When I own a Ferrari I will expect to have to pay for regular servicing.

Having said all of which, I did notice the costs of the project mounting. I wonder if one could correlate the time spent with SAS to the project budget, and compare this to a benchmark project in which they were not involved in some way?

When I own a Ferrari I don’t plan to call on the services of a chauffeur too often.

It just goes to show, I suppose, that best practice can be expensive and this is, perhaps, why our SAS environment is in the mess it is. One thing I’m very wary of is the amount of resource assigned to seemingly mundane tasks where I can’t see the value they will add. This may be myopia on my part, and I am optimistic that experience (to be gained) will teach me the error of my ways, or that experience (already earnt by colleagues) will find the budget for the SAS resource.

Technorati Tags: criticism, SAS

Any ideas for more imaginative, or non-intuitive uses?

How about

proc alpha;
   q='';
run;

Technorati Tags: alpha, SAS, wolfram

In his recent blog post Stephen Wolfram tries to answer some of these questions:

What is Wolfram|Alpha, then?

It’s going to be a website: www.wolframalpha.com. With one simple input field that gives access to a huge system, with trillions of pieces of curated data and millions of lines of algorithms.

What does that mean?

…one would be able to ask a computer any factual question, and have it compute the answer…
I’d always thought, though, that eventually it should be possible. And a few years ago, I realized that I was finally in a position to try to do it.

But that’s never going to work is it?

I wasn’t at all sure it was going to work. But I’m happy to say that with a mixture of many clever algorithms and heuristics, lots of linguistic discovery and linguistic curation, and what probably amount to some serious theoretical breakthroughs, we’re actually managing to make it work.

Wow! Google, your days are numbered!

Technorati Tags: alpha, google, wolfram

Well, whilst they may help, I think Edward Tufte has identified some of the key issues in his paper on the Cognitive Style of PowerPoint. I’m a great fan of Tufte and devoured his books on long commutes: the only time a fellow commuter has ever taken an interest in what I have been reading was after they had been admiring his rich pagination over my shoulder – I was too engrossed to notice.

Having bought the books through Amazon, their alogrithm suggested that, based on my spending patterns, I might enjoy certain other of their offerings. Two spring to mind as having been pushed relentlessly through “Amazon’s recommendations for you” pages: Information Dashboard Design and Presentation Zen. Neither was in the same league, nor did they aim to be. IDD was interesting, had some neat ideas and a cunning new graphical scheme for displaying a lot of information densely; coming from Tufte’s work, I found Presentation Zen attractive but very shallow. This disappointing performance has turned me off Amazon’s recommendations.

But is this fair?

I suppose I imagine reading as a journey, with each step taking me ever upwards. The problem is that if great books are distributed on a power law basis and you read one of the very best, then there is likely to be a lot of (relative) dross in the foothills, and I wonder whether this exposes a problem with Amazon’s analytics: whether they take into account the sequencing of purchases (a reading vector), or just correlations between purchases. (Of course, I’m assuming that my estimation of the relative merit of these books is absolute, which may be very far from the case!)

I took a route to Tufte via Jon Moon’s books; perhaps other people have travelled via Presentation Zen, and Amazon is simpy showing me the footsteps around Mount Tufte. I’d be interested to hear from anyone who knows more about how Amazon works: do they have someone interested in BI and Performance Improvement?

… talk of which, brings me neatly to Tammi Kay George: I was inspired to write this after reading her recent blog post in which she mentions Presentation Zen. I couldn’t help wondering if she had been bludgenoned into buying it by Amazon, too: I suspect there may be an overlap between her library shelves and mine, and if there is, my decision to buy Presentation Zen undoubtedly will have increased the likelihood of Amazon recommending it to her! It’s all my fault!

Technorati Tags: amazon, analytics, bi, powerpoint, presentation zen, recommendations, SAS, statistics, tufte