Hello dear reader.
I'm going to be taking a leave of absence from Sun soon so I can study Chinese full time (I'm very grateful to have this opportunity). So, this is likely to be my last entry in this blog for a while. Indeed, the preparations for my trip are why this week's entry are as late as it is.
One thing I've been doing is setting up an account on gmail (free pop access to email seems useful). I've not really used gmail much before, and noted something interesting about it today.
The layout here looks like an ordinary email program. Each row seems to be a message, with columns for sender, subject and date.
The interesting thing is that this isn't really quite what is being presented. Each row is actually a (hopefully related) group of messages in one conversation created by a set of people. The "From" column lists the people who have contributed to the conversation in the order they've responded (bold indicates you haven't read that entry, and a double dot indicates when there are more entries than can be shown in the column width)
There's nothing inherently wrong with this. While it may be a little unexpected at first ("What does it mean for a bunch of people to have sent a message?"), it is close enough to the "traditional" model that it is not hard to learn what is going on.
As I looked at this, though, I found myself wondering if there might be a more fruitful way of displaying this information. The "From" column (really, the "active participants" column) is trying to show a lot of information in a really small space. My first thought here was that if you could associate a picture with each person, then you could actually get a longer list of people in the conversations here. Rather than two or three names in the column width, you could get ten or more.
In theory, this seems like an improvement. In practice, I'm not sure the pictures would be large enough to let you identify the people:
Yet, when I look at this, I still can't help but feel like something is wrong with this design. Each row is a whole conversation, not an individual message. The prominence of the people involved is interesting and unexpected. It would seem to me that the system could quickly begin to notice that a particular set of N people are routinely involved in conversations with one another. Perhaps it could begin to deduce some social clustering. At that point, it might be more fruitful for the software to present messages in terms of the social groups first, and the individual conversations setcond. that way, one could effortlesly look at all the conversations with one's project group at work, first (and ignore messages from outside the group). Maybe this would look something like this:
Or, perhaps it would be better to show which group members participated in each thread:
Or you could imagine many other possible designs for something that is social-group oriented as much as it is thread-oriented.
Of course, at this point in time, email serves many purposes for people. It is a to-do-list, a paper trail, a file exchange medium, a reminder mechanism, a communication channel, and so on. something like this wouldn't fit all the uses of email well. So, maybe that means this is a bad idea. Or perhaps the fact that those usages don't fit into this design well actually suggests more design innovation that could be done.
If I have a central point to this entry, it might be that sometimes when your interface design seems a little strange, it may actually reflect an opportunity to do something novel. (and I suppose a variation on that idea would be : if a design is bad, it might be good not to dismiss it as a bad design, but investigate whether it suggests other possibilities that you hadn't considered before.