Monday, 2 August 2010

R I P Tony Peluso

Tony Pelso is the guy who played the guitar solos on The Carpenter's Goodbye to Love and along with Richard Carpenter invented the Big Ballad Guitar Solo. He wasn't the greatest guitarist in the world, but he didn't need to be. He just needed to do this...



In Heaven as I think of it, that's enough to get you in.

Friday, 30 July 2010

Why Experts Don't (Usually) Work in Service Companies.

A service company uses stuff that other people make and design to provide a service to other people. Telcos use optical fibre and highly specialised computers ("switches" if it's a SS7 / C7 network, routers if it's a VoIP network) built by someone else running software that implements very abstract and specific standards designed by someone else. You don't need to know the innards of C7 signalling or electromagnetic theory to run an telco operation, you do need to know how to make relationships and negotiate. Banks are service companies, so are bus companies, railways (unless they are vertically-integrated), travel companies and all retailers. So are hospitals and GP's - medics don't know how drugs work, but surgeons at least know their way round their bit of the human body.

Service quality can range from a five-star hotel with a concierge paid a very high salary for his knowledge of the town to a one-star bed and breakfast with a single girl on reception who seems to be there (as far as you can see) all the hours the place is open and who can just about direct you to the local railway station. The service sector tends to drift down the skill scale: if you need a palladium widget you have to pay for palladium and that puts a minimum price on the widget, but you can always live with a little less hotel service on this trip to save some money. A sustained bout of cost-cutting customers later and none of the hotels can afford to offer good service, so they don't. Quality decrease is the other side of the inflation coin: heads we put up the prices, tails we reduce the quality and quantity.

After a certain point on the way down the price/quality slide, the senior management of a service organisation gets concerned about the organisation's ability to handle any tasks requiring specific abilities, talents and knowledge. It make sense to outsource your freight forwarding to a specialist company because you probably don't do enough to keep the people you would need busy. The same applies to advertising, as very few companies have a culture that's friendly to the kind of erratic sparks who think up good ads. The problem starts when you outsource all your creative thinking about anything to outsiders, so that all your people are doing is managing agency relationships and the internal bureaucracy. Telcos and banks are nothing more than information-processing machines, yet many outsource their serious operational IT work (The Bank doesn't run your ATM's and current accounts, EDS does.). As a result, no-one in the business actual understands the systems any more and if they ask, it costs them a fortune in fees for meetings and Work Requests. So they don't ask.

A service organisation can wind up consisting of a large number of people who graft at heavily-supervised very specific front-line tasks, a senior management with a small supporting staff, and a middle management that is little more than groups of people managing relationships between the people who actually know something. In fact, that's a pretty good description of a lot of people in the product areas of The Bank. Relationship managers, project managers, team leaders, business managers, account managers - call them what you like, they don't need and often don't have much industry knowledge or technical skills. What matters is that they can handle people, politics and bureaucracy, and they don't rock the boat.

Which is not going to produce a culture that encourages skill and knowledge development much beyond the lower end of the advanced beginner on the Dreyfus scale. Why not? Well, what kind of person gets to be really good at something? Or to ask the same question a different way: do you really think that Roger Federer, Tiger Woods, Alexander Grothendieck or for that matter Brian Kernighan and Denis Ritchie are (or were at the time they did what made them famous) well-balanced "normal" people?

Uh-huh. Normal people do many wonderful things, but only after those wonderful things have been invented by their neurotically-driven inventors. Think of the famous story about the discovery of penicillin: just how desperate for a discovery do you think Fleming was that he thought there might be something in what a normal person would have seen as a dirty Perti dish and cleaned without a further thought? Exactly.

