I just finished reading Joel’s—yes, that Joel’s—The Best Software Writing I.
It’s a collection of blog posts from 2003/2004ish, along with commentary-laden introductories from Spolsky.1
It’s a thought-provoking, if not enjoyable read. Also made for great line- and waiting room-reading.
A few of my favorites, from his collection of favorites:
- Ken Arnolds’ Style is Substance: I’ve only been “in the industry” for… four years now? But at every software shop I’ve worked, I’ve had to have the “let’s use a style guide for our code”-series of meetings.
They’re always painful.
The fact that I, some “pissant, just-out-of-college” junior developer, has had to bring up discussing whether or not it matters what the code looks like is something our industry… should be ashamed of.
Arnolds’ solution is tempting.2
- Rory Blyth’s Excel as a Database: I couldn’t keep from giggling every time I read this. All I could think of were the… “non-technical people” at VMware, who used to kill the mail server3 by sending each other and clients .vmdk‘s—you know, the gigabyte-sized virtual machine disks—as attachments.
- Ron Jeffries’ Passion: We talk a lot in The Community ™ about what makes Mozilla special and why people are engaged and interact with each other in the way they do. Jefferies’ explanation of passion really captures that idea, at least for me. It’s also provoking, because it’s made me think about areas for personal improvement on some of the aspects which he uses to define “passion.”
- ea_spouse’s EA: The Human Story: this made a big splash in the blogosphere back when it was written; even traditional media4 picked it up. Times are, I’ve heard, different at EA. Admittedly, any correlation, what with being a publicly traded video game company, are muddled.
Just the same, I’ve heard similar sentiments expressed by ea_spouse from close friends and this notion of “mild” crunches and “pre-crunches” and “heavy duty crunches,” a veritable endless crunch-treadmill that won’t ever slow down is… disturbingly familiar.
- Eric Lippert’s How many Microsoft employees does it take to change a light bulb?: Another familiar sentiment.
No, we don’t have that many development steps (nor that many levels of management), but I have been stared at with blank faces when trying to explain that “Adding this five-minute feature to Tinderbox5 or installing this little upgraded utility on that machine is problematic, because we use Tinderbox for releases, and it will have an effect on releases, or we can’t change that machine’s configuration now.”
The cost is never five minutes.
- Rick Schaut’s Mac Word 6.0: An interesting [set of] anecdotes about how you can make what seems like the 100% correct technical and architectural decisions at the time, but twelve or eighteen months later when you ship, the product is just a piece of garbage.
Secondarily, an interesting set of anecdotes about how telling your customers that all the right technical and architectural decisions were made, doesn’t make the product any less a piece of garbage.6.
- Eric Sink’s The Hazard’s of Hiring: my last couple of experiences have been joining a company on the beginning of growth curve. Since that typically translates to “lots of hiring,” I’ve been thinking about the subject, in both global and localized contexts, lately, so reading this added some good fodder for introspection.7
- John Gruber’s The Location Field Is the New Command Line: An admittedly rehashed8 treatment of The Web as Application Platform and Its Impacts on Desktop Software, but what I find interesting of the treatment is that it questions the value of the “Browser War,” under the predication that Microsoft won. I’m not one for all the squeeing over the “Browser” “Wars,” but it’s interesting that the argument is premised on a point I’m not sure is valid.
Many of these discussions of this nature seemed to be premised upon this [incorrect or debatable] point.
Since all of the content in the book is on teh Intarnets, the only real value here is Spolky’s commentary. And the increased portability of paper.
Both are worth it.
1 And a crapload of footnotes with nothing but URLs, since that’s what you have to do when you translate the web to book-form!
2 But mostly for the ticket sales opportunities he mentions.
3 Which wasn’t that hard to do, since it was Exchange.
4 Soooo Web 0.5
5 Mostly because adding anything to Tinderbox takes much more than five minutes…
6 At least in their eyes…
7 This, which just received a facelift, too.
8 But maybe just for this audience?