Tuesday, September 28, 2004

The Growth Habit

What's with this growth imperative that all companies seem to follow?
Why is it mandatory to consume? You know: shop till you drop. Bloat till you burst.
'Cos that's what happens in the end: you kill off and gobble the competition, outgrow your available market, and then where are you?

Stuffed.

Semler muses on this topic in 'The Seven Day Weekend'. He was specifically concerned about the impact on Semco culture: growth -> more jobs -> more people -> cultural dilution. Since Semco culture appears to be unique, vibrant, and overall a very precious asset, you can see his concern.

So, I got to wondering: why can't a company just putter along with a steady income? Why does it have to grow? What are the forces that drive company growth? For a publicly listed company, the answer is likely the ongoing need to give the shareholders a regular income: they want an ongoing return on their static investment.
But what about proprietary companies? The answer I came up with has to do with Risk Management.

I think there is a 'sweet' spot for company size where it can optimise the service and support it can provide its customers, and the level of nimbleness it can display in providing that service in a changing environment:
  • Too small, and you risk being wiped out by random fluctuations in your market opportunities, inconvenient invoices, lapses in your available skillset.
  • Too big, and you start to consume more resources to sustain your corporate structure than you can provide to the market.
For the standard business model applied to companies, that sweet spot would seem to be unstable: in astrodynamical terms, it's at L1 or 2, rather than L4 or5.

(Semler actually identifies a few examples that have managed to achieve an equilibrium: one of them is a small Amsterdam restaurant that had been in the family, pretty much as is, for 200 years or so!)

Since a newly hatched company tends to be on the small side, it clearly needs to grow to avoid being 'swamped by the waves'. So, it tries to put a bit of fat on and get more product out. In order to increase production, it hires more staff to handle the production requirements, and identify the market.

Now we start to see at work the forces that define productivity*:
  • overhead: what needs to be spent in order to produce the product
  • inventory: the investment in the actual product on the shelf (but *not* sold)
  • throughput: the income derived from placing product into a customer's hands
  • productivity = throughput - inventory - overhead
To replay the last sentence using these terms:
'In order to increase production (throughput - inventory - overhead), it hires more staff (increases overhead**) to handle the production requirements (increase inventory), and identify the market (hopefully, to increase the throughput)'.
Put like that, it isn't immediately clear that overall productivity has been increased, since the increased throughput is offset by the increased overhead and inventory. Furthermore, bear in mind that throughput is a lag indicator: unlikely to turn up on the balance sheet until the end of the month, or next. No such luck with the bills, eh?!

A manager's life become one of juggling these values so that productivity stays positive, and a very bumpy ride it can be too.

Positive? Why not just zero (to return to the original argument)? Well, the problem is that productivity can fluctuate wildly from month to month. If one is prudent, one will aim to have a bit stashed away for a rainy day, no? Just in case the next months figures aren't as good. Or really terrible. This prudence becomes a habit.

And so it grows.
* See 'The Goal' by Goldratt for a discussion of measuring productivity. It's presented in novel form (!). While it won't win the Booker prize; it makes for a fair read.
** Folks, don't take being referred to as overhead too personally. It's a purely technical term. I'm sure your company loves its organic assets ;-)

Monday, September 27, 2004

Interlude

It was a nice day last Saturday.
We took Little Missy down to the local park and had some afternoon tea.

Lying back on the picnic rug, looking up at a flowering gum catching the evening sun and framed by a vivid blue sky, I reflected that, if you can't remember the last time you did something like this (lying back and looking up at the sky that is), then it was probably far too long ago.

Thursday, September 23, 2004

A Little Something of Mine Own...

Anyone who has been following this monologue may have noted a certain degree of frustration with my current position, and the glass walls which seem to prevent any able but inexperienced person (ie me) from bettering themselves. While I don't believe that learning Java, Perl, PHP etc to be all that difficult (certainly not the stock three years difficult) I do realise that some ability is expected. Maybe if I could demonstrate some real work, a little project, such as...

Hello? Are you listening?

...Is there anybody out there???
(no answer from the great, empty back paddock.)

I have a knack for being invisible.

It can be a useful knack. It would be an even more useful knack if I could control it! It is a knack which seems to have extended to this blog page. Judging from the search results: not even the Google bots have sniffed me out yet!

While it wouldn't surprise me if there isn't anyone in the great back paddock. The thing is, unless someone leaves a comment, I wouldn't know ...unless I add a web counter.
I could drop one of the ones listed in the Blogger addins , or...

Aha! I think I've found a project!

I'll let you know how I go


Tuesday, September 21, 2004

She's Ba-a-ack!

