Old Boar's blog (Old Boars blog)

"One more day to go, one more day of sorrow..."
I've been a bit busy today. Completed the descriptions of a load of "transformation rules". I'm a bit dubious about the value of that, but some of it _was_ worthwile.

I've almost finished an overview education presentation. Should make it by tomorrow, just! Now, I've off to someone's leaving do (not mine).

This is going to be a short post, because I'm busy.
I _was_ having some fun with Python - then I tried to integrate a little bit of Tkinter and after an hour I just couldn't get it working at all. I gave up in disgust. This morning I had another go, and now I've got it working. The explanation, a mixture of:

• Trying to run before I could walk (or maybe crawl)


• Working through the problem in my sub-conscious?

You'd think I would have learned better by now! Maybe it will be back to having fun now.

Yesterday evening had planned to join "the krauts" watching their team play the Czech Republic. The plan was a simple one - watch the football, eat crisps and nuts,
drink beer (not necessarilly in priority order). The weather was so awful that I cryed off.

Instead I sat with fiddled about with Python and watched the match from the corner of my eye. Well, the company was rotten (there wasn't any), the Germans lost and I have a silly problem with Python. I'm not sure if it is looking in the right place for files. Single module stuff works but imports don't seem to be found. I'm sure I've set the paths but maybe I've got it wrong.

I should check this evening but I'll probably watch England v Portugal instead!

Sometimes it's fun to watch a fight. It appeals to some of the less attractive aspects of human nature, particularly if you're not involved yourself.

Here's one I can see developing near where I am and I'm going to try and not get involved with. Everyone I know has experienced the arguments that develop between departments - "You should pay!", "No! You should!" etc. Well that is what I can see developing and in the end "the little guy" is going to get the blame, I think. What has happened is this -

• Where I work many of the "permies" are paid an accommodation allowance. You know the kind of thing: covers the cost of a flat or lodgings, but less than the cost of living in a hotel. Some choose to live in luxury, some choose penury but pocket the difference. If it works right, it is a win-win for the employer and the employee, and the tax-man makes sure it isn't abused.


• It turns out that some of the charges have been going to the company personnel budget, rather than the project living expenses budget.


• The reason: there are two versions of the claim form: one with a box for the "charge to" code and one without. All these forms are checked (?) and countersigned by the world and his wife, including representatives of the project and the budget owners, and nobody noticed that some of the forms were different.


• Of course the official line is that the people raising the requests are responsible for its content (even if there isn't a question) - "You should have known that was required".


• Naturally there is no way for the individual involved to see where their money is coming from. They can see Whether they are getting paid, but not where from which account.

If I took this at all seriously, I'd suggest that someone should look at this "business process". But I don't, so I'm not going to get involved. You've got to laugh, haven't you?

I spent the weekend at the North-West Office, sort-of on call. Nothing happened.
So long as you can mentally switch off (or at least, not be disturbed by) the part of your brain which is "waiting for the phone to ring", while leaving the device itself switched on, hanging around can be pretty pleasant. (should cue "The Stranglers" at this point)

I made time to catch up on some reading on the programming language "Python".
I've got to be prejudiced in its favour, simply by being a one-time "Monty Python" fan,
but that aside, I'm quite impressed. I particularly like the way Python allows introspection and the way it can cooperate with Java. Maybe I'm being naive but the Java/Jython mix looks like a marvelous way of combining the best features of compilation and interpretation.

It's ages since I've done any programming, so I've downloaded Jython and, armed with that, the Java SDK, a book and a couple of tutorials we'll see how far I get before I get distracted or manage to confuse myself.

This article (IT Doesn't matter) has caused a bit of a hoo-hah in the industry. I had missed it until a colleague pointed it out, and I found it very interesting. It provides lots of "food for thought". I think Nicholas Carr is deliberately being provocative. I don't agree with the conclusion of the article - "IT doesn't matter", on the contrary I think the article makes a excellent case for "IT is essential", but the challenging argument should encourage businesses to review how they are investing in IT.

Carr's main point is that IT is becoming a commodity (or a utility), and that the important thing about a commodity or utility is the negative effect _not_ having it can have on the business. Some people fall into the trap of assuming that things they have are universally available. Even in the Europe access to motorways, mains water, sewerage, electricity, telephone and broadband is _not_ universally available. Assuming it is are has contributed to some very bad business decisions. This is one reason behind the (local) political campaigning for broadband, and why governments invest in or subsidise "infrastructure".

Commoditisation is why IBM and Microsoft are positioning themselves as suppliers of infrastructure and utilities and Apple are moving towards consumer goods (iPod) and consumer services (iTunes). Carr points out that systems need to be enhanced or replaced. His example of AHS shows that a system which provided business advantage can eventually become "a millstone" if it is not replaced in time. The reasons for this are not purely technical:

1. Competitors, customers and partners are changing how they work, so the business needs to move forward,.


2. Administrative systems (of which IT is just an implementation) seem to suffer from a form of "entropy" which the originator of "Parkinson's Law" (C Northcote Parkinson - "The task expands to fit the time (resources available)") called "injellititus". This means that if not overhauled regularly, the bureaucracy serves itself and not the business.


