July 242006 Infrastructure Discussion

comments:

1) Embrace open-source. Philosophically, open-source is a mission-fit with educators, and platform-dependent systems like Microsoft servers, MS-Access, and ASP web coding is antithetical to AECT’s orientation. There are also operational problems with combined systems as we have now. This means eventually moving AECT’s main web site from its current Windows server to a Unix-based server; in practice this would likely mean hosting AECT on the same server as the aect-members.org site. Upside - Downside - change

2) Move to mySQL. The main membership database is in MS-Access and needs to be converted to a database that was actually designed for the Web. This also enables a certain synergy between the membership database and the database that supports the MovableType publishing platform. At least one program (called, interestingly enough, Access2mySQL) makes the database conversion a simple chore. The scripts, then, need to be changed to accommodate the new database, and those changes would best be accomplished while we…

3) Modify our log-in procedures. Last name/membership number is simply a bad system. We need to allow members to choose their own password (a membership number is not a password and is not secure). I have a plan to accomplish both the database conversion and the login problem at the same time and can explain the procedure in another conversation.

4) Create member info pages that enhance the notion of a community of practice. Each member can have a page that presents information about that member which is actually informative: for example, if I went to Trey’s member page, I should be able to find his email (unless he blocked it) but should also be able to find abstracts for all of his conference presentations, links to any ETR&D or TechTrends articles he’s written, and links to his latest weblog entries.

5) Realize the value of our conference database (dating back to Atlanta 2001). These abstracts are valuable and demonstrate who we are and what we do. Right now, it's all invisible.

6) Support member weblogs. We are professionals, we are educators, we are writers. We should be able to aggregate members' weblogs to demonstrate our expertise in our field, and we should be able to offer weblogs to members who don't already have them. As a matter of fact, our Drupal installation makes member weblogs easy to create and support. We have it, we just aren't using it. The only reason we wouldn’t support member weblogs is an irrational fear of our members’ comments about AECT, once stated as "…sometimes, freedom of speech isn’t a good thing." In a member-driven association, in a country with a Bill of Rights, this is indefensible.

6) Move the mail server. Our aect.org mail server is a Windows NT server in AIT’s basement (Windows NT hasn’t been supported by Microsoft for years now). It is separate from our web server, which is unusual to say the least. The reason we haven’t moved it yet is because our listservs are linked to the membership database (still in MS-Access) and a Rube Goldberg-like, manually-managed script is needed to keep it updated. Inefficient, to say the least, and horribly limiting. We haven’t made this change because of a confounded belief that ListServs are member benefits, that “member benefit” means we need to limit ListServ activity to current members only and that our hands are tied. So, we’re tied to an unsupported operating system on a remote server, with the membership database kluged to the mail server – all because we havent' explored open-source ListServ options.

7) Take TechTrends to the web. This may be impossible given Springers’ taking control of the bimonthly, but hear me out: TechTrends, as a freely-available online magazine, exposes it to thousands of new readers. New readers who can then see the value of membership in this professional association. We may hold back feature articles as premium content, but most of the articles published each issue are short items that have more potential value in promoting AECT than restricted value in a limited, “member-benfit” model.

There are more things we can and should do, but in general I think we’re in need of a new philosophical underpinning:

Also refer to http://dragonfly.unco.edu/mwiki/index.php/Electronic_Services_Policy