Thanks for the comment, Alister! I’m excited to see how this works out too.

Plus I don’t need to worry about it going down or running out of RAM. WordPress.com know what they’re doing performance wise when they host some of the biggest blogs on the planet (CNN, Flickr, RedHat, Time, BBC) on the same infrastructure.

They automatically add things like iPad and iPhone specific versions without me needing to do anything, which I think is great (view http://watirmelon.com on your iPad to see what I mean).

You can choose not to have advertisements on your self hosted wordpress blog, but that’s a paid option anyway on wordpress.com, and I personally find them so unobtrusive I don’t mind having them around.

Just my opinion obviously, I’ll be keen to see how you find self hosting.

Yes, your link points to the Phoenix checklist Andy and I were discussing.

Thanks for pointing out _About Face_. I’m a fan of the book and in other situations have lobbied for error handling policies such as the ones he advocates.

A little more context may help here though. The application in question is software to program a pacemaker. A cardiologist is allowed to enter conflicting parameter values into the GUI. The fields and values are not constrained, because sometimes a handful of parameters interact and we don’t want the user to have to worry about the order in which he/she sets the values. At the same time, we want to warn the user of a “bad” input, because a life is at stake here.

Basically, when a field is changed to a “bad” value, we don’t know whether:

a) The cardiologist accidentally chose an incorrect value.

b) The field change is part of a larger plan. After the cardiologist changes a few more fields, the conflicts may be resolved.

In (a) we want to alert the cardiologist right away while the action is still fresh in his/her mind, whereas in (b) we don’t.

What is constrained is the “Send these values to the device” operation. That button is not enabled until the user sets all parameters to non-conflicting values.

While a freshman in college, I took an introductory physics class where one of the professors, though a really nice person, insisted on using the adjective “obvious” to describe most concepts. Every now and then, something was “more or less clear” or “…

There are quite a few aspects to this that I think are important. As I considered the comments posted here and other verbal comments that I have received, it has become clearer that I think the factors fall in to at least two main groups: classroom …

That doesn’t mean to say that extroverts can never succeed in the field — obviously. It just means that extroverts prefer a routine that gives them more opportunity to interact with people.

This might be a good time to clear up some myths. Extroversion does not imply disorganisation and introversion does not imply an inability to stand up for oneself. They are purely measures of gregriousness.

You’re absolutely right of course – these are generalizations, and I’m very happy for any candidate to show me that they don’t belong in this generalization (in fact the first two people I hired did just that!). And equally I’ve rejected perfectly good candidates because they were too introverted, and couldn’t stand up for themselves in the kind of lively debate you suggest.

In the face of so many potential candidates, however, there are tendencies that I look for, and this is one of them. It would be a problem if it were part of a pass/fail checklist (hires here must have a degree, which to me should be a ‘good to have’), but as an indicator I think it works out fine.

Cheers, Paul

I’ve been doing some interviewing on potential hires recently, and a certain amount of introversion is something I actually look for. I associate extroverts with a certain lack of organization and attention (a generalization of course), which are not characteristics conducive to spending many hours doing the same kind of things in a systematic way.

There are definitely limits to this, and I think testers are more outgoing than developers as a whole (but less than the next step ‘up’, business analysts!).

I’ve also seen statistics that point to self-selection of introverts into software related fields. As I recall, the dominant type is INTJ across software disciplines. Introverts are definitely the minority in the general population.

Esther

You might check his web site, http://www.asktog.com/ for leads.

http://www.bo.uiowa.edu/~TrainersNetwork/handouts/Handout_1.pdf

It’s missing at least one key question, “Whose problem is it”.

I find Pete’s example interesting, after reading Cooper and Reimann’s “About Face”. Error messages are at best annoying to end users, at worst (if you believe Cooper, and I think I do) a constant low-grade humiliation. Cooper makes a strong and convincing case that much effort should go into eliminating error messages. The application should either constrain input (e.g. by the use of controls such as the “spinner”); or accept unbounded input without validation, but prevent the “incorrect” input from doing damage, and provide unobtrusive feedback about the possible consequences of such “incorrect” input.

Also see an older blog entry of mine about error messages from the tester’s viewpoint (http://bossavit.com/thoughts/archives/000029.html).