People get to be good at anything because they push their current limits of knowledge and accomplishment a lot of the times they practice and work. There's a whole literature on this: the 10,000 hour thing and the idea of Deliberate Practice - see Geoff Colvin's Talent is Overrated. Try it at the day job: you have to read the manuals, acquire the background knowledge, find time and projects to learn on - which means you have to dodge the time-consuming routine junk work and be prepared to miss those fake deadlines so you can learn and experiment with something new rather than hack out an answer with what you know. You're going to do this how while you have a day job, a commute, a wife and children or you want a social life and to spend the weekends wind-surfing? Why bother even ascending to competent when you can get paid just as much for being an advanced beginner? Especially when that's all the company expects?

If you want to be good at anything, and you are in a large service company, you will leave for a small firm, a consultancy or private practice so you can concentrate on what it is you want to learn and work on. Which is the final part of the explanation as to why service companies don't have experts working in them, and precious few proficient people either. Now you know why governments, banks and other organisations don't have someone who Knew Better when they made that crazy decision. There was, but they left. Before they went crazy.

Wednesday, 28 July 2010

The Dreyfus Skill Levels

A few weeks ago I picked up a book called Pragmatic Thinking and Learning: Refactor Your Wetware by Andy Hunt of Pragmatic Programmers fame. Hunt discusses a thing called the Dreyfus model of skill acquisition, which applies on a per-activity basis to people. So you can be an expert chef and a novice windsurfer. Or vice versa.

The top level is the expert: experts write the books, give the conference addresses, get called on by governments and television producers and as practitioners have "forgotten" what they know so that it has become second nature to them. It all looks intuitive and mysterious, because you can't see the thought process.

Which you can with the novice, who needs strict instructions and has no experience or background knowledge with which to exercise judgement. Think offshore IT help desks. Next up from the novice is the advanced beginner. These people know enough not to telling what to do every time, but not so much they can sort stuff out when it goes wrong. These are the "party trick people", who know the tricks needed to do their job but not much more and who copy code from the Internet. They frequently use tools and refer to ideas that they don't really understand but can "use properly". They still need showing how to solve something, rather than being able to get a hint and figure it out from there.

Figuring out an approach to solving a problem and troubleshooting are the hallmarks of the competent. Competent people take on new problems because they will learn more by solving them. They have a fair chunk of background knowledge and can use general principles in their problem-solving. There are a couple of books on their desk, strictly utilitarian reference textbooks.

When you start reading books like The Pragmatic Programmer or How To Solve It and want to know the abstract ideas behind a subject - from theories of how food tastes to the grammar of the language compiler - and you can comfortably interpret and use maxims and rules of thumb in context, then you are at the proficient stage. There's a lot more background knowledge, not just of the technical stuff but of the wider business and social context. The proficient person hears a conversation about how someone fixed a problem on the corporate website and gets an idea about how to solve a database problem - they are operating at a much higher level of abstraction.

The news is this: forget all that stuff about "subject matter experts" and "excellence": most people are at best advanced beginners at everything. You shouldn't be surprised: how much time do you have left in a week to get really good at something that isn't your day job? And how much time do you have in your day job to learn and practice new stuff? After all the routine crap work is done?

Monday, 26 July 2010

Rules for the In-House Training Course

At the start of all the courses I've been on recently, there's a little ceremony where the trainer(s) ask us "what we want the rules to be" for the duration of the course. There's a moment's silence and then someone says something fairly applie-pie-ish. The rules have turned out to be fairly consistent:

Confidentiality: what is said on the course, stays on the course. These are not prescriptive courses - for complicated reasons to do with institutional denial, The Bank can't do prescriptive courses - but rather ones where we're presented with some ideas and invited to discuss what various concepts mean to us and what we think of this or that fairly simple case study. (It's a lot like Teacher Training, and for the same reasons. I feel another article coming on.) The value of these courses is what you learn about yourself and the benefit of that is the out-loud admission of your faults and revelations. A lot of frustration, confusion and a little hurt gets expressed in these sessions. If we thought any of that would get back to the Bosses, the sessions would be useless.

No Mobiles: no comment needed.

Let everyone have their say - no talking over people: people like me need to be reminded not to do that. It's a really bad habit I have.

