Kyle Smith

As part of the preparations for our move to the Boston area next month, I’ve been shredding old receipts I had saved for some reason. It was rather enlightening to look back on what I’ve been doing for the last two and a half years.

Most of the receipts from 2005 were from Potsdam favorites such as Eben’s Hearth (big beers & wings), Maxfield’s (beer), P&C (beer), T&R (liquor), and Sergi’s (fat bags). Looking back on it, I can see where my spending and temporary drinking habits started, as illustrated by the daily Maxfield’s receipts between April 1st and May 8th and the weekly trips to Eben’s before COSI meetings that spring.

You can even see how my habits have changed. There are plenty of receipts from Starbucks and the Eveready Diner in the fall of 2005.

There are better ones too, like December 17, 2005, when Stacy and I went to New York City for the first time. The receipts reveal the plans for the day: a train ride down & back, coffee at Dean and Deluca, cheesecake at Roxy’s in Times Square, dinner at Planet Hollywood (it sucked), a quick stop at Starbucks, and then Fiddler on the Roof on Broadway. Reading through them the whole trip came back to me.

Stacy always asks why I keep receipts. Now I have an answer. Because they’re just as revealing as photographs.

Had an interview recently with a manager about a development position.  When I asked for his organization’s view on spending company time to work on patents, publications, and Redbooks his answer was that they certainly valued patents, and depending on my family situation I should be able to find some time to work on those types of things.

When I later asked about working on an M.S. or participating in a BizTech project he started talking about turning down my workload to 80% and the management team understanding that I’d be working on other things.  Elsewhere in the conversation he mentioned “turning it to 100 or 120.”

Am I the only one out there that doesn’t want a job where I’ll be expected to sacrifice time with my family in order to work on intellectual property for the company or am treated as a machine with a knob that can tune how much and how hard I work?

One of the main selling points of software as a service is that people don’t want to maintain their software and don’t want to worry about security. There are plenty of services out there today that people love to use — Gmail, Google Calendar, Google Notebook, Twitter, Flickr, IBM’s Dogear and Plaxo.

I’ve had my own domain name and web site for almost 8 years and in recent years I’ve grown tired of maintaining the various software packages that my site is comprised of — WordPress, Gallery 2, WikkaWiki, and others. Software as a service is a partial solution to this.

The problem with the current generation of services is none of them are integrated all that well. I don’t really want 7 distinct services to manage and keep up to date. It’s no more efficient than maintaining the software myself.

I would like to see those service providers team up with web hosting companies to offer an integrated solution where the hosting provider provides a breadboard of sorts that the services plug into. The hosting provider would be responsible for my site, including billing, but the actual functionality of the site would be provided by the service providers.

If that vision isn’t working for you, think of the hosting provider as an EJB container (which provides all the common supporting services you want) and the service providers as the developers of the add-on components (the actual EJBs) that are deployed in the container. I like this notion even more, because it suggests that if I’m not happy with a particular service, I can swap it out for another that’s written to the same API.

No more building your site from scratch and then trying to integrate it. You log in to your hosting provider, pick your domain registrar, your webmail provider (Gmail, Squirrel Mail), your photo management solution (Flickr, Picassa, Gallery), etc. and then click the Make It So button.

The people that talk to me on a regular basis know that I liked to read in cycles. Around the beginning of the year I was focused on Isaac Asimov. More recently I’ve returned to my people/project management kick and went through a dozen or so books from the IBM site library. Of the dozen, there were a few that were particularly noteworthy that I’ve added to my list to recommend to people. The list:

• Peopleware by Tom DeMarco and Timothy Lister
• The Deadline: A Novel by Tom DeMarco
• Behind Closed Doors: Secrets of Great Management by Johanna Rothman
• The Mythical Man Month by Fred Brooks

(Sorry, cleaning out the queue.)

For once I am impressed by the level of documentation available for both GNOME Screensaver and Pidgin.

I’ve been using Sametime lately in Windows and like the fact that it sets my status to away when the screen is locked. The open source community has been talking about doing it for years with Pidgin, but I couldn’t find an existing plugin that did it.
I was looking forward to working with DBus and the Pidgin plugin API when I came across the GNOME Screensaver docs and FAQ. Those two pages plus a faint memory of gaim-remote and voila:

my($cmd) = "dbus-monitor --session \"type='signal',interface='org.gnome.ScreenSaver',member='SessionIdleChanged'\"";
open (IN, "$cmd |");
while (<IN>) {
if (m/^\s+boolean true/) {
s y s t e m ("gaim-remote 'setstatus?status=away&message=Away'");
} elsif (m/^\s+boolean false/) {
s y s t e m("gaim-remote 'setstatus?status=available&message='");
}
}


Update: sorry for the spacing of some of the code, but it looks like there’s either a WordPress, PHP, or DreamHost issue with me putting in a particular Perl function call that also happens to be a PHP function.

