine callouses to boot); along with pain in the left wrist, hand and fingers (I have a Bb7 claw hand at the moment). The pain comes from contorting my long-standing limbs and digits into new shapes and combinations.

That physical pain is nothing to the pain of emotional embarrassment at hearing notes and chords come out hideously (in front of friends and family); the feelings of incompetence that weigh down my shoulders; and worst of all the nagging, uncomfortable pressure just under the solar plexus that is my bruised self-esteem. You just suck at it, for quite a while. Nobody wants to suck. So we (quite logically) just avoid pain and sucking by staying the hell away from learning.

The second thing is that learning is tiring. I’m trashed after an hour of practice – listening, trying, re-listening, trying again, playing along … keeping up with band-mates and having fun jamming leads to the best night’s sleep.

Photo by Jamie Supple

Thirdly, learning moves at different rates. Some days it just clicks, others a small chord change can take a week to master. It doesn’t pay to plan (in my case) that I can master a song a week for example – too much uncertainty in that! Then, just as my left hand gains competence at changing from an F to an Am chord, my right hand can’t master the strum pattern that makes it sound rich and funky. The ‘team’ are learning at different rates!

The internet both helps and hinders learning – willing amateur musicians seem to have tabulated (aka written the non-notation cheats for) every song every written, and loaded them up to sites like Ultimate Guitar tabs. But most of them are wrong! As you can see in the example above I spent literally months in the wilderness with that Fly My Pretties song trying to figure out why it didn’t sound right – only to have a competent guitarist point out it was a B-flat, not a B.

Learning works so much better with feedback – and not after a week of doing something wrong, mastering the wrong damn thing only to find the chord progression was different all along – and quickly deciphered by a guitar master (see the hand-written chords at right on the photo above). Ten to 20 minute intervals of feedback are the natural human learning cycle. Which makes you wonder how useful a weekly retro can ever be!

Most of the time is spent on my guitar doing it wrong, listening and adapting. It takes a lot of repetition, and I often find it is best to just walk away and come back later – and then be amazed what your muscles have learned while you were cursing those chord changes.

Photo by James Pierce

Launching songs for real into the world at a band gig, or even a band rehearsal is a world apart from practising solo. Everything you thought you knew from doggedly playing along to the original song on the iPod goes by the wayside as the team rhythm takes over, and what matters is the skills and techniques more than the precise process of hitting all those notes in order. The order, in fact, can change with alarming speed!

And finally, a good teacher makes all the difference. I am very lucky to have several, professional and talented amateur in my life. And now I can reflect on why coaching is such a fundamental part of learning to play agile in business.

At YOW! in Melbourne this month we heard Jeff Patton talk about agile adoption, and he told us the story of how Ron Jeffries responds to people who whine about their failed agile project “oh yeah, we tried agile and it didn’t work for us”. He says that is akin to  saying “oh yeah, we tried baseball and it just didn’t work”.

You need to practice, chances are you just suck at baseball!

And it is practice that is the biggest secret of all – for guitarists, baseball teams and agilists.

Footnote: for the curious, here’s a live recording of the original (and beautiful) Fly My Pretties song tabbed so badly, and scribbled all over in my song book above.

At first I was cheering on all the tweeters, going ‘right on’… But, then I realised what Paul Graham and co are doing is exactly right.

Clearly they have a successful recipe for making startups work, Reddit, Disqus, Dropbox, Posterous, Airbnb and on it goes, those are just the ones _I_ like.  So what they are doing is innovating their own recipe, iterating and trying something new, why not, what else are they going to do ? Wait for all the copycats to get just as good at making pie as they are ?

The other point is of course that I agree with Y Combinator, teams, culture and ways of working are much more valuable than ideas.  I think we just don’t like saying it out loud sometimes because it sounds soft, so we have aquihires and we use words like ‘pivot’ to make it sound more legitimate.

Your ability to succeed in the face of an uncertain environment or problem is directly related to the quality of your team, how well they work together and how good they are at finding and responding to new intelligence and customer insights – the ideas will come, go and change along the journey.

Addendum: These three insights into Google, Goldman Sachs and Yahoo give a window into what happens to companies that ignore their people, ignore their customers and lose the ability to innovate.