A couple of things I've (Lynne C) been thinking about.

1. A few years ago a student did a project on duplication of letters in CMC (e.g. sooooo). It occurred to me that this kind of thing should be used less on Twitter, where every character counts. Is that the case? Do Twitter users still use duplication when they use other forms of CMC?
2. I'm not quite sure what might make a doable project in this area, but I'm intrigued by the increased use of writing without spaces. We had to start getting used to it when we had web sites named with phrases or sentences without their spaces (e.g. www.doyoulovefood.co.uk) and it can lead to some hilariously unfortunate ambiguities (kidsexchange, for example). But now people are doing this in sometimes very long twitter hashtags. Before the days of Windows people using computers used to use the underscore _ to replace spaces in, for example, filenames. Why don’t Twitter users do something like that to avoid ambiguity? (There are two separate issues here – the restricted number of characters, which applies to spaces and underscores equally, but also the requirement that a hashtag has no spaces.) Or another way to look at it would be to ask whether the space is overrated – how often do we actually have trouble reading phrases without spaces?