Having dismissed Pauline Hanson in a recent posting, what do we find?
Lo and behold! She's back standing as an independent senator! What's more, she even seems to pull enough credo to get an interview with Andrew Denton. This makes me prick up my ears. Denton started off as a comedian with a razor sharp wit, and has, in the recent years, proved to be a very astute interviewer, capable of selecting some unusual but worthwhile subjects, drawing them out and letting them speak their minds in their own words. His show is called 'Enough Rope'. Need I say more?

But back to the Hanson interview, which was aired last night, and which I happened to watch. Here we have somebody who, in her maiden speech in parliament, stated that we were in danger of being overrun by hordes of Asians, and that aborigines were in no need of social welfare.
Eight years on, we see someone who is perhaps a little sadder and wiser. Someone who is perhaps a little more articulate; if still very nervous. Someone who claims that she's not a racist, but a critic of political correctness: that the point she was trying to make was that she doesn't see aborigines as any different from anyone else, so why do they need preferential treatment?

Oka-a-ay.

That said, she still doesn't have my vote (she wouldn't anyway, since senators are elected by state). In the interview, she still seemed quite uncertain of herself, and Denton, having covered the issues raised by her speech, moved on to the naivety displayed when the One Nation party she co-founded imploded through sheer incompetence, and then raised questions about whether she was up to being an independent senator who could, conceivably, hold the balance of power (scary!). Denton got a bit relentless at this point, trotting out figures on the number of pieces of legislation reviewed by the Senate annually, and the amount of paperwork involved. He then queried whether someone who is a competitor in a ballroom dance competition is showing enough dedication to the job. I, personally, thought this a bit low, since someone with that sort of workload ought to have a bit of relief. Nonetheless, I had nightmare visions of 'Springtime for Hitler'

It was an interesting inteview, and I came away reflecting that, with the world becoming an increasingly more polarised place, the ability to keep an open mind is a precious thing. I'm willing to accept that leopards may change their spots, or that the light might have been a little dim to start off with.

Mind you. Once, long, long ago, I had an open mind about John Howard too.

He has since shown himself, in not so many words, to have precisely the attitude that Hanson claims she does not. What's more, he is a far more intelligent and astute politician than she will ever be.
That's the real worry.

This is someone who will demonise a group of desparate and destitute people in a leaky boat in the middle of the ocean as fifth column terrorists who think nothing of throwing their own children into the water to force the Navy to rescue them (and thereby let them in to work their wicked ways). This is someone who will deliberately incarcerate said people in complete isolation indefinitely. This is someone who seems incapable of showing any sense of decency to anyone non-white who originates from outside our little exclusion zone.

So, sorry John! Your moral values and mine are clearly not on the same wavelength. I don't care how competent a manager of the economy you and your team might be, come election time this boyo's going to be one of 'those who walk away from Omelas'!

Thursday, September 16, 2004

Open Your Windows!

There's something in the air. Something new and fresh, which I'm only just beginning to notice.

No, I'm not referring to the fragrances of early spring in Melbourne (and no, I haven't forgotten those wattle photos; be patient) .
Nor am I necessarily referring to Linux, although that is closer to the mark (and should give a hint as to where the title of this piece is coming from!).
Open source is certainly a part of it. A very big part.

But there's something else. Step back, and see:
  • Open source (obviously) : groups of people coming together spontaneously to produce code to share
  • Internet: the underlying architecture of which was designed to survive nuclear warfare by avoiding central, easily targeted, switches
  • Blogging: online diaries accessible to all, and visible to all (if that is desired)
  • Wikis: webpages to which anyone can contribute and correct
  • Wikipedia: an online encyclopedia to which anyone can contribute: and correct
  • Companies (a few): abandoning the old 'management knows best' philosophies, in favour of a workplace democracy, where all staff get to participate in defining the work (check out Ricardo Semler)
  • Surveillance: check out David Brin's thesis in 'The Transparent Society' that the best answer to the old question: 'who watches the watchers?' is: the watchers themselves.
  • Disaster Response: New York: Sep 11. Official Emergency services were overwhelmed by the magnitude of the attacks. However, volunteers appeared out of nowhere to assist. A similar thing happened in Bali a year later.
Two common themes are emerging here:
  • 'decentralisation': It's a very human reaction when faced with a crisis to tighten controls and yet it's the worst thing for a manager to do. Why? Because it prevents other people from contributing: and actually helping out!
  • 'self organisation': given sufficient information, and effective means of communication, people don't usually need to be told what to do and where to do it. They will usually manage to organise themselves into effective teams. Yes, there will be leaders, but they will change with circumstances, and different skills are needed to be applied.
Decentralisation? Self-Organisation? Ugh! Both these terms are rather clumsy so, for the moment, let's lump them under the one label:
'Openness'
It isn't really that new but, in the past, it's foundered on the tyrrany of distance that diminishes and distorts the free interplay of ideas that nurture it. Not to mention the vested interests that thrive in closed, tightly controlled environments. Control is a habit they are very loath to give up.

