Home / Pretending to be the user

Pretending to be the user

As someone who crafts experiences, I’m often engaged with product owners, developers, marketing folks, etc. As such, I’m like an information funnel. Good or bad, I listen to everyone involved. So you can imagine, I hear a lot of ideas. Usually those idea’s take the form of “If I were the user, I’d expect it to work like...”

This unfortunately is a bad practice. You and I, we are not the user. We’re merely one of many. As architects of the applications we’re building, we are so much more advanced than the average user. Drupal is no exception.

Currently, the users of Drupal, are by-in-large also the architects of applications - so my argument is slightly diluted. But is Drupal always destined to be used by technically savvy folks? I'm hopeful it is not. So how do we get over this hump?

To start with, stop pretending to be the user. If you must pretend, acknowledge that you're far more informed than the average user - and biased because you're the architect. Said differently, of course I understand and like my designs - I built them. Put your ideas in front of less savvy users. Build for the novice, and you'll also be building for the expert.


Posted on by Tj Holowaychuk (not verified).

Seems like things are starting to get better with D6, although I must admit I was confused even by the new submit button placements which still through me off. From all of our clients it was mainly workflows and terminology that threw them off. They did not understand that when they want to add "Gallery Album" that they need to create a "Taxonomy Term", those sorts of things.

I will admit as a developer I have been thinking much more with the end user in mind lately, one main project being my Drupal specific WYSIWYG. URL is replaced by 'Location', alt is replaced by 'Title', etc. I believe tools like this could perhaps have a 'developer mode' which would enable dev lingo as well as additional tools that would merely confuse the average user.

Posted on by Gregory Heller (not verified).

Good point Jeff. I agree we are definitely not the end users we have in mind. Every time i work with a client to try to bring them closer to the level of "power user" able to really take charge of their website I find myself saying things like "Ok, the language here is really confusing, try not to get hung up on it..." or "well, you'd think you would find this over here, but actually it is in this other place..."

I see a tendency to push new functionality out the door in drupal modules without strong consideration of the usability. Often in the first few betas the usability gets improved, but this confounds early adopters like myself who may learn it early one way, only to have it changed later. It would be great to see more "measure twice cut once" planning, especially of complex modules (thinking cck, views, panels) with usability high on the list of considerations.

What would this look like? Probably alot more user interviews, paper prototyping and user testing to improve and simplify the UI/UX early in the process, before it is too late to make changes.

Another opportunity is the tremendous number of newbies coming to drupal all the time, I think that many more experienced drupalistas overlook the opportunity to ask "how would you expect this to work?" in a rush to explain how it does work. Newbies are actually a vast and untapped resource for user testing and design improvements, we should make use of them, rather than simply forcing them to learn how to use the sometimes convoluted UIs we have become complacent about.

Posted on by peach - all dru... (not verified).

Luckily I happen to know the most suitable test person to test if something is easy to use for 99,999% of the population. (my mom).

Posted on by eigentor (not verified).

I appreciate these comments coming quite often from the acquia folks lately. Yes, I also believe newbies are a big chance. Recently I have again come across the "where is my content" problem. You create a node, navigate away: where the hell do I find it.

Hopefully Dries will stand to his word to not commit D7 before some of these major flaws are removed. There is so much more, but I also believe once it is started, It might even become fun to make things intuitively understandable. It is not that hard.

Posted on by Stijn Debrouwere (not verified).

One of the problems with the current focus on usability by drupalistas, though, is that we all too often assume that there will be solutions that will work for everyone. Take the node entry forms: there have been a lot of proposals to improve these over at the Usability drupal group (and some very good ones, too) but they are all focused on creating a one-size-fits-all solution - whereas the reality is that e.g. someone who uses Drupal to manage projects or as a CRM may find usability improvements geared towards article entry to be counterproductive in his case. We should keep the immense versatility of Drupal in mind when making usability improvements.

Posted on by gaele (not verified).

Well said.

There's no such thing as "the user".

Don't pretend to be the user. Pretend to be the users.

Posted on by Jeff Noyes.

I couldn't agree more. One of my upcoming blog entries will be aligned with this subject - titled "there's no such thing as a silver bullet in use case resolution.

Jeff Noyes
Interaction Designer

Posted on by as if (not verified).

Agreed. This problem is inherent in the development methodology of contributed modules: there is a pressure when developing a module to make it "work for everyone". You want it to be a "one size fits all" / "one-stop shopping" experience, because you don't know what types of sites it may be implemented on, and you don't want to release something that worked well in your original circumstance - only.

So on one side of the fence, you get a lot of clunky and confusing modules which are full-featured APIs, but require hacking skills to use. Then on the other side of the fence you get modules written for relative newbies, which do such basic things they really shouldn't even be modules at all.

Unfortunately, I don't see a solution that doesn't sacrifice the whole spirit of OpenSource, and that would be throwing out the baby with the bathwater. The *best* implementations of OpenSource systems will always be the sites that take what they need from the generic functions of the released code, and then step *away* from the rest. You live, and eventually, you learn.

Posted on by roy (not verified).

Nice post, a very important point. Also, are you coming to Szeged? You should.