A Week’s Reading For May 3rd

Continue reading →

A Week’s Reading For April 25th

Continue reading →

Stephen Few’s “Bricks”

Stephen Few has introduced “bricks” as a possibly clearer method for geospatial analysis. I think the 9-block version is an excellent idea. Here are my thoughts.

  • One could say that the area of the shapes can’t easily be compared. I don’t think a bar-chart-style comparison even applies here. I can’t ask my preattentive mind, but I felt that I instantly understood the relative size of the 9-square bricks without any regard to arrangement.
  • Given my sense of how I perceived the 9-square bricks, the profiles of each shape could be made more distinctive, especially those in sequence. There’s no need for one side to remain flat. Similarly, why should 3 be a linear extension of 2?
  • I don’t see any problem with color shading, or any need for interior lines in the 9-square brick, and less so with more distinctive shapes.
  • I don’t think overlapping bricks are an issue either. And I’m quite sure it’s not a problem if the bricks are semi-transparent. Semi-transparent bubbles are easy to distinguish when overlapped, and so would semi-transparent bricks. Distinguishing shapes would be easier if the outline of each brick was intensified.

So, I’m suggesting that this makes the most sense as a shaded, semi-transparent, darkened-outline collection of 9 distinctive brick shapes.

I was distracted by the 81-square version. I found a fire truck, a capitol building, a hotel and and tugboat in seconds. The subconscious shape association added a real challenge to comparing the sizes.

The DDoS That Almost Broke the Internet

As someone in charge of DDoS mitigation at one of the Internet giants emailed me this weekend: “I’ve often said we don’t have to prepare for the largest-possible attack, we just have to prepare for the largest attack the Internet can send without causing massive collateral damage to others. It looks like you’ve reached that point, so… congratulations!”

The DDoS That Almost Broke the Internet

Cumulative Identity

Admirable, and a little disconcerting.

Most commentators on the Government Digital Service site have welcomed the move. “There are eight federated ID providers that a claimant can choose – but they will do more than just provide ‘Facebook’ type logon,” said Brian Wernham, author and commentator on project management and government IT. “They will be carrying out background checks on claimants to ensure that they are who they claim to be. PayPal and the others will be providing a ‘Level Of Assurance’ (LOA) on each claimant’s identity using the recently published Cabinet Office Identity Assurance ‘Good Practice Guidance’ (GPGs).”

Resilient Business Software

The outline of the resilient (almost symbiotic) business software platform includes:
  • Capturing as much reality as possible. This is the untapped potential of the big-data revolution in business. It is also the critical benefit of “creating feedback loops”.
  • Representing reality with as little transformation as possible. Graph databases, naturally.
  • A deep concept of data provenance, meaning that attached to each piece of data is everything you may want to know about how it came to be. This is really applying the graph database concept to individual pieces of data, be they input, imported or calculated.
  • A related concept is deep versioning of code, data and processes. This enables agility in business processes and avoids further complicating exceptions due to institutional amnesia.
  • Visualization as an exploration of data, process and human interaction. See exceptions sticking out farther than their cohorts. 3D density and sparsity shows positive and negative space. Graphs don’t have one representation; play with the levers to change perspectives, change the domain and compare to the past.
  • Pervasive predictions and recommendations. This takes the modern approach of “look at a ton of data” rather than complex algorithms. Predict the data that is needed by the shop foreman at 12pm because it’s what has been requested in the past. In a graph database, paths become well worn like hiking trails, showing what activities and people successfully resolved the problem.

Still, I see friction. This is a resilient system that helps people fit software processes to the business needs. But it’s only an aid. Business won’t run on auto-pilot. Repeatable, accountable actions will always be necessary. Rule-breaking exceptions always happen, even with a system designed to minimize them. And there will be plenty of counter-examples where minimizing exceptions will destroy some competitive aspect of the company.

What would emerge? Some things like running an entire production facility from your phone because the system can anticipate patterns and predict needs. It then feeds you only what you likely need to know, with links for more details. Eventually the body of knowledge includes a good number of exceptions and the path to resolution gets recorded (and thus reinforced) time and again.