To Do
Forward leaning…
-
This thing in blog format -
Author posting data on right y-axis -
Comments enabled on this todo list -
Smarter regular expressions on entered URLs - Sidebar with recently generated graphs
- Avoid overlapping the graph with the block of tags on the right
-
Date based zoom feature on graphs (begin and ending dates) - More user data in a graph (how many, when, …)
June 1st, 2005 at 11:45 am
[...] lution even when site has been very popular but just in a very few days. You have a great todo list, can we not have comments enabled there so that our suggestion would have a [...]
June 2nd, 2005 at 5:09 am
I’m terribly interested in seeing MY tags, how my use of tags has evolved over time (examples: “1 year ago I was interested in “p2p” and now nomore”, “I used tags a lot over christmas and not during the summer”, …). A picture says more than 100 words so being able to see the *evolution in time of tags used by one user* would be great! So instead of focusing on “the tagged URL”, this additional page will focus on “the human tagger”. That would be extremely useful (at least to me, since I have no idea [until you code the tool
] of how I changed my interestes over time! Did I explain it clearly enough? I’m not very sure …
June 4th, 2005 at 12:12 pm
I suggested some ideas for tag related graphs here that you might be interested in. Instead of showing how things change in time, these graphs give “profiles” of URLs, tags, or users at a point in time, with a focus on answering a question (related to whether the profile is dominated by its “long tail”). The user graphs might be interesting to Paolo, from his comment above. I’d be interested in seeing any of these actually generated, if they even can be…
December 2nd, 2006 at 3:11 pm
I am also interested in seeing the growth or stagnation of certain tags of mine over the years. I have a rough guess as to how many I had 6 months ago, but I know the progress is not exactly linear, so it would be neat to see where I have stagnated and where I have surged ahead.
This could also be aggregated to see how groups of users track along certain subjective taglines-a la google trends, but with much more effective terms and a more dedicated base. Meaning, how long did we go tagging articles about wiretapping as wiretapping until we started to call it spying? Right now we can see that on a specific article, but this isn’t too helpful, because the readership for any given news article generally falls of logarithmically from the peak readership within 24 hours. In other words, we would have to compare thousands of possibly dissimilar articles on one overarching subject with the same rubric in order to get an aggregate change in opinion over time.
HOWEVER, if we could track del.icio.us user tags over time, or track some bulk listing of those tags, we can see where tags correlate on similar articles. We don’t have to make the rubric, it is already constructed for us by the del.icio.us architecture and the userbase.
Thanks!