. I made this using Joe Gregorio's excellent sparkline service, by the way. (Here is the URL for that sparkline; handily, it encapsulates the raw data.) It would be great to do this easily with del.icio.us history, for a range of time scales.
The del.icio.us URL info page, which records historical interest in a particular URL, has morphed. According to the del.icio.us blog:
We've prettied up the URL info page so it is easier to get at the conversation around the links.For me this change has added value in one way, but subtracted value in another. Here, for example, is the del.icio.us history for the Live Clipboard example page. I can't recall exactly how the del.icio.us history used to look, but I'm certain it reported dates for each posting. I relied on those dates to sense the ebb and flow of interest in the ideas represented by landmark URLs. At one point, for example, I used them to track citations of the ACLU's digital identity nightmare.
Date information is still available on a per-month basis, so memes that have been floating around for a while, like the ACLU screencast, can still be tracked on that basis. But in the case of new memes, like Live Clipboard, I cannot now see or chart the daily ebb and flow.
Donning my wishing cap, I'd like both capabilities brought back in a more robust way. On the history page, I'd like to be able to see a chart that reports the count of postings by day. Through the API, I'd like to be able to retrieve that data and work with it.
Here, for example, is a sparkline showing monthly citations of LibraryLookup in del.icio.us since Jan 03: . I made this using Joe Gregorio's excellent sparkline service, by the way. (Here is the URL for that sparkline; handily, it encapsulates the raw data.) It would be great to do this easily with del.icio.us history, for a range of time scales.
While I'm wishing, I'd like to apply this kind of analysis to sets of related URLs. Live Clipboard, for example, is represented in one way here (the examples) and in another way here (the screencasts). Del.icio.us knows these only as c175db19ba3a085aced11051e9f40b7b and 26d9143562a51010791df0a5a09dadf8, two identifiers that share a tag. I know they're close siblings, but I can't tell del.icio.us that.
In terms of data visualization, the sole innovation to emerge from social bookmarking and tagging has been the tag cloud that correlates font size with frequency of use. Tag clouds are cool, but they're just one of the interesting stories these services ought to be telling us.
Former URL: http://weblog.infoworld.com/udell/2006/03/14.html#a1405