From MWCSWiki

Contents 1 Project Litstream 2 Wiki 3 Action items 4 Notes 5 References 5.1 Systems we're using 5.2 Competitors to look at 6 Dev Stuff 6.1 What I had to do to get sproutcore and papercube running on ubuntu lucid:

Project Litstream

Mission (working draft): to enable knowledgeable users to easily add semantics to the information in a particular domain (in this case, academic publishing) such that novices to that domain can take advantage of them to accelerate and improve their understanding.

This project is staffed by Lucy Bain, Aaron Clemmer, Stephen Davies, and Chris Donaher. (And perhaps, Jesse Hatfield.)

Wiki

Personal Knowledge Base

Action items

For September 8, at 4 in the Underground

• Register litstream.org domain name (SD)
• Create project Wiki (SD)
• Explore: how to get the data? (CD)
  • Can we get it from Google, UMW library, individual databases, CiteSeer, screen scraping (last resort)?
• Graphical toolkits: which one should we use? (AC)
• Explore: Lit search on lit search tools (LB)
• Explore: What do other tagging systems do, exactly? (delicious, Stack Overflow, etc) How do these work, and what can we learn from them? (LB)

For next meeting (possibly September 10, at 3)

• Does any tool use use keywords in general? (SD)
• Talk to the guy from PaperCube (SD)
• Talk to sysAdmin for rosemary about removing database size limit (SD)
• Does CiteSeerX have the keywords packaged in with the data available? (AC)
  • No → do a lit search text mine from abstracts
• Branch PaperCube code base (AC)
• Send everyone a hg cheat sheet (AC)
• Play with PaperCube to make SURE you understand EVERYTHING it can do, think about how to expand the given interface, read master's thesis (LB)
• read up on the hg stuff from Aaron(All)
• Think about the interface - how does the user actually work on the site? What do we want in the future (All)
• look into the code, get familiar with it (All, especially AC and CD)

For September 13, 3-5

• Tagging systems, how do they work? (LB)

• Look up the "competitors" (see "references" section on this wiki) and make notes (inline on the wiki is best) about what they can do, so we know how they compare to us (??)

Notes

These excellent notes are provided by Lucy.

• 6-Sep-2010
• 8-Sep-2010
• 10-Sep-2010
• 13-Sep-2010
• Quick notes from 14-Sep-2010
• 14-Sep-2010
• 17-Sep-2010
• 20-Sep-2010
• 22-Sep-2010
• 24-Sep-2010
• 28-Sep-2010
• 30-Sep-2010
• 04-Oct-2010
• 06-Oct-2010
• 08-Oct-2010
• 13-Oct-2010
• 15-Oct-2010
• 20-Oct-2010
• 25-Oct-2010
• 29-Oct-2010
• 8-Dec-2010
• 13-Jan-2011
• 21-Jan-2011
• 24-Jan-2011
• 28-Jan-2011 (I think)
• 4-Feb-2011
• 11-Feb-2011

References

Systems we're using

• [PaperCube]
• [CiteSeer]

Competitors to look at

• [Google Scholar]
• [SCIrus]
• [getCITED]
• [EBSCO]
• [EI Compendex]
• [SciTopics]
• [citeulike]

Dev Stuff

What I had to do to get sproutcore and papercube running on ubuntu lucid:

# we need ruby
sudo apt-get install ruby

# we need gems
sudo apt-get install rubygems1.8

#debian systems pkg dev tools separately
sudo apt-get install ruby1.8-dev

#we need rubygems > 1.3.5
sudo gem install rubygems-update
cd /var/lib/gems/1.8/bin/
sudo ./update_rubygems


# we need deps!
sudo apt-get install libopenssl-ruby1.8
sudo gem install json

#we need sproutcore!
sudo gem install sproutcore --version=0.9.23

# I didn't have java, and we need java for YUI compressor apparently.
sudo apt-get install openjdk-6-jre-headless

#so let's finally build the thing.
cd ~/litstream
sc-build


#happiness.
ls tmp/build/
  nodegraph  papercube  prototype  sproutcore
sc-server

Go to: http://localhost:4020/papercube