Of course, like any new idea in its early stages, it's very easy to get enthusiastic, abandon all caution, and extrapolate wildly (and Openness is, by its very uncontrolled nature, particularly susceptible to this). How widely applicable is Openness? Could a company really be run by its philosophy (Semco seems to have managed for 20+ years) ? Could a government?

Could a world?

But there! I'm allowing my old insular habits to assert themselves. For now, let us sip this heady brew; see where the winds of change take us.

And, I'll let you be the judge.

Tuesday, September 14, 2004

Monocultures

Consider this:
A field of wheat. Six billion identical heads, rippling in uniform waves in a gentle breeze. Provided with all that is considered necessary for their wellbeing. All primed for maximum productivity.
Total population: six billion
Now consider this:
A field of wheat. The nodding heads are a little fewer, and, perhaps, smaller.
Some signs of nibbling here and there as fieldmice partake of the bounty, taking the leftovers back to their young in nests suspended between the nodding stems
Weeds grow among the stems. Cornflowers provide startling blue highlights among the muted gold.
There are bees.
Total population: six billion (not all wheat)

Which would you rather?

I'm bored!

It's one of the reasons why I started this blog.

In between testing, having nothing to do at work for a week can be a liberating experience. After months with nose to the grindstone, one can relax (one should take time out anyway, of course), look around, see what other people are doing.

After two weeks, one may begin to chafe: should one find something to do? Is it worth starting something before one gets dragged back into the production ratrace?

After a month of relative inactivity, anxiety sets in. After all, you're getting paid for sitting on your expanding rump (comfort food cravings) surfing the net and generally doing nothing! Surely, somebody, somewhere, is going to notice? All you can do to counter that is to let management know you're not very busy and listen as, with typical 'finger on the pulse' insight, they act surprised, mutter something about seeing what needs to be done, and then forget about you. Hazards/advantages of working in a large-ish organisation.

Does anyone else need a hand? I ask. The trouble with this place is that everyone is so budget conscious that they won't even take on an extra body for nothing! (sheesh!)

Three months. OK this is getting serious. Shall I start looking around elsewhere? Not a bad idea, although, Micawber-like, we are in imminient expectation that something will turn up. Meantime, one can do some serious self-improvement. Not those rubbishy courses on personal interactions that are whacked together to fulfil the training requirements, we'll get some serious programming skills under our belt. I start investigating Agile methodologies and, as luck would have it, some in-house Python enthusiasts get me interested in that as well. This leads to a brief, but useful couple of months work on another project.

Then, phase 1 ends, as does the budget, and I'm back to... testing! Only, it's even less enthralling than before and others realise it also because, after another month or two, I'm let off that as well.

Back to management. Same uneasy shuffle. Some half baked internal projects follow.

Arthur C Clarke once observed that a truly educated mind is never bored: the presumption being that it can always find something with which to occupy itself. I've managed it for a year or so, but I think I've got ot the end of my education.

Definitely time for a change!


Friday, September 10, 2004

Of Time (and Terabytes)

'Life caching', it's called. Attach a wireless webcam and microphone to your bonce and let a continual stream of all your comings and going get squirreled away for viewing some other lifetime.

Actually, the idea has its uses:
  • '...unhand me, you randy brute! You're on candid camera!'
  • '...if you want a witness, my website saw it all!'

There is a catch of course: all this information has to be stored somewhere. Part of the recent interest in the idea is that mass storage is getting pretty... well, capacious! It wasn't that long ago that terabyte storage was the province of major computing installations. Now it can be obtained off the shelf for only a few hundred dollars.

Is it enough?

Well, I started doing a bit of calculating:
Suppose we wanted to store all sight and sound experiences for a lifetime. If we take our alloted span to be a generous 100 years , that's about 3e9 seconds. Smooth streaming video requires about 30 frames/sec, so that works out to be around 1e11 frames.

Now then, how much data in a frame? We want full surround vision (ie 4*pi steradians, please!) in 24 bit colour.
The average pixel on a 40 cm (1280x1120) screen is about 400/1280 ~ 0.3 mm across.
Given we have comfortable viewing at about 50 cm, that requires about 4*pi*(500)^2/(0.3)^2 ~ 35 million pixels, or 1e8 bytes to paste our entire viewing sphere to the same resolution in 24 bit colour.

So, that works out to be 1e19 bytes, or 10 million terabytes. I'm not going to include sound, as it needs a good deal less storage than vision, and could easily be packed away in some corner of the figure quoted here.

OK, so I've been rather generous with my requirements, and I've given no thought to data compression. The basic point is that full on life caching is a ways off yet.

Still, you could get started now, and ramp up your storage as it becomes cheaper.