3. These "regular overhauls" make me optimistic that there will be work for Business Analysts like me (us?) at least long enough for me to draw a pension. Even electrical wiring and plumbing are replaced periodically.

Finally, Carr overlooks some of the opportunities technology is creating:

• Cellular phones and wireless networks are creating opportunities (and problems) for interaction between businesses,


• RFID (Radio Frequency ID) allows individual items to be tracked and the records to be integrated with business processes in a way that promises much.


• Goodness knows what is coming in the way of embedded systems.

In conclusion, I think Carr has a valid point about IT becoming "infrastructure" but that IT matters all the more because it is taken for granted. Maybe we will have to change the name "IT", or invent something new. In any case there is still sufficient techological innovation to create opportunities for businesses to gain strategic advantages.

I spent the weekend wrestling with tax and related forms. I'm not sure if it was entirely necessary, but it seemed to be a good idea to get it all out of the way.

The reason for the urgency (necessary or not) was that my wife is having problems with the Tax authorities in Eire. They say they want some evidence in the form of a document from the (British) Inland Revenue. I have a feeling they are not really being reasonable, as they are asking for something which will be produced in due course, but we would not normally have now. Since there is a chain of dependencies here, the easiest way out of it seems to be to get everything done as soon as possible, rather than in the normal matter of course.

Those who don't like dealing with bureaucracy should try dealing with two bureaucracies (plus a wife who's getting worried by it all)! Still, 'tis on it's way to the post box right now!

I did a foolish thing last night. I read my book "Foucault's Pendulum" while lying on my bed.
Guess what? I dozed off and the next thing was I woke up and it was dark, very dark. Needless to say, I did not sleep well for the rest of the night, and woke up this morning with a stiff neck as well. Serves me right!

Got to work today and sorted out the little back-log of things to finish from yesterday.
I needed to check if any of the registered minor issues had been addressed incidentally
by the other activity I was aware of. Well, it was a nice idea, and definitely worth the effort of looking, but no luck. None of the open "issues" (not that I would count any of them as being that big) have been touched by what I have been doing. The best news there is I think I've spotted one which can be closed for another reason. Even I get lucky sometimes.

While in the kitchen getting a mug of coffee, I started talking with one of my colleagues (JS, from Australia) about "Xtreme" programming. In case it isn't obvious, my current project is anything but "XP". At times I have joked about it being the "glacier" subset of "Waterfall". It turns out that JS also has experience of the approach and, like me, he sees a whole lot of good in what they do. The question we were discussing was: Is XP applicable to all situations, or if not, when and where should it be applied. I wouldn't claim to be an expert on XP. My "programming" experience is rather stale,
and my inclination is toward solving a "Business" problem (with or without) programming.
As a result neither JS nor I expressed any really strong opinions, and we certainly didn't reach a conclusion. On the one hand, I have seen some good things from XP (and from DSDM), but on the other I will admit to distrusting the "hit the ground coding" attitude. One thing that did find interesting was that JS thought that a vital prerequisite for XP to work was that the "Customer" bought in to it.
I think that is quite a perceptive observation.

Now, back to the grindstone!

And something has just reminded me that I still have a copy of the original "Frankenstein" to read. I downloaded it from Project Gutenberg a little while ago. I have to admit, I found reading the text a little hard to read, so I converted it to a pdf (by way of LaTeX) but then I got diverted onto other things. I think I will have to finish off reading "Foucault's Pendulum" first (even though I started F first). I've just conjured up the image of Mary Shelley and Umberto Eco fighting over whose book I should read first. I'm obviously either hallucinating or getting delusions of grandeur (if they aren't the same thing)!

So far today has been spent working through a batch of specification changes. They've all been approved by the customer already, and some of them are actually retrospective (in as far as the code they define has already been changed). It's not too exciting, but it is better than sitting around doing nothing. While I'm doing it, I can't help thinking that the Xtreme Programming and other RAD types have go a point. The question is, how do you maintain the necessary degree of control, while allowing things to progress. I definitely think that the administration associated with some of the changes I'm handling is inhibiting change in an undesirable and costly way. Still, I'm one of the costs and summer (holiday) is a coming.

On a different tack, I've been wondering what to do this evening, and whether I should plan an overhaul for my website. The current site was always intended to be temporary and has been static for far too long (well over a year).

So far today; I've sent two bugs on their way (unfortunately they were trivial ones, more to do with interpretation of specifications that anything else), done the next draft of that d*()%d report, sketched an illustration and written a few pages of notes for "the book" and sent a few eMails. In many ways all of that is pretty satisfactory, but I wish that testing were uncovering more bugs. I can't help thinking the reason we are not seeing more is that we don't have sufficient "coverage" yet.

