Hello Sweden, where are you? .se FAIL

Posted: October 13, 2009 | Author: | Filed under: Uncategorized | Tags: , , | Leave a comment

How stable is the foundation that the web is built on? You kind of wonder after the incident last night where the entire .se-domain was unreachable for about 30 minutes. I still had problems accessing some sites this morning.
Apparently something went wrong during a maintenance routine at the Swedish Internet Foundation.

“For an unknown reason, an extra .se was added to the end of web addresses.”

I wonder what the cost of that little “woops, hey, it worked before” is. Considering how our lives are weaved in to the web to an ever larger degree, it is worrisome that things can go so utterly wrong.


Make it work. Make it pretty. Make it fast.

Posted: October 4, 2009 | Author: | Filed under: Software Engineeering | Tags: , | 4 Comments

That’s my priority when I build stuff: Make it work. Make it pretty. Make it fast.

Work: it does what it’s supposed to do and adds value to the user.
Pretty: it’s easy to use and looks good.
Fast: there’s no performance bottlenecks and the software is scalable and easy to extend.

I’m a programmer but I also want my software to look good and be easy to use. The last thing I worry about these days is performance. Not because performance isn’t important but because it’s usually not a problem until you have enough users and you will never get enough users unless you focus first on the other two.

Ten years ago my list would probably have been: Make it work. Make it fast. Make it pretty. Or even with “fast” being number one.

Not that I’ve ever been a performance freak, tweaking stuff to win an extra millisecond, but there was a time when I was obsessed with doing the Perfect Extendible and Reusable Software Design. My education is software engineering (I have a masters degree) and when I was fresh out of school (which I was 10 years ago) I used to spend a lot of time drawing class diagrams and object interaction flows and lots of other paper stuff that didn’t take me anywhere closer to a working program. I even wrote huge requirement specifications.

Make it work. Make it pretty. Make it fast. That’s my list. What’s yours?