Paper Reading #24: Have a say over what you see: evaluating interactive compression techniques

Title:
Have a say over what you see: evaluating interactive compression techniques

Comments:
Cindy Skach
Luke Roberts

Reference:
Tucker, S. and Whittaker, S.  Have a say over what you see: evaluating interactive compression techniques.  IUI '09.  http://delivery.acm.org/10.1145/1510000/1502659/p37-tucker.pdf?key1=1502659&key2=1951472031&coll=DL&dl=ACM&ip=165.91.4.169&CFID=16209705&CFTOKEN=47524431

Summary:
This paper goes over the idea of interactive compression (IC).  This is the process seen in Figure 1 of reducing the amount of content.  This is able to be done at different levels of compression which is different than most summary software, hence one can choose the level of compression wanted.  IC controls the amount of information a user sees.  It is split into two types in this paper: Excision - removing unimportant words or transcripts from a sentence and highlighting - keeping all the text there but highlighting important words. 

The researches did an exploratory survey in order to test the design principles in which they were interested.  The survey basically featured ways in which people skim a document and asked whether each way was helpful.  The results showed that users had a preference for word over utterance techniques for IC and that highlighting was also liked. 

The second study tested these findings in an implementation backed by the control of a regular full document.  It was found that the IC techniques made users skip over more important information more easily, but also allowed them to view documents faster.  Users also preferred IC to unmodified text.  Excision allowed faster reading than highlighting, but highlighting allowed the user to skip over less information. 


Discussion:
This paper was interesting.  I am definitely a skimmer, and I think if I had this software to skim the papers that we are reading I would be in good shape and would appreciate it so much!  Where is my copy?  But I like that they did two surveys and based their implementation on an initial survey rather than just going out and thinking of a design themselves.

Future work could be to the effect of summarizing.  They were railing on summaries, however if they claim their way is so much better, they must have some insight in to why summaries are weak and revise it.