testing just 2 users per design reduced the probability of being wrong from 50% to 24% — cutting it in half.
Quit debating what works and go gather DATA. Then, go where the data leads you. Anything else is examining chicken entrails.
Monday, June 22, 2009
Data Trumps Guesses
Jakob Nielson's Alertbox has been one of the best sites on web usability for many, many years. A recent post does a nice job of showing just how pointless it is to argue in a meeting about what design is probably the best.
Quit debating what works and go gather DATA. Then, go where the data leads you. Anything else is examining chicken entrails.
Quit debating what works and go gather DATA. Then, go where the data leads you. Anything else is examining chicken entrails.
Monday, June 15, 2009
What is great blogging
Last year Dave Weiner did a great job of defining what makes for good blogging. The criteria don't just apply to blogging, but any writing in general.
Of course, Joel Spolsky, who I think is great, rants about how anecdotes are not science. This is true, but what Joel misses is that popular-science writing should be considered examples of good journalism, not good science. If the observations and areas of study are new and interesting, then by all means someone who is an expert at digesting and presenting the information (a journalist) should help us all understand it.
People talking about things they know about, not just expressing opinions about things they are not experts in (nothing wrong with that, of course).
Of course, Joel Spolsky, who I think is great, rants about how anecdotes are not science. This is true, but what Joel misses is that popular-science writing should be considered examples of good journalism, not good science. If the observations and areas of study are new and interesting, then by all means someone who is an expert at digesting and presenting the information (a journalist) should help us all understand it.
Friday, June 12, 2009
GPG for data at rest. TLS for data in motion
It takes a while to read, but it's very much worth it. If you still think you should be writing your own crypto code, or even using native crypto libraries, you are DOING IT WRONG.
P.S. Same goes for parsers! Learn to use ANTLR or any other of the hundreds of parser generators out there.
P.S. Same goes for parsers! Learn to use ANTLR or any other of the hundreds of parser generators out there.
Sunday, June 7, 2009
Eclipse upgrade
I was trying to get eclipse to install an update and it just wouldn't do it. Specifically the m2eclipse plugin. I kept getting errors about being unable to fix a dependency.
"maven eclipse can not find a solution 3.4.2.M20090204-0800"
Turns out several things help. 1) turn on every site as an available update site, 2) refresh everything, 3) starting at the bottom of the resolve dialogue begin unchecking items one at a time until the error goes away (yeah, obvious), 4) starting at the bottom again, turn things back on to ensure you get everything you need.
"maven eclipse can not find a solution 3.4.2.M20090204-0800"
Turns out several things help. 1) turn on every site as an available update site, 2) refresh everything, 3) starting at the bottom of the resolve dialogue begin unchecking items one at a time until the error goes away (yeah, obvious), 4) starting at the bottom again, turn things back on to ensure you get everything you need.
Saturday, June 6, 2009
Subscribe to:
Posts (Atom)