X-Git-Url: https://sipb.mit.edu/gitweb.cgi/wiki.git/blobdiff_plain/0e6f037754eb2f9caae159006e9e5867545401f1..ed04867ea0947d6f6e10cae8b8727eab72900cb4:/doc/ProjectIdeas

diff --git a/doc/ProjectIdeas b/doc/ProjectIdeas
index d9ee187..23339ba 100644
--- a/doc/ProjectIdeas
+++ b/doc/ProjectIdeas
@@ -10,8 +10,13 @@ changing.
 
 If you have a project and want people to help, go ahead and add it.
 
- * SIPB website (sipb.mit.edu)
-  * (More information here)
+ * SIPB website (sipb.mit.edu) -- talk to `price`
+  * We're working on making the website a wiki, which will hopefully
+    help it get and stay up to date in the long term.  The core software
+    should be up before the hackathon.
+  * write stuff about SIPB
+  * convert material from the old sites and this doc wiki
+  * help style it to look good
 
  * mailman.mit.edu hacking -- Talk to `kcr` or `nelhage`
   * There are two ways to create mailing lists at MIT -- Moira, and
@@ -24,15 +29,21 @@ If you have a project and want people to help, go ahead and add it.
   * There are a number of ways this could be implemented, including
     moira <-> Mailman sync, direct Moira integration in Mailman, or
     something in between or different.
-  * MIT runs Mailman 2, but Mailman 3 has been released as alpha and
-    may be worth investigating. It's likely to be far easier to get
-    significant patches back upstream for mailman 3, if we go that
-    route.
+  * MIT runs Mailman 2, but Mailman 3
+    [http://mail.python.org/pipermail/mailman-announce/2009-January/000126.html has been released]
+    as alpha and may be worth investigating.  The author is explicitly
+    open to major changes for mailman 3, so we could push some of our work upstream.
 
  * Dodona -- Talk to `jhamrick`
   * `jhamrick` is working on a Zephyr bot designed to answer technical
     questions using a natural-language interface, and is looking for
     people to help hack on it.
+  * create a web interface using pyjamas (or something else?) for people who don't use zephyr
+  * figure out the best way to store and retrieve technical data.  Currently
+    Dodona pulls from a text file and parses that information into a dictionary.
+  * improve Dodona's UI
+  * IF (and only if) we finish all of the above, start the NLP! (the fun part!)bbb
+
 
  * `scripts.mit.edu` -- Talk to `geofft` or `quentin`
   * Find some software we don't have an autoinstaller for, and write an