Honesty: within reason, of course. What this partly means is that people must not say "I've never done that" or "That doesn't happen here" or "Of course I do that all the time". Not so much dishonesty as what the psychologists call faking good. The other part of what it means is that everyone has to share a little: no sitting there saying nothing.

No deferring to the senior guy. Ever noticed how everyone in the room shuts up and does deferential listening body language when a Senior Person makes a contribution? Even if you didn't know they were a Senior Person, they minute they start talking, you know it: there's the quietly confident body language, the measured tone that doesn't expect to be interrupted, the measured, placatory, ambiguous language and the sense that they are delivering a message, not speaking from the heart or soul. The training people advertise the grades who are eligible for the course so this doesn't happen. But when it's an open-grade event, the deferring happens.

No apple-pie and no jargon. This is mine and I'm horrified at how few people know what "apple pie" means in this context, or for that matter, what "drinking the Kool Aid" means. How can you identify the action if you don't have the concept? Apple pie, if you don't know, comes from a 1950's phrase: Mom, the flag and apple pie, three things no-one is going to find fault with. It means anything that is uncontroversial, unarguable, received wisdom and hence bland. Drinking the Kool Aid is what you do when you decide to go along with the prevailing or required beliefs - with the suggestion that it's a little bit cultish, uncritical and yet that one's acceptance is slightly ironic. I'm asking that the participants don't hide behind platitudes - as one woman did when asked for her action at the end of the Resilience course: "well, it's about making time for me, really" she said. Apple pie.

I liked the guy who said that he wasn't keen on profanity. Everyone at The Bank is far to well-behaved to be profane of speech, but it did perk the debate up a notch. He was a nice guy as well.

Friday, 23 July 2010

Whatever Happened To....

Procol Harem's Whiter Shade of Pale would not sound the same without it. Booker T played one with the MG's on the reference-funky Time Is Tight




Larry Young played one with Tony Williams's Lifetime and Georgie Fame played one with The Blue Flames. Stevie Winwood played one with Traffic, Blind Faith and in his solo career: that's him playing the blues solo on Voodoo Chile with Jimi Hendrix. Keith Emerson played one and so did Rick Wakeman with Yes. John Lord of Deep Purple played one, as did Keith Jarrett when he was in Miles' Live Evil band. Brian Auger backed Julie Driscoll with one to make this psychedelic version of a Byrds song...


... and then it vanished. RIP.... the Hammond electric organ. 


Wednesday, 21 July 2010

A Six-Point Plan For Removing Depression - If You Don't Commute

An article in the Guardian the other day reported a doctor who treated depression without using pills. Since my religion forbids mind-altering substances, he has my support. His six-point prescription is:

1. Take 1,500mg of omega-3 daily (in the form of fish oil capsules), with a multivitamin and 500mg vitamin C.
2. Don't dwell on negative thoughts – instead of ruminating start an activity; even conversation counts.
3. Exercise for 90 minutes a week.
4. Get 15-30 minutes of sunlight each morning in the summer. In the winter, consider using a lightbox.
5. Be sociable.
6. Get eight hours of sleep

Let's take the good doctor seriously and see where this gets us. Let's assume you live in the suburbs, have a 9-5:30 (thirty-seven-and-a-half hours a week) job and an hour's commute. How to get that 30 minutes of sunshine in the morning? We're looking at walking some of the journey to work (lunchtime isn't the morning). If you use the car, you're going to leave it half-an-hour from work, so you'll have half-an-hour's walk back in the evening. By public transport, you can keep the original time for the return trip. So with that eight hour's sleep, your diary now looks like this:

Return home: 22:00
In bed: 22:15
Asleep: 22:30
Wake up, ablutions and breakfast: 06:30
Commute (public transport or car) starts: 07:30
Walk: 08:30
Work starts: 09:00
Work ends: 17:30
Commute ends: 18:30  (car: 19:00)
Ready to party: 19:30 (car: 20:00)
Return home: 22:00

You can take the vitamins at breakfast with no loss of time and all that walking takes care of the ninety minutes of exercise requirement.