I'm trying to work up some energy. If I was bored it would be one thing, but actually I'm not, it's just that I'd like to focus on things it isn't really appropriate to be doing here.

In the meanwhile, I can look forward to continuing with "Foucault's Pendulum". I have to admit it is a bit dense but it is definitely intriguing

Another matter altogether... Quite by chance I stumbled on a website which contains some stuff on Aspin engines. Now that really is a blast from the past.

Over the weekend I skimmed through the book
"Whatever Happened to the British Motor Cycle Industry?" by Bert Hopwood which has been reissued by Haynes.

For those who don't know, this book is about how the British Motorcycle industry managed to go from being a world leader to scarcely existing in under 20 years.

I can't claim to have read it thoroughly, but I will. It's almost Bert Hopwood's autobiography and he was there when the events described took place. You have to be interested in motorcycles (now pretty old motorcycles) to get anything from this, if you're not, then you're going to find it dull. However, if you get beyond that then it is an interesting story - almost a tragedy. Parts of it are like a horror story where you want to shout to the characters "Don't do that! Don't open that door! Don't waste money on that hopeless model! Don't cancel one of the few projects which might get you out of this mess!". But, and this is where I want to read it more carefully, the question it doesn't really answer is: "When did it _start_ to go wrong?"
Clearly when he wrote it Bert felt bitter about some of the things which happened,
and some of the decisions were just stupid, but I kept on looking for lessons I could apply elsewhere.
So far all I've come up with is a list of negatives:

• Don't lose sight of your customer


• Don't show contempt for your customer


• Don't ship poor quality product


• Don't allow the board to lose touch with marketing, manufacturing or development
  (and certainly not all three at once)


• Don't start gambling on "silver bullet" (or "killer app") products

And two observations:

• The whole thing from fledgling industry, through world leader,
  to lame duck to stone dead took place in a man's working life-time.
  In fact, many of the same names appear at different times.


• At that time (70's) a lot of British industry seemed to have problems.
  It wasn't just the motorbikes.

But "When did it _start_ to go wrong?"

I spent part of yesterday evening working on the picture outlines for the new book. It is making reasonable progress, but I expect to encounter some form of "block" (you know, "writer's" or perhaps "draughtsman's") some time soon. Producing the drafts for the illustrations has been good in several ways:

• It has confirmed the general outline that I had laid out (that is good)


• It has (as planned) given me something to hang bits of text (bullet points) onto, and


• It has provided a first bottom-up estimate of the size of the whole production.

However, it has also identified a few weeknesses:

• I'm not sure the parts support the overall arguement well enough. I'll have to work on that


• Trying to crystalise the use of cross-references between "Business Rules" and "Processes" has shown me that I need to think hard about how this will be done and how the cross-reference will be applied. I know how I do it intuitively (at least I think I do), but that isn't good enough. This is another point which I will have to work on, and the one which may produce the "writer's block".

Of course, looking on the bright side, finding structural problems at this stage is certainly a good thing - especially if I fix them!

Finally, another subject altogether. I've got the reviewer comments back on that draft report. I guess that with that and the book I'm going to have busy weekend writing.

I got round to listening to the lecture by Rusty Schweikart last night. It was a pretty interesting experience on several levels.

First of all, "The asteroid threat for the next 100,000 years" is an interesting subject. If you're not at least a little interested, then you haven't seen any disaster movies. Rusty is a good speaker. Listening to an audio recording of a pretty visual lecture is strange. From time to time Rusty referred to simulations of asteroid orbits and it was clear that he was pointing at something on a screen (which I couldn't see). Sometimes it reminded me of listening to radio coverage of a football match! "this asteroid makes a break on the wing and we can predict that it will score/collide with the earth here. But if we accelerate it, just a little, here, then it will miss. (Not a) GOAAALLLL!!!" (commentator pauses for breath). The predictions are less dramatic than I expected and Rusty's proposed solution sounds like it might be achievable. Certainly a gambling man (who was going to live long enough) would bet on it!

On a different tack, my vacation plans are turning into "planes and boats and trains", but we are going to get the relevant family members into the right place at the right time (unlike the asteroids).

This Monday (yesterday) was a Bank Holiday. I was feeling a little down until I reviewed what I had done over the weekend. Actually, it was pretty good! I spent the weekend in London and I achieved a lot. The last stages of the office refurbishment are underway. The last order has been placed, and so far everything is within budget. I have even managed to slip in some minor improvements for no additional cost! The worst thing is that two of the items are on two-to-three week delivery. Well, that really isn't that bad. It means that I will have time to arrange for cover for the installation.

Apart from the office stuff, I managed to do a little in the garden. Again, I achieved rather more than I thought. Despite my good intentions about going "organic" I think I am still going to have to take recourse to chemicals to keep the more persistent weeds under control. The brambles and the bindweed (convolvulus infiltrate into my garden from next door and I always let them grow too long (in terms of both time and length).

Anyway, I'm back in the North-West for another week. I wonder what excitements it holds?