Defintion: Idea Rat, and Twitter

Quite often we get asked to define what it is we do for a living. This is incredibly difficult. Unlike our friends and family, we don't have an easy job to define like pastry chef, dietitian, or teacher. The only one of our friends in a worse situation than us is the one who is a genetic counselor, but at least there are TV shows about people with similar jobs. To fully explain what we do requires a piece of paper, a pencil, and two cups of coffee.

There is a Dilbert strip that comes close to what we do. In this one, our part is played by Ratbert.

It's a bit of an exaggeration, but not by much. What we do is examine our product, look at suggestions our user base has, or things some of our competitors might have done and figure out ways to make our product better reflect all those visions. Sometimes it's tweaking an existing service. Sometimes it's designing something completely new from the ground up, designing the screens, all that fun stuff.

One of the first things we learned, and that we are still learning, is the interconnectedness of all things in an application, or a web site. Even something as simple as adding a logo to the upper right hand corner of a page can have far ranging consequences: Is there a logo already there? Where does that one move? Does that mean that the rest of the header information moves over also? Do we let the client customize that? If so, how much control do they have, and how much correction do we do on it?  This, again, is oversimplified, but right away it's involved layout, programming, graphic design, and general usability for something that someone at one point said "This should be easy to do..."

There's a little quote we keep in our heads every time we start to work on something. We're sure it comes from some place very profound, but we first heard it on Tiny Toons. To wit:

Nothing easy is ever simple

Just because the concept of an idea of good, or sound, or even perfectly logical does not mean that the design and implementation of the solution is going to be that easy. A great example is when a client wanted control over the visibility of a certain field on a profile screen. It SOUNDS easy: just give the admin an option to make the field visible or not. However, it's not: what happens to existing profiles that have the visibility turned on? Does the visibility that field impact the way the profile reads? Will other information make sense if that fields is not visible? How many records does this impact. This is before we even get to the point of determining what the option will actually do: will it just hide the field? Will the field still be editable by administrators? And then, what will the option look like: Is it a radio button? Will it be on a profile-by-profile basis? After all that has been done, it has to go to a developer who then looks at it and everything behind the scenes that will have to happen, and then lets us know if this is something that can be done as a hotfix, or needs full development, and then how long that might take.

This ties into two things.

1) Our ongoing battle against FAIL as a valid way of making a point. There are a lot of moving parts, and to just look at something and saying FAIL is the equivalent of a kid saying they don't like carrots and then holding their breath. But we've discussed that to death here.

2) Twitter trends. Recently twitter integrated trends into their sidebar. This has already created a number of problems

1. People adding the high ranking trending topics to their spam posts
2. Scammers using it to get access to personal information, albeit in a roundabout way.
3. Most topics when they get to the top ten become polluted by people asking "Do you know what #trendytopic means?", pushing valuable information further down the list.

None of these were much of a problem before, because they weren't featured on the Twitter home page (they were part of search.twitter.com, which was a horse of a different colour). Of course, people have said it should be fixed, and immediately. "Fix it! It's a bag of FAIL."

Great. HOW?

It's not a simple fix. You could build algorithms to monitor the signal-to-noise content, but that will take time and need a lot of tweaking and will probably hurt some very legitimate threads. You could pay some intern to monitor the topics and manually edit their rankings. This is a horrible solution and will lead to messes like Apple's App Store approval inconsistencies. You could also just take it down until you think of a solution, but that's the worst solution: a solution should never take away existing functionality.

The point is that the problem is simple, but the solution is not. The best course of action might be to just inform users that Twitter makes no guarantee about the content or validity of any trending topic, and that people are using them at their own risk. We're certain that something like that exists in the terms of service, but it might need to be posted in some section of the site, like the one that currently is used to promote Twitter related services.

Who knows, maybe it is as simple as flicking a switch. But chances are it's not, and just "fixing it" quickly might create more problems. It's very easy to say something is broken and discuss how you'd like it work, but just saying 'it has to be fixed' or 'FAIL' and stamping your feet helps no one.  

But don't ask us. We're just idea rats.