UMBC ebiquity
2007 May

Archive for May, 2007

Animated character to talk at UMBC Computer Mania Day

May 2nd, 2007, by Tim Finin, posted in Uncategorized

Jennifer WebbThe UMBC Center for Women and Information Technology will host Computer Mania Day on Saturday May 5th to provide a half day of technology-related activities geared for 6th – 8th grade girls and their parents. The invited speaker will be Jennifer Webb, described as

a digitally animated 8th grader at Computer Mania Middle School. Although she does not always receive straight A’s, Jennifer really enjoys school, especially her algebra class.

While boys are welcome, the focus is on girls because of their continuing under-participation in information technology classes and careers. This program is designed to provide a broad-based introduction to information technology applications to different careers.

US Army adds new restrictions to military bloggers

May 2nd, 2007, by Tim Finin, posted in Uncategorized

The war in Iraq is the first one that has been blogged about by the troops, whether on individual blogs sites or via aggregation sites like milblogging.com. Reading blogs by active military members stationed in Iraq and Afghanistan has been both informative and interesting and offers an unfiltered perspective on critically important issues.

Wired news reports (Army Squeezes Soldier Blogs, Maybe to Death) that the US Army has ordered soldiers to stop blogging or sending personal email messages without first clearing the content with superiors.

The new rules obtained by Wired News require a commander be consulted before every blog update. … Army Regulation 530–1: Operations Security (OPSEC) restricts more than just blogs, however. Previous editions of the rules asked Army personnel to “consult with their immediate supervisor” before posting a document “that might contain sensitive and/or critical information in a public forum.” The new version, in contrast, requires “an OPSEC review prior to publishing” anything — from “web log (blog) postings” to comments on internet message boards, from resumes to letters home.

Wired reports that the directive applies to civilians working for the military, army contractors and even soldiers’ families as well as to active duty troops.

Installing telegraphCQ

May 2nd, 2007, by Tim Finin, posted in Uncategorized

These are the steps I used to install and troubleshoot my tcq installation. For official documentation visit –http://telegraph.cs.berkeley.edu/  Installing Telegraph CQ 2.0 on kubuntu linux. Steps:

1. Create a new user for usage with telegraph – Command = useradd -d /home/tcquser -m tcquser

2. Change password to whatever you want.

3. Go to http://telegraph.cs.berkeley.edu/telegraphcq/v2.0 to get TelegraphCQ version 2.0. My observation is that there are some problems with version 2.1 for unarchived streams. I could only get archived streams to work on it, so I couldn’t query it realtime. I had to query the archived stream. I found it on the mailing list that people worked with 2.0 and could use it for unarchived streams.

4. su tcquser

5. Create two directories – /home/tcquser/tcq-tools and /home/tcquser/installs/telegraph

6. To install mm- unpack the .tar.gz file provided at – http://www.ossp.org/pkg/lib/mm/ into /home/tcquser/setups/mm

7. Go to /home/tcquser/setups/mm and run –

a. ./configure –prefix=/home/tcquser/tcq-tools b. make c. make test d. make install8. To install judy- remember to pickup the patched version for tcq from http://telegraph.cs.berkeley.edu/dev/judy-tcq-0.0.1.tar.gz unpack it into /home/tcquser/setups/judy 9. Go to /home/tcquser/setups/judy and run –

a. ./configure –prefix=/home/tcquser/tcq-tools b. make c. make check d. make install10. To install telegraphcq –

a. export $PGRSHOME=/home/tcquser/installs/telegraph b. Unpack the telegraph package into /home/tcquser/installs/telegraph/TelegraphCQ-2.0 c. Go to /home/tcquser/installs/telegraph/TelegraphCQ-2.0 d. ./configure-tcq e. make f. make installCommonly encountered errors –

1. During configure-tcq – readline library not found – install readline (for ubuntu/ kubuntu install the libreadline and libreadline-dev from the package manager) 2. During make of TelegraphCQ, I got some error in the code of the form “invalid lvalue”. Apparently there was some code of the form –

Class1 * ptr1; ptr1 = (Class1 *) ptr2 -> ptr3 = NULL;I had g++ version 4.1 on my machine and it did not recognize the “(Class1 *) ptr2 -> ptr3″ part as a valid lvalue. I broke the code into 2 lines.

ptr2 -> ptr3 = NULL; ptr1 = (Class1 *) ptr2 -> ptr3;3. While running ./createdb you might get errors such as libpq.so.3 not found. Generally there are newer versions of these files in /usr/lib. Create a symlink as per the required version e.g.

sudo ln -s /usr/lib/libpq.so.4 (existing file) /usr/lib/libpq.so.3 (reqd file)4. Sometimes ./createdb or ./psql fails saying –

psql: could not connect to server: No such file or directory Is the server running locally and accepting connections on Unix domain socket “/var/run/postgresql/.s.PGSQL.5432″? This means either the server is not running or client is looking for the socket file in wrong place. This error message show that it’s looking for it in /var/run/postgresql whereas on my machine it was actually present in /tmp. To overcome this I created a symlink.ln -s /tmp /var/run/postgresql

Map of online communities

May 2nd, 2007, by Tim Finin, posted in Uncategorized

XKCD has an amusing “map of online communities and related points of interest” in which “geographic area represents estimated size of membership”. The map cautions that it is not a complete survey and should not be used for navigation.

map of online communities and related points of interest

Semantic Web Markup for Blogs ?!?

May 1st, 2007, by Tim Finin, posted in Uncategorized

I read The Definitive Guide to Semantic Web Markup for Blogs expecting it to be about how to make effective use of RDF in blogs to add explicit semantic information. I was suprised to find it offered some pragmatic advice about how to use the most basic of HTML tags (e.g., TITLE and H1, H2, H3, ..) so as not to confuse search engines.

Unbelievably, nearly every WordPress, MovableType, or TypePad theme that I’ve come across in the past year fails a simple test for truly semantic (and Google-recommended) XHTML markup. … Your goal is to describe each page to the search engines through the use of hierarchical XHTML tags (

While the post contains lots of useful advice, it’s clearly not about big-S-big-W Semantic Web or even (IMHO) lowercase semantic web. If anything, its about another SW — SEO Web.

There are several plugins for WordPress and other blog frameworks that add RDF data to your blog, such as SIOC Exporter for WordPress as well has various plugins to add SKOS and FOAF data. Many people add additional RDF information in their RSS 1.0 feeds. Blogs make heavy use of folksonomies through their tagging and category systems and some have information embedded in microformats and RDFa. Finally, a number of blogs, e.g., FieldMarking, are experimenting with the notion of “data blogging” — using a blog as a medium to publish a stream of data.

If we generalize a bit to other forms of social media, like Wikis or Flickr, there’s even more going on that clearly falls in the upper or lower case views of the semantic web — Semantic MediaWiki and machine tags spring to mind.

You are currently browsing the UMBC ebiquity weblog archives for May, 2007.

  Home | Archive | Login | Feed