Staying Current

When was the last time you heard "how can you consider yourself a software professional if you haven't read X?", only to realize that X was a foot deep in your "must read" pile, had been there for six months, and there was another foot of "absolutely must read" material on top of the first pile? StayingCurrent these days can seem like a losing proposition.

(O.K., so this isn't the real problem. The "read me now or die" piles are just a symptom. The real problem has to do with holding on to unreasonable expectations, and not letting go of the crazed notion that it's possible to ride multiple runaway trains simultaneously. Let's leave that for another day.)

With the problem thusly redefined, what to do?

There are services that summarize books. You've probably seen their ads in in-flight magazines, which is where they target the harried executive market. Earlier in the year I noticed that one service (http://www.summary.com/) had recently summarized several books that where at that moment growing mold in my "read me soon or your career is toast" pile. Well, I thought, what the heck. A hundred bucks gets about three dozen eight-page summaries, and if even one or two could save having to buy and slog through the latest fad business book, it seemed like a fair trade-off.

(I used to snicker at corporate drones who immediately pounced on whatever book was last seen on the boss's desk. It was only later that the value of being forewarned, and thus forearmed, became apparent. A boss who has just been inspired by reading the most recent TomPeters book can seriously destabilize a project.)

Six months after subscribing, it still seems like a good deal. I've retired three books from the "read me soon or never again be promoted" pile after finding their summaries sufficient, but have added two new ones based on intriguing summaries.

That helps the business part of the book pile, but not the technical part. My "read me now or be forever bypassed by the neat stuff" pile is still pretty big. What we need here is some sort of summary service for software books. Somewhere were we can go to get a quick sense of what books are must reads, learn which can be (and should be) ignored, and scan summaries of those for which the summaries suffice. Does anyone know of such a beast?

If you've got a strategy for StayingCurrent, or if you heard one whispered about at a party, please share it.

-- DaveSmith (9/17/96)

I've added a new tactic to my job interviews; I mention the last few cool work-related books I've read, then I ask the interviewer to reciprocate. If the interviewer hasn't read anything lately, that's a warning sign; if the interviewer is passionate about books I also like, that's a sign that we at least have compatible tastes. This strategy means that no interview is completely wasted.

The three books I'm recommending most often at the moment are the 20th anniversary edition of Brooks's MythicalManMonth, Bringhurst's TheElementsOfTypographicStyle?, and Tufte's TheVisualDisplayOfQuantitativeInformation.

I find that the reviews of trusted peers are my best guide to professional reading; if a colleague who loves software for the right reasons recommends a book, it usually contains useful ideas. I haven't yet found any professional software reviewers I trust.

-- BetsyHanesPerry

You know, this makes me wonder why people bother to write books. Why don't they just write summaries instead, and save everyone a lot of time? -- DaveHarris

Books cost too much. Books about software cost way too much. In addition, they tend to have about a 6 to 12 month shelf life when they are about current topics (or trends).

I tend to stay away from books that fall under any of the following conditions:

The topic is so fresh that no one (including the author) could have possibly mastered the subject yet.
The book's title include any of the following words: Dummy, Idiot, 21 days, 3 weeks. (FooOneOhOneInSevenDaysForDummiesInaNutshellSuperBibleUnleashed)
Book length over 400 pages (with large fonts - gee, it must be definitive, the book is so thick!).
Garish typography (always a warning sign...)
"..updates and corrections available at web site.." on the front or back jacket. (This usually means the book is already obsolete).
Published by fly-by-night publisher (I won't name names here...).
Obvious typographical mistakes in the preface or first chapter.

I want to stay current, but I don't believe in throw away books.

-- ToddCoram

A quibble: I buy a lot of software books, and among them I've bought a few Dummies books (HTML for Dummies, VBA for Dummies and others) and, on balance, they are good, certainly well above average for content and editorial quality. They've sold millions of these. One reason, in my experience, is relative quality. -- StevenBlack

For an account of one author's experience with writing a technical book, see http://philip.greenspun.com/wtr/dead-trees/story.html.

I'd like to offer one more item for Todd's list:

Crude hand-drawn or badly-scanned graphics and artwork. I should not see "jaggies", dithered backgrounds, and so on. If the publisher doesn't have time, money, or willingness to do publication-quality artwork, the book is probably no good.

-- TomStambaugh

I have discovered that the number of books & articles I want to read will always expand to fill my available time. It helps to decide which areas you want to keep up with. As obvious as this may sound, I have known people who have attempted to keep up with far too many areas than is humanly possible.

As far as which books to read, I choose by author and then by publisher.

LarryConstantine wrote a good article in the April, 1995, issue of SoftwareDevelopment about staying current. He talked about how practical knowledge of a given topic will only last a few years but learning the theory behind it will last far longer. I believe the article also appears in ConstantineOnPeopleware.

-- AlanHecht

Me too. Here's a tactic that's worked for me since it's drastically reduced browsing time. All new magazines go first through a triage, using post-it notes, wherein promising articles marked. This way, when I later grab the magazine to read it, I can go straight to a meaty article since the triage and ad-browsing has been done. -- StevenBlack

I have books I bought 20 years ago that I'm just now getting around to reading. Likewise, there are books I read 20 years ago that I'm just now getting around to understanding. -- EdBuffaloe

One comment about the "Dummy" books - I was looking for a good book to help me write better GUIs. There aren't many out there that try to be an introduction - there are lots about tying widgets together (I can do that, the results just didn't make sense) and a few about usability that went too far into the weeds, but none that covered the whole picture.

Then I found a reference to "GUI Design for Dummies" ISBN 0764502131 . I balked, but checked out the author's web site (http://www.wordfixers.no), was impressed by her tone and the advice she has on her site, and ordered the book. I was impressed. It's a great introduction that tries to get you to focus on the important aspect - what the user is trying to do - not on the database design or "logical" steps. I'm not going to claim it's made me an expert GUI designer overnight, but I feel a lot more confident about it and the few GUI elements I've made recently have that "feeling" of being "right".

I don't think it could have been published as anything but a "Dummies" book, either. It covers too much ground - usability, navigation, graphic design - too shallowly to ever be a textbook, and it doesn't focus on any given tool or platform enough to be published by a vendor.

And then a sister gave me "Cooking for Dummies", and while it wasn't really needed, it does cover some information no normal cookbook does. I'm still not thrilled with the titles or the covers, but I'm not going to be (too) ashamed of buying "Dummies" books anymore.

-- RobCrawford

Actually, I do need what I think "Cooking for Dummies" might be. Put another way, I'm definitely a dummy when it comes to cooking. Do you think it fits the title? -- KielHodges

Crude hand-drawn or badly-scanned graphics and artwork ... If the publisher doesn't have time, money, or willingness to do publication-quality artwork, the book is probably no good.

Don't judge a book by it's pixels ... ChristopherAlexander's masterpiece "A PatternLanguage" is full of scrappy little scribbled diagrams, and is none the worse for it. Content wins over presentation (though it's nice to have both of course). -- RogerBrowne

I don't think the drawings in Alexander's book count as bad presentation. They're simple, yes; hand-drawn, yes; unbeautiful, yes - but, also, clear and evocative. They add to the book, not detract from it. I bet Tom wouldn't reject the book on account of the drawings even if he didn't already know it's famous. -- GarethMcCaughan