Let's look at the logistics of socialising. Forget the theatre or movies on the way back from work: unless you get a movie that starts before 18:30, there is no way you're getting back before 22:30 by public transport. Anyway, it would take you about half-an-hour to unwind from what would have been a very long day which ended by travelling on a train or bus full of drunks. If you go locally, allow twenty minutes to get there, so your evening meal is rushed to make the theatre at 19:15 (takes time to get to the seat). Ditto the movies and evening classes. Supper with friends can start about 19:30 and you might be back at 22:00 if you don't stop at theirs for a chat. There's the pub (unwise: alcohol is a depressant and lowers the quality of your sleep) or round at theirs or yours chatting (over a few beers, right? Not so much) or you could be ten-pin bowling. Whatever you do, this is pretty focussed and maybe hectic "socialising". In my experience, socialising with a time limit is only partially satisfying, closer to networking than a moment of friendship. Maybe you could do your chores during the week and socialise at the weekend? Except everyone else is doing their chores at the weekend - because they don't share your new lifestyle.

Or you could live like this:

Return home: 23:00
In bed: 23:15
Asleep: 23:30
Wake up, ablutions and breakfast: 07:30
Commute starts: 08:30
Work starts: 09:00
Work ends: 17:30
Meet friends in bar / cafe: 17:45
Leave bar / cafe: 18:45
Return home: 19:00
Ready to party: 20:00 (for e.g. 20:30 movie twenty minutes away)
Return home (again): 23:00

You can do this if you are living and working in the centre of a decent-sized town somewhere near the Mediterranean that doesn't close at 18:00. It's inland commuters who have a hard time fighting depression.

I'm not disagreeing with the good Doctor. The bit about not dwelling on negative thoughts is a tad glib, but it's what depressives do. I suspect it's a symptom, not a cause. The cause is a deadly job with a nasty supervisor that pays just enough to meet the bills, a marriage that went dead two years ago and having no-one you can talk to about any of it. The real cure for depression is to get out of the life that's killing you. Which is tougher to do than it sounds. If one part of happiness is the feeling of having choices, one part of depression is the feeling of being stuck with this shit. I suspect the good Doctor's six part plan works not because of what it does but because of a Hawthorn Effect - you're paying attention to yourself and your life. Pass the fish-oil capsules. 

Monday, 19 July 2010

A Vision For Analysts

We've been having meetings about "what sort of MI team we want to be". I try to miss these as all anyone does is serve up apple pie. After the Leading A Team course I went on, I wrote this in an hour on Friday afternoon...

"Our first job is to help the business. We do that by responding to their requests and by spending time talking with them about what they are doing so that we can suggest MI and analyses they may not have thought of.

Our second job is to produce an overview of the business. Most of our colleagues are focussed on details and don’t have the time or resources to do this.

We take time to discuss with the customer what they want and need, why, when, for what audience and for what purpose. And we keep the customer informed on progress and discuss any issues arising during the project.

We keep up with what other MI groups are doing so we can benefit from their work and not duplicate it.

When we have ideas for improving reports or find things that strike us as interesting during the course work, we will not be disappointed if does not fit the current business agenda.

We are in MI because we like to understand what a business and market looks like and we like solving intellectual problems. We should be looking at things we think might be interesting even if no-one suggested it and we should be extending our knowledge of the tools we use even if we don’t have an immediate use for it.

We are surrounded by people whose job it is to influence, persuade and manipulate other people into working on their projects. We need to develop the skills and temperament to handle those people in a polite and mutually beneficial manner.

High-grade MI is produced by smart analysts using the right tools and adequate data. The team will hire people with potential and train them so that in two-three years they will leave for salaries way above what they could earn here. In the meantime The Bank will have had one or two years of a quality of work way above what it is paying for."

... and sent it to the boss. You may safely assume that if I'm saying we should do something, we're not doing it and indeed some people think we shouldn't be. This might look obvious, but none of it is showing in what we do. And I'm going to bet it isn't in your team either.