June 2009 Archives

I am currently leading a team at my workplace that's evaluating our use of the project management software Quickbase. Quickbase is essentially an online database that manages data points such as (in our case) people, courses, programs and tasks. The tool was originally intended to facilitate collaboration and project management, enabling the use of templates to generate task lists and notifications for courses running on a repeat basis. It was also intended as a platform for designers to manage the development of new courses; pulling all their team members and resources together on one timeline. Managers have a stake too; using Quickbase they are able to easily assess at any time the status of courses or programs.

Our Quickbase use has fizzled a bit (or a lot).

One reason for this is likely due to the interface: It's very database-y:
QB.png
Our workplace consists of instructional designers and assistants who may not be accustomed to a database interface, and honestly are not expected to be power database users as part of their skill set.

My charge, along with the team joining me in this journey, is to find what might actually work to achieve our ends. We know that the software is very good at managing data, tables and relationships. We also know that our users are drawing a blank when attempting to use it. So the question is: how do we build an interface that our users instinctively know how to use and serves the purpose of keeping team members honest and on track? There is yet another layer of complication here. The team interacts on a nearly daily basis with various data housing systems outside of Quickbase - systems that store more detailed course information, manage time tracking, store course multimedia, etc. Here in fact is a snapshot of the whiteboard from our last meeting where we were discussing all these systems:
QBwhiteboard.jpg
You can see how tangled this problem is. Many of the systems are out of our control and will probably still have to be interfaced with manually, but for the in-house solutions, there is great potential for integration, saving time and duplication of effort for everyone who uses them.

Luckily, we have Jeanette Condo on our team, a programmer with database and API experience. (yay for us!) With her help, we may come up with a win-win solution that puts Quickbase in the background while fully taking advantage of its powerful database capabilities.

Our next team meeting is next Monday. We are still in the discussion and exploration stage and have not begun any nitty-gritty development or testing yet, so the question I think I want to ask is, if we were to design a system for managing our work from the ground up, without consideration for existing systems, what would it look like?
Wayfinding the Semantic Web

AD-54 Penn State policy will not be updated any time soon to require semantic metadata

instant history of the Web
  • web 1.0 - series of linked documents
  • web 2.0 - communications
  • web 3.0 - semantic web
purpose of semantic web - add meaning to Web documents
burners-lee - semantic web makes it possible for people and machines to understand web content

We all come into contact w/the semantic Web every day - Google is a good example when it comes up w/intelligent search results.

An Argument For Semantics

words can have different meanings in different contexts and depending on the bias of the observer - machines don't have bias and can't read context well

Question: can we pull non-sensitive enterprise data from a central source?

Semantics: meaning of words or symbols, become really useful when shared betw. individuals and w/in community

Keys - in the absence of meaningful identifiers, we invent a (usually numeric) key to unique data

Demonstration - Firefox add-on Operator. find and save events. Will have to work with it to see how it really works or is useful.

Code demonstration - possible implementations:
  • microformats - class attributes in CSS that are standard, can be parsed out, eg. class="time"
  • metadata
  • xhtml attributes - 
  • RDF - THE language for presenting semantic info on the Web - it is essentially embedded in the html. WILL NOT VALIDATE. Creative Commons licensing is an example.
  • RDFa - embedded as attributes
  • Content management systems do a good job of handling semantic web aspects. They make it easy for developers but don't "make" you do it. Recommend a person w/the role "final release editor", someone w/a librarian mind.

Brian Panulla - get rid of the words semantic web! Moving from metadata to linked data
tagging is NOT the semantic web - too loose and open. can lead to good results w/enough tagging, but not good enough.

higher ed = walled gardens. leads to redundant data. relational databases are about structure and not data. convention may allow us to integrate data across systems, e.g. usps state code, campus 2-letter code. problem - ad hoc identifiers and multiple copies of same data in existence, e.g. college/campus name changes

New W3C language - OWL: Web Ontology Language

Semantic Web means essentially:
  • further the separation of content, presentation, behavior, logic
  • breaking down data silos
______________
Augmented reality
Current AR technology - GE smart grid
Mini automotive ad
topps 3d baseball cards
bmw augmented reality - example on youtube shows AR assisted car repair!!
AR island on Second Life - may be fun to explore

______________

About this Archive

This page is an archive of entries from June 2009 listed from newest to oldest.

April 2009 is the previous archive.

July 2009 is the next archive.

Find recent content on the main index or look in the archives to find all content.