I was just reading Marlena’s blog post on her ultimate nerds and this sentence in her section on Charles Joseph Minard caught my eye:
“There is so much focus in tech on completing your most important work in your twenties or maybe early thirties at a stretch, and it’s bullshit.”
This sentence made me think of one of my ultimate nerds: Donald Knuth. The reason? He is 74 years old and still working to to complete his most important work: The Art of Computer Programming (TAOCP).
Other reasons that, if I had a list of Ultimate Nerds, Knuth would be at the top:
- While writing TAOCP he was so frustrated at the state of the art in electronic publishing that he took years off before completing it to invent TeX and METAFONT, considerably advancing the art of electronic publishing and typsetting.
- He used to pay $2.56 (a hexadecimal dollar) to anyone who was the first to find a bug in his books.
- TAOCP includes the definition of a theoretical computer and assembly language instruction set he invented to illustrate the algorithms in the book.
- His first technical publication was in Mad Magazine. He talks about it here.
There’s so much more that could be said about Knuth but I’ll stop here. I guess you’d have to say I’m a fanboy…
OK, one more thing (this one appeals to me as a tester), a line in a memo from Knuth: “Beware of bugs in the above code. I have only proved it correct, not tried it.” (Other Knuth quotes here.)
This was my first week as an IC (at least in the global address book) and I have to say that I don’t feel like much of what I’ve done so far has been what I think of as IC-related work. I guess the transition will be gradual.
My current main focus is on understanding the compatibility space for our product (which we haven’t traditionally focused on in a global way) and put together a structure for a compatibility test suite. I’m working with 3 other SDETs on this project and up until now I’ve kind of been the lead for this project. Our current working definition of compatibility is the ability of a version of our product to handle artifacts created with a different version of the product. Our compatibility story in the past has pretty much been that if you want to work with an artifact created with a different version of the product and find a problem you should work with the version of the product that produced the artifact.
We have had capabilities of upgrading artifacts (for example, you can upgrade a SQL Server 2008 R2-produced PowerPivot workbook to SQL Server 2012 PowerPivot-readable workbook but you won’t be able to work with the workbook in SQL Server 2008 R2 PowerPivot any more.) but we’d like to, in future versions of the product, do better than that. Getting a test automation suite in place to check compatibility behavior will be key to knowing the state of the product as we work on future versions.
Also, this week, I attended a day long training course that was an introduction to Testing in Production (TiP).
…and realizing how much modern Windows UI assumes mouse rather than touch input. Lots of stuff assumes that you can hover but with a touch interface you can only press…
“People who are more than casually interested in computers should have at least some idea of what the underlying hardware is like. Otherwise the programs they write will be pretty weird.”
-Donald E. Knuth
Today, the HR system transfer of my remaining direct reports to their new manager became final. I am now an individual contributor. I have a challenging first assignment to work on and a new area of the product to learn about. And a lot of other “new”s swirling around.
…there’s not many of them. Brent Jensen gets a place on my blogroll due to his great post on The Test Lead’s Job (although I don’t think he’s currently a test lead). Anita is the other blogger I know who’s in test management. Any others I should be following?