This includes things like bookmarklets, toolbars, greasemonkey scripts, and so on. For a nice discussion which puts some of these in context see Ross Singer's Access 2005 presentation. He talks there about the 'sloppy underbelly of the web'. I used intrastructure for a while to talk about this type of thing, but as nobody else does I gave up ;-)
Recently, I have been wondering about the impact of 'distance from the default' on take up. One might surmise that the more a user has to do, or in other words the more they have to adapt their default settings, the lower the uptake will be, other things being equal. And most of these tools do require installation of some sort. Even small effort dampens use. Greasemonkey allows you to do very nice things. What proportion of your user population will you reach though? Well, IE is still the default browser for many. So you immediately have a narrowing to those who have installed Firefox. And then a further narrowing to those who have installed Greasemonkey. And then a further narrowing to those who have installed a particular script. You are getting down, unfortunately, to small numbers at this stage. You can tell a similar story with toolbars, widgets, extensions, dashboards, and so on, with different degrees of narrowing as the distance from the default changes.
And the issue is compounded by the variety of approaches. We do not have a generally adopted, simple 'composition' framework in the browser which narrows the 'distance from the default'.
The ability to very simply compose external services in your browser/desktop/devise seems to me to be so valuable that it has to happen ;-) Something will come along that unifies our work environment more. We know some of the candidates, but maybe something new will appear.
Incidentally, Kudos to Meredith Farkas for setting up the Library Success Wiki. A good and simple idea which is valuable and deserves to succeed.
Link to the browser extension page via Science Library Pad.