I’ve been taking a small break from reading Isaac Asimov’s Foundation series to read some other books I picked up from the newly-discovered IBM Poughkeepsie site library. One of them I finished was Cem Kaner’s book Lessons Learned in Software Testing, which had one lesson in particular on evaluating tester performance that caught my eye.

How should a tester (or other employees) be evaluated? Rather than look at arbitrary metrics like bug counts, familiarize yourself with their work (the actual bug reports, documentation they’ve written), collect comments from coworkers or project stakeholders that have interacted with them, and consider the following items:

• What fights does he get into and why?
• How well does he meet deadlines?
• How well does he keep his own promises?
• What kinds of problems does he miss?
• What types of assistance has he provided to other testers and
  programmers to make them more effective or more productive?
• Is he gaining new skills? How well is he transferring his skills to
  other testers?
• What issues has he taken stands on in your company? How do these reflect on his business judgment and personal ethics?
  (Source: Lessons Learned in Software Testing, Kaner et al)

For those of us that have worked at IBM, being evaluated on the above items would be a pleasure compared to Personal Business Commitments (PBC) ratings and results, which never seem all that concrete to me.

My second redbook was published on Wednesday.  The book shows some of the ways that z/VM can be useful when testing and developing applications.

In other fun news, three of the original savages (Eli, Jason and I) will be reuniting in June to work on a z/VM basics textbook here in Poughkeepsie.  If you’ve used z/VM before and found it painful, this will be the book for you.

A second article that caught my attention earlier was from Seth Godin and talked about sheepwalking.

I agree with Seth that the employee can’t be faulted initially, but after the typical probation period I think it’s entirely the employee’s fault if they aren’t able to break from the flock of average performers and demonstrate some initiative. As much as I’d like all companies to have creative cultures like Google’s or that of a startup, it’s not going to happen. Once the company is established the happiness of the shareholders or funders inevitably becomes the primary concern.

Thankfully, I think as long as a company is willing to allow people to innovate for a few hours a week that’s enough. If you’ve hired the right people they’ll take the initiative and locate or create opportunities to be innovative and make the most of that time.

When I first started full-time at IBM I quickly got bummed that my manager wouldn’t let me go off and work on any project I wanted to. After several months I’d had enough and started working on small projects on my own in my spare time. Once I’d demonstrated to him that I was able to accomplish my regular work as well as my side projects, he started suggesting other opportunities to me and giving me more freedom.

Since then I’ve worked on two redbooks (and am scheduled for a third), started writing an article series for developerWorks, signed up to teach a weeklong workshop in Ontario, and participated in two “patent farms”.

The point? As long as you’re willing to take the initiative being creative shouldn’t be a problem. An easy place to start is to determine where the boundaries are between your job role and your coworkers job roles and then fill the gaps (don’t try to take over their roles).

If you’re not willing to step outside your comfort zone, be prepared for a very boring 9 to 5 job and an equally mindnumbing career.  If you’re looking to change that behavior you might check out Robert E. Kelley’s How to be a Star at Work (thanks to Bryan and his “How to Kick-Ass at Work” discussion group). I highly recommend reading it.

I doubt many (if any) of you are involved in hiring employees but I came across a post on TrueTalk by way of Jason Yip which had an awesome quote:

Friend of mine used to say, if you need to hire something that climbs a tree, hire a squirrel. You could train a turkey to do it, but it’s a heck of a lot simpler to go with the squirrel. Or, “hire for temperament; train for skills.”

Unfortunately I’ve encountered a number of employers that do the exact opposite. They hire the people that have a particular skill they need right now and neglect to look at the rest of the candidate. As soon as they see “Linux” or “Java” on a resume tunnel vision develops and everything else is forgotten.

When the next project starts and the manager (or a coworker) realizes the employee doesn’t have the skills or adaptability to do it properly, it’s too late.

When I interviewed at Microsoft in 2005 I got the sense that even though they had an immediate need for a particular skill, they refused to give in to their desire to hire someone that had it without looking at them as a whole and considering how they would perform on project N+1 or N+2. The writings of Joel Spolsky and William Poundstone seem support my view. I also get the sense that Google looks for more talented people (like my friend Dave from elementary school).

Sadly I’m pretty sure companies won’t improve their hiring process even though they’re already being bit by this problem.

This morning the first article in a series on IBM Director that my coworker Jay and I are writing was published on developerWorks.  You can read it here.