On Strategy

World War I is, bare none, the most tragic war fought the last thousand years. Some say the war started in the 1870's era and ended on the eve before World War II. From a technology and strategy point of view, this is true. Nothing quite boosts human mental prowess more than a deep and enduring crisis; the war started with outdated ideas and technology and ended with all the concepts that were used to wage the second tragedy in the same century.

Take, for example, the approach on the battlefield -- long lines of young men, trodding slowly towards their doom. And even those who arrived at the other end of No Man's Land could not hold on. There were no supplies, no backup, no mates to back up their positions. Seldom was terrain conquered for long. Reconquests of trenches bought dearly with the blood of many were the norm.

The British did not innovate their fundamental tactics. Their society at that time was highly stratified with little room for initiative in the lower strata. On the other hand, and contrary to popular belief, Germany was almost the inverse; it had a strong middle class. It invented a concept that nearly blew the Allies away -- Stormtroopers. Soldiers with decentralized decision making authority, often by non-commanding officers (NCOs in jargon, ie those below officer level). The tactics were aimed at deep penetrating units, circumventing tough defenses. New weapons were added to the arsenal; flamethrowers and grenades. The units had a high degree of self-coordination and self-reliance, without overbearing staff interference up to the level that they did not, and indeed, could not, know about. The unit was very successful and mostly due to incompetence of the over-hyped general Ludendorff (suffering from a nerve breakdown after his failed Peace Offensive), was the advantage thrown out of the window and could the Allies win the war.

Strategy permeates everything we do, even though we may not be conscious about it. It is about how we deal with everyday aspects. What do you do if someone treats you unkindly? Will you put up a fight? Offer friendship? Walk away? Strategies are most often not designed. Indeed, they evolve. You try something; doesn't work? Toss it out and try something else.

The best examples are those seen in nature. Think about the arms race between prey and predator, or how parasites and influenza viruses keep on adjusting to outrun what medicine can throw against it. The relentless pressure of evolution is the best evolver of new strategies.

Nor is strategy a single thing. It often consists of a number of characteristics that strengthen a generic theme; in companies, these could consist of culture, processes, skills, technology. The best companies in the world have strategies that work. If you break those down, you will find many, many parts that all interrelate and enforce each other. Do not even bother cherry-picking; it's bound to fail, because you do not have the overall strategy.

It makes me think. Can a really good strategy actually be designed? Can it come into existence without the pressure to change? Can it evolve into the target strategy withouth the cultural preconditions being in place?

I think pressure is essential. And so is the holistic view. Without pressure, no one will feel the need to change and without that energy nothing will happen. Without the holistic view, all the measures are shrapnel shots going nowhere. It needs to be part of a bigger picture, the parts reinforcing the whole.

As for evolution or design; strategy suffers from the same problem as software development. Everyone accept nowadays that the waterfall model of software development does not work and smart people invented the concept of iterative development, which for sure works a lot better. So it is with strategy. I firmly believe that it is not possible to design a 100% sound strategy upfront. You use your instincts and experience to aim as good as you can and then evolve the concept as you go.

Culture might well be the hardest one. How do you know you are not shooting for an impossible target? The Will to Succeed may well be blind to impossibilities. Do you have the right people on board? Are you yourself the right person to pull this through?

Personally, I stick to these rules:

• design and look at the big picture, see how it all interacts
  
• know what not to do and block that out
  
• relish a crisis; it offers the opportunity for change
  
• don't be too harsh on yourself for not seeing the entire picture (yet)
  
• be fair in appraising weaknesses
  
• learn from others, read biographies of successful people
  

Despite an intensive, lifelong fascination for anything that has to do with strategy, I still cannot claim to have mastered the subject, though I do claim to derive a lot of intellectual pleasure just thinking about it.

Geeh... migration

Last saturday we migrated two separate email domains and our Google Apps domain into one Google account. Besides the mailboxes, we therefore also had to set up Google Talk buddy lists, export/import the calendars and copy the Google documents.

We started by having a general reset on all passwords, so we could execute all phases easily.

The mailboxes went very smooth. Google has the right tooling to enable this. The only hiccup was the transformation from folders to Google's labels. Deeply nested folders come out really ugly, but hey, what can you expect? I prefer labels to folders, especially since folders do not cater for multiple annotations like labels do. And let's face it, like an animal that is both bird and fish, some mails refuse to be defined by a single label.

Google Talk is easy as well. There is a global setting which automatically makes domain users accept any buddy request -- this is key! After this setting is enabled, it is just a matter of copying the entire email list into the invitation text field for a user and the deed is done. Google could improve its service considerably by having a setting which automatically connects all domain users.

Calendar has the same ease of migration as Google Talk. Just export the calendar and reimport it, mapping to the right user names.

Google Docs is a bit of a pain. There is no way to change ownership of a document to someone outside of the domain (@Google -- improvement?). Luckily, the Google Docs API is splendid. I made a custom script that does the following:

• analyze all documents and flag those that are owned by a user
  
• check the access rights on that document
  
• download the document to a temporary location
  
• upload the document to the new Google Docs account
  
• reassign the access rights to the uploaded document (also using new email accounts)
  

The script performed well. Noteworthy is that on uploading a document in an automated way, a lot of garbage is created by Google. I think it has something to do with uploading a high number of documents in quick succession. Anyway, the script also cleaned up the Google garbage, which was not hard since the garbage documents still had the original filename used for downloading.

After the fact, I noticed I forgot about document labels and email notifications when a document changes. Not a big deal for people, but if you want to be all-inclusive, you should have this as well.

The migration took less than a working day with a minimum number of complaints.

Here's to Google!