thesis: analysis of visual study #3 A…

I feel pretty good about this study for now. I think it's a solid idea, and that there are a lot of places for me to go with it. It certainly embraces and highlights the idea of triage.

From Martha's feedback (and reaction) I can see the the whole shifting/sliding/filmstrip notion won't work. Which is fine. I think the solutions she and I talked about (making the left-hand space a history line, and the right-hand space a place for saved material) will work well. Or, at least, will be good things to explore.

I think what's working now is the pared down aesthetic. I might play with this a bit more, but I like the grays, the stillness to it all. Maybe a texture for that background gray would be nice.

I like keeping the type large and simple, with just some basic typographic weight/style changes, and one (maybe two tops) color changes for emphasis. I think using the center panel for main content is good, and the surrounding real estate needs to be kept minimal, only allowing for pertinent material and maneuvering through the system. Everything about this system needs to be about isolation and leading someone by the hand.

The language is going to be very important as well. Exactly the way in which the questions are asked, and the way things are worded will be key. I need to establish a consistent tone for the wording (to match and enhance that of the visuals).

Some things Martha mentioned:

• Should the recommended queries be listed with the keywords?
• When you need to come back to the system, how would that work? Could there be a search mode and a saved mode?
• You need to think about the emotional components for this. Especially since I'm looking at decision making and such. My personas and scenarios are (as Martha said) "fraught" with emotion. How to account for that? Or exploit that?

From here, I think the next step will be to story-board how the rest of the system works/can be demonstrated. Then from there, I need to really scrutinize the story-board, and then produce the visuals.

Finally, I also need to be cognizant of the factor of time. My three personas are very busy people. A system that slows you down to really help you move through information in such a methodical way can, in the end, be a time saver—but the time investment up front may feel like too much. So the challenge will be to balance that, to move through these initial steps quickly, and to make the history and saved material functions really useful and intuitive.

thesis: analysis of visual study #2 B…

So after much thinking, and some sketching, I revamped and re-presented Visual Study #2 to Martha. I think this a much better direction. It's still very loose in my mind (and in form), but I believe it will function in a much more interesting way.

Here, the system is taking over more directly, however, the user is still able to manipulate and move through information. I had envisioned this encompassing many different visualizations (like Viewzi), but after talking with Martha, that feels superfluous. How can I manage this system, so that it can intuitively function on one plane? I have so many variables I can work with. What I'm struggling with is balancing the practical with the triage.

Really, I think this study is now all about the visualization of the information more than anything else. Perhaps the difference between this and my other two studies is that this system will allow for more personalization than the others, more movement through the system's results. I think this will be all about the system suggesting organization and categories, and the way in which the display of the information helps to triage.

A thought just now....what about bar codes as a metaphor? Thinking here about the way each element means something, and goes together to create a unique tag for an item/object. Could I chunk information in this way? How could that visually work? I'm starting to do that with this system, with the spine, with the color, the hierarchy. But what else?

I really need to take a close look at Martha's comments about this system (below), and need to definitely keep going:

• Why are these alphabetical? Couldn't it just sort by ranking, or by amount of information?
• How does the saved function work (if there is one)?
• How does tagging work?
• Should think about universal categories: advocacy/opinion, explanatory, practical, commerce, news could change into latest/breaking info.

Next, I need to run this through a full scenario, and possibly try it out with one of my other personas. Or, think about different moments within the system that the three personas might exemplify through use?

thesis: analysis of visual study #2 A…

It is obvious I had to start somewhere, and I certainly consider this study just that: a start. Just a beginning. Pulling back and looking at what I've done, I can see that I really got caught up in the conventions already found on the internet. Search results will be sorted into a gridded list, they will be clickable, they will all look the same. Users can implement tags, which will function in a very predictable way. Users can filter results using a left-hand column list, again in a very predictable way.

It's funny, I poured many hours into this first execution, thinking these were sound, practical, interesting ideas. However, when I look at them now, I can see that they expand nothing, reveal nothing, and triage nothing. I have created a system with many functions and features, but it doesn't establish an atmosphere of triage at all.

Denise mentioned that triage is really about getting at the essence of the problem, introducing a user into this whole new world that they don't really understand. That it is about the system taking over (and not about allowing a user to still make all their own decisions). This idea of the system completely taking over makes me very uncomfortable (which is why I haven't attempted it yet). I have been very caught up in trying to create something that will do a lot of the "work" for the user, but will still allow for complete flexibility. However, many of the systems out there already do that. Google works because it is simple and flexible. However, there are things about it that don't work because it has been kept simple and flexible. The engine has a certain sense of universality, for good and bad.

I think I need to get out of the mindset of trying to "fix" Google. It is what it is, and many people on their staff work hard to look at the system and make it better, and make it work reasonably well for as many people as possible. Google does what it sets out to do. This is why it's still the most used search engine, and why it has become so ubiquitous in so many ways.

What I think I'm attempting to do, is to highlight other ways that people might search, using other kinds of tools....which really, in the end, need to function in extremely different ways. Looking at Viewzi has been very helpful in this regard. This is a system that someone built which aggregates information FROM Google and others, and then extrapolates that data, that information, in numerous and novel ways. And the power that comes from that is seeing the data sets in these unique shapes and forms.

So, to move on, I need to let go of some of this sense of practicality and resonable-ness, and look at the problem a bit more obliquely (maybe even obtusely). How might a system truly, TRULY triage for a user? What would that look like? And what would the experience be?