Be careful with what you ask for

An article, posted almost 2 years ago filed in , , , , , , , , & .

Since we all seem to know that every other field in the registration form is another percentage of users failing to register*, we think of alternative ways to gather information. We gamify the user profile completeness by adding a progress bar to our user account, we present a full form after the confirmation link or we ask questions while using the application.

But there is another reason why we might not even ask all the questions. Ask the wrong questions and you may alienate your user.

It can be relatively minor things like picking your favourite colour, where the form just lets you pick one colour, while many have multiple. But it may also be more personal (or one could argue, more political): not everyone defines oneself as male or female, so why only present just these options (and do know that it isn’t particularly nice being referred to as ‘the other’ all of the time).

These issues are well discussed in this [talk by Ca…

Continue reading...

No signing in

An article, posted more than 6 years ago filed in , , , , , , , & .

IMPORTANT: the assumption made here is incorrect. I suggested using a hashing function, but one should make a special message authentication code function such as HMAC

A thing I've been rediscovering as of late is the bookmarklet. Not that I use many, but in contrast to many of the browser extensions, bookmarklets are really minimalistic and hence very simple to use (although installing them on mobile devices is not) pieces of software. Currently I use the Tumblr, Instapaper and Pinterest bookmarklets, but they all share a common problem: they require you to authenticate before you can actually use them.

Not satisfied with the third parties, not satisfied with hot they work…

The three step bookmark proces typical among bookmarklets that post something

I'm using the Tumblr blogging service simply because it makes posting, via its bookmarklet, easier than posting s…

Continue reading...

Wat een beetje murb al niet kan doen.

An article, posted more than 8 years ago filed in , , , , , & .

Mensen vragen mij wel eens wat ik doe. Zo’n gesprek loopt vaak uit in schaamte weglachende opmerkingen over dat zij ook zo veel moeite hebben met het gebruiken van de nieuwe apparatuur (veelal oudere mensen), waarop ik gelijk op kan reageren dat dat nou precies is waar ik mijn bijdrage denk te kunnen leveren. Jongere mensen reageren vaak zo van, uhuh… (ze zijn immers alleswetend) en beginnen over coole gadgets, waarna ze in de loop van het gesprek er ook achter komen dat de techniek toch eigenlijk niet zo werkt zoals ze wilden dat die zou werken. Techniek frustreert toch nog steeds te vaak. En dan kan dat ene apparaat wel perfect werken, toch moet er vaak ook informatie van het ene apparaat (lees ook b.v. software) naar het andere apparaat. En daar gaat het, ondanks dat we in dit moderne leven zo vaak informatie uitwisselen, nog vaak mis.Dus. Wat kan ik voor uw organisatie betekenen? Wel, volgens mij kan
een bedrijf op de lange termijn alleen maar succes…

Continue reading...

Why we don't need multi tasking

An article, posted more than 8 years ago filed in , , , , , & .

Because computers can multi-tasks doesn't mean they should. People are actually quite bad at multi tasking. Computers, like any other tool, are made to support tasks. Make things easier to accomplish. Requiring users to multi task is far from supportive. Every interruption, which switching interfaces is, takes time to recover from. Hence, instead of promoting the idea of multi-tasking, computer makers should think more about completing tasks users are confronted with.

Instead of designing top notch 'solutions' for an entire 'office in a machine', computermakers should design solutions for a single clerk's job, or maybe even just a part of that job and find a way to nicely integrate in that clerk's job. Computers, or the softmachines powered by them, should be attempts to support an entire task without forcing the user to switch interfaces during task execution. Bad and good designs should be tested against the vision of perfect support for a single (ty…

Continue reading...

Don't design to make it usable, but useful

An article, posted about 10 years ago filed in , & .

It turns out that I finally get to understand what I would like to specialize in, a designer of user experiences in the field of I(C)T. But what should that entail? What is it that a User Experience designer can add to the value of a project?A project starts with a goal, or maybe more like a question, e.g. 'I would like to have something that helps you in getting X done'. If you think about it, solving X may be simple. You gather a set of requirements, translate these requirements in a nice graphical user interface, which may even correspond to all accessibility guidelines available at that time and would work of course flawlessly from a technical perspective, and get it implemented. If done correctly, the client will initially be happy (because it does exactly what it was supposed to do, and you may have gotten him or her to agree on a few sketches and flowcharts you drew in the meantime). Such a process, however, is not considering the needs of the actual users. It is hard to get …

Continue reading...

murb blog