hmm, i think there has been a bit of confusion about the technics and  
techniques in regards to dissemination versus aggregation.  Most of  
the ideas mentioned were more about aggregation of information than  
dissemination.   These are two significantly different things.   
Dissemination best practices center around systems like http:// 
arxiv.org, http://repec.org/, http://eprints.rclis.org/, and http://dlist.sir.arizona.edu/ 
    all of which have significant collections some of which address  
the issues of their formation and use.

it seems to me that 'aggregation' is why we have colleges,  
universities and it is their core functionality (contrarily it isn't  
teaching or research) as their library, but not just the library as  
repository, but the library as aggregator of technologies and social  
systems surrounding knowledge and its concentration.   As such, I  
don't think much of the hand-waiving about 'whatever shall we do' is  
necessary or really fruitful.  Design does not have to redesign the  
library, university, or college to function, it just has to learn to  
participate in those social systems, which it likely does anyway.

Let's start with the basics:
"How do you currently use publications (paper and digital)?" paper i  
throw away via recycling, digital I archive.  the only time i read a  
paper on paper is if it is mailed to me or i need to edit it.

- how do you search? scholar.google.edu or the publisher's website

- how do you read a paper article? (first: check author-title  
combination, second: flick through the pages. Third: Check the  
illustrations/diagrams and captions. Fourth: Go to the references:  
scan these for familiar names. Fifth: Go back to begin: read abstract.  
There is a strong preference for structured abstracts. Sixth: make a  
photocopy/print and store it for later. Only if it is really  
interesting: postphone all other activities and start reading it.)

the only time i read an abstract is if it is the only thing  
available.   usually I don't read the article either, I will use  
devonthink pro to find the specific areas of the article that should  
be pertinent to my work, and if they seem like they are then i'll read  
the article.

- how do you store photocopies/prints and pdf-files of articles that  
are relevant to your work? How do you categorize them?
http://pictures.tmttlt.com/main.php?g2_itemId=23248  is what i do.   
this database has 27490 papers, books, essays, etc. in it, which is  
2.2 million unique words and 2.1 billion words total.  It is arranged  
by smart groups, which are words or phrases that interest me, there  
are around 1600 of those, some of which you can see in the  
screenshots.  all the documents came from journals, the web,  
government and ngo reports, etc. etc.