Tuesday, September 07, 2004

Resistance is Futile! (so why compete?)

It's axiomatic: when criticising a point of view, go for the assumptions. The trick is to identify the assumptions that are being made, and these can be so universally accepted that most people have a blind spot about them.

Now, it's popular to take and apply natural behaviour to a business setting. The most fundamental thing about businesses that you will hear is that they compete. They jostle each other at the free market trough and strive, by any means, to maximise their share. This is good, you hear, because the evolutionary forces select for efficiency.

The assumption here being that it's the only way to select for efficiency

I have no doubt that competition does, indeed, select for leanness (meanness, too, but I'm not going there today...I think)

And yet...

Maybe I'm missing something, but it seems to me that all this fighting and competition can be an awfully inefficient way of ensuring efficiency!

Let me give you an example. Some years ago, when the Australian Telecommunications industry was being opened up. Optus was being set up to compete with the incumbent government body: now known as Telstra. Vast sums of money were lavished on advertising campaigns to woo customers. All for one fundamental choice: which carrier to use when making international calls. To this day, Optus has 'Yes' embedded in just about every advert they make!

OK! OK! I know! It was a stunning example of how free markets don't operate: Telstra was set up like a papingo and, as I recall, Optus still weren't able to make much of a dent.

What struck me at the time was the amount of energy companies are willing to spend just neutralising each other. (Hmm! A more obvious example: US and USSR up to a decade ago)

What a waste. Wouldn't it be better if they could somehow agree to disagree and get on with doing something productive?

Ah! but what about efficiency? And anyway, it's the way of the world, isn't it: winner takes all?

Yes... and no. Consider your typical alpha males that beat the crap out of each other for mounting rights. In fact, they don't. Full-on fights are fairly rare in nature: all of the effort really goes into display and bluff. This 'anything you can do...' approach allows the maxim 'live and let live' to flourish (... but just think twice if you want to start something, buster!). What happens is an agreement, not to compete to the fullest extent, and results in two fairly intact beasties in their prime rather than one dead and one as good as.

The trick being applied here is an efficient means of demonstrating fitness: a stag's horns are a relatively small investment compared to the protein behind it. (Nonetheless, it can still get out of control: think Irish Elk!)

I've already made the point I wanted to make:

competition is an inefficient way of being efficient

What other ways are there?

Well... open source projects seem to demonstrate a model of cooperative strategy. The point being that, without a concept of ownership, anyone is free to contribute to an existing endeavour. So, if you want to build a better mousetrap, you can start by helping to make the existing mousetraps better.

Monday, September 06, 2004

Oh, Beslan...

Three years on from September 11, 2001, and it is now Russia's turn to confront the grinning face of homicidal insanity. I hope they learn from the experience of others, and decline to retaliate in kind.

Not that they can be blamed at this time. I imagine it's almost impossible for the bereaved not to feel vengeful themselves (I can well remember my angry reactions when it was only my house that was robbed: how it must feel when it is your family? To lose children like this...?!)

I can also remember my reactions to 9/11: one of the earlier queries in the confusion was: who has done this? No one was claiming responsibility, and what that implied to me was this:

'There is no need for a face, for we have nothing to say to you: we simply mean you HARM!'

This sentiment rises from the doctrine of despair:
'We have nothing to lose. You have everything. Go figure!'.

Faced with an attitude like this, no amount of arms will prevail. What you have to do is go after the root causes: remove the environment that nourishes this sort of desparation.

It's the only way and no, it's not easy. It will be a long and at times painful process to learn why people, many people it seems, are willing to fly fully laden passenger jets into office buildings, or blow up nightclubs, or mow down school children.

It will be long, because the root causes run very deep, and are very old, and will not be soon forgotten. (We're talking generations, I'm afraid)

It will be painful because the West must be willing to accept responsibility as and when necessary.

And, I'm afraid, we're not doing it. Not even starting to. Three years on, the US seems to have reacted in as wrong a way as it possibly could. Bush has been following bin Laden's script to the letter. The script presented to every dictatorship by just about every geurilla freedom fighter since... whenever.

And that is this:
  • irritate, and annoy the rulers intolerably
  • provoke reaction, but do not present a clear focus for it
  • instead, let the peasantry feel the clumsy displeasure of their masters
  • let the peasantry rise up in outrage and displace their masters
  • ...and then step in to fill the void
Do you really think that every Afghani is a Taliban extremist? Every Iraqi a hater of 'liberation'?
Every Chechnyan a homicidal lunatic?

There is a better way of dealing with the new world order. It is one that does not require overwhelming force of arms (indeed, they may be a burden). It is the one practised by the British during the Malaysian emergency:
  • learn to gain people's trust, not their fear
Mr. Putin, please consider.