The iNexus

As I’m sure you are aware, Apple had a media event recently covering a topic near and dear to my heart - the Mac. Outside of the new iMac and iLife ‘08 and iWork ‘08 (wow, that’s a lot of excitement already to pack in an hour and some odd minutes!) there was a little announcement about your favorite online service and mine - .Mac.

And what a doozy the news was! The standard storage space for each .Mac member was multiplied by a factor of ten for a total of 10 GB (yes, GB) of storage. All those who doubted there was any value in the ‘paltry’ service should now see there’s a lot more to offer there in terms of webmail, storage [you could set up 9GB for mail and 1 GB for iDisk if you so choose], and even “Web 2.0″ features for photo galleries and web pages (authored in iLife ‘08, of course.)

I think this is great news and I was excited to read about all of the changes. It means dotMac Menu’s iDisk size calculations are a bit dated now and don’t account for this, but I’ll take care of that along with many other great enhancements in the next major update

I read a link on TUAW to a post by a Handbrake developer about his view of what open source development is and isn’t. My shorthand response: never attempt to speak for an entire movement, it tends to come back quickly.
Disclaimer: By and large I agree with TUAW’s framing of the topic and do not hold them at fault, and my apologies for the title inferring this.

(more…)

There’s a quick followup I’d like to make that better elucidates what I was vainly saying in my Interviewing for Programming Jobs post.

If you haven’t, you should really read Wolf’s great followup on programmers and coding.

Jerry W. Walker really says it all when he states, “… would suggest that the most successful programmers of that age were those who were good at algorithmic reasoning. The languages (any number of assembler languages, Fortran, COBOL, PL/I, C and such) were generally easy enough to learn within a few days and you could get good with them within a month or two. I believe that the most successful of our programmers today, on the other hand, are those who are good at learning natural languages. I would suggest that the ability to learn, not only the Java language, but the Java packages and method calls therein, is a great deal more like learning French than it is like learning algebra or calculus. The active vocabulary has become huge.”

I too think that modern programmers are better measured for their language breadth and skill, not necessarily the need for logarithmic analysis. A well-balanced programmer hangs many tools from his proverbial belt, but in the end I think it’s how you use the tools and what you can do, not arbitrary challenges, that should be the standard.

There’s a question I’ve often wondered - why do technical teams feel a need to use academic questions that only serve to test logic challenges when these have nothing whatsoever to do with the actual job?

Yes, I’m aware these can condense logic/problem-solving skills down to some degree, but ultimately they do not have a direct application to the abilities relevant to the job, which is writing code (which requires language skills, toolset knowledge, and dedication, not just problem-solving.)

I’ve interviewed with several companies throughout the years and each time a revolving series of questions about b-trees or linked lists comes up. When is the last time a practical (Mac OS X) programmer spent time coding up a linked list (I’ve yet to see one in any project I’ve worked on) and in discussing implementations and designs I’ve yet to have someone ask for the big O notation for its run time.

Yet over and over these questions come up. Do they actually serve to prove ability? I’ve had the (mis)fortune of being entirely self-taught yet I’ve been able to work on several notable projects, and build a company, without using these computer science abstracts. Does that make me an inefficient programmer? I think not. But in the eyes of interviewers, apparently so.

So here’s my take - you want a computer science degree, say so and stop taking in candidates that don’t meet that exact criteria. You want someone experienced in writing Objective-C, say so and you’ll likely get a qualified candidate with enough screening. But don’t waste time conflating the two, they are not mutually inclusive.

I was working on some QA (quality assurance) for the upcoming Compare Folders public beta and was looking at the top of the files in Xcode, noting some of the dates. It’s so great to step back a bit and wax nostalgic about what’s came before.

In November Compare Folders will officially be 2 years old, and this month marks the 1 year anniversary of the introduction of .mac/iDisk support. I still feel CF is still incredibly fresh and new, and working on it brings all the fun back, with each improvement, change or tweak bringing back everything.

So to all who code, take a moment to step on through your projects and feel out the dates… it’ll bring back memories, I’m sure. To those who inspire, contributing on lists, in code, or by sharing ideas, bugs, and crash reports, and especially the customers, thanks!