Becoming a data scientist
I'm not a data scientist, really, although I work with a bunch of people who are. This O'Reilly Radar article is great reading.
Book review - "Being Geek"
Book Review:
Being Geek, by Michael Lopp
http://oreilly.com/catalog/9780596155414/
I guess I've got three criteria for shaping my reading list: Will I learn something from this book? Is it something I want or need to learn? Will I enjoy reading it?
"Being Geek", by Michael Lopp, scores high on all three points. As someone with a background in software development management, "Being Geek" struck home with me. I recognized myself, my staff, and many of my friends in the pages. I also got some good ideas on how to become a better software developer and development manager myself.
The book's genesis is in blog posts from the Rands in Repose blog, but edited, expanded, and organized. There's also some new material. The book is organized into several sections - how to manage your own career and job search process, how to deal with management: managing up, managing down, managing sideways, and managing toxicity, tactics and strategies for dealing with the day-to-day panics and crises of life as a developer, and how to think about how, why, and when to consider your next gig.
The two sections that I found most valuable, and will be returning to, are these: first, prioritization and keeping both a task list and a "trickle list" - strategic items that need a little bit of attention every day, not necessarily a big box of attention and then they're done. Second, the chapter on managing managers - communication styles, meeting schedules, and dealing with surprises - led me to some useful introspection about my own strengths and weaknesses as a manager. The section on "The Nerd Handbook" didn't really resonate for me, but the stereotyping in the chapter certainly has plenty of basis in fact.
Lopp has a clear and conversational style of writing that gives you the feeling of sitting with him at a table (probably on your second glass of good beer each) and a notepad between you for quick diagrams, while he explains his view of how software development teams work. He's opinionated, he doesn't pull punches, and he's occasionally pottymouthed for emphasis, but it all works. The book offers advice and food for thought both on how to get through the day, and how to get to where you want to be in your career.
Several other reviewers commented on bad editing and proofreading in the book. I also found this to be true in the first advance copy I was sent. However, a subsequent drop of the ebook had fixed the dozen or so errors I found in the initial draft, and I think they're also addressed in the print version.
If you're a software developer, you owe it to yourself to read this book. If you're not sure, check out the Rands in Repose blog as a sampler of the material that's here, but also be aware that there's material in the book that's not on the blog - the book is more than a collection of posts. I wish this book had been out 15 years ago when I was contemplating moving into management; I think I'd have gotten better faster.
Disclosure:
O'Reilly offered me a free copy of the ebook in exchange for the review. I'll likely be buying the dead tree version anyway, in order to loan it to some friends who really really really need to read it.
Yeah #yyz #rush http://twitvid.com/PPCZG
Yeah #yyz #rush http://twitvid.com/PPCZG
$70 don't buy what it used to at a concert... #rush #statefair
$70 don't buy what it used to at a concert... #rush #statefairLaurel at the Macalester family picnic
Laurel did some facepainting too...
Are generics killing java?
Eric Armstrong writes a great post http://www.artima.com/weblogs/viewpost.jsp?thread=299081 that captures how I've been feeling when I try to write java lately. It hurts way more than it used to.Working on Scala
I'm feeling a little annoyed with the books on scala. None of them talk about IO in detail, and only Venkat Subramaniam's book even bothers to touch on it. The other ones that I looked at don't even mention it. At all. C'mon guys - real programs do IO!
I can't help feeling that scala is too big for its own good. It's not as elegant as ML or OCaml, and not quite as unsurprising as Java.
But I'm not giving up yet. Learning ML wasn't easy either, and that proved to be worthwhile.



