The web site of the University of Waterloo (UW) Computer Science Club (CSC) has existed in various forms since at least the Fall term of 1994. It was initially hosted on calum, a SPARCserver 10/40. At the time, this was the only computer the club owned. It was eventually moved to aquata (with an alias of www), which was a 486 donated by two club members for this purpose. This computer has since been upgraded, in both hardware and software. The web site is divided into many sections, each corresponding to a directory tree in the local filesystem. The home page has a list of direct links to each of these sections, followed by a more descriptive introduction to the site contents. As the WWW admin for the CSC for several terms, it has been my responsibility to see that this web site operates as intended, and fulfills its purpose. The site has two primary goals. The first is to disseminate information about the club to its members, and the rest of the world. The second is to host web pages for club members, and other UW clubs.
There are numerous places around the site where information is disseminated. The most obvious is the Announcements and Upcoming Events section of the home page. As the name implies, any announcements to the club (and the world) at large are put here, as well as notices of upcoming club-related events. The major problem with this section is technical. Upcoming events are announced in a somewhat automated way, and this would be an obvious place for automation; yet, there is no mechanism in place to insert notices into this section of the home page, and it must be done entirely by hand. Sometimes there is little time to perform this task, and sometimes it is simply forgotten. Other less transient information is available in several of the other sections.
The Documents section contains various "Official" documents and information about the CSC, such as the constitution, and the minutes of meetings. This section exists mostly for reference and archival purposes, and it performs that function quite well. The About section contains some general information about the CSC. The "W5" format of the first page was my invention, and is a rather informal contrast to the Documents section. The only criticism I have ever received in regards to it was a CSC regular ranting about the evils of bergamot oil.
The Past Events section is mostly for reference and archiving as well. Certain events are repeated regularly, and it is handy to have the old notices around. There are also some online materials and resources made available for certain events, and links to these are maintained. The Flash section is a relatively new one, and still somewhat under construction. It serves the same sort of historical function as the Past Events section, by simply archiving for posterity the CSC flash. HTML pages for individual terms have not been written yet, and the last couple of terms haven't been archived yet. As well, the HTML in the individual flashes does not approach the strict quality of much of the rest of the site. These are problems which are not difficult to fix, just time consuming.
The Pics and Quotes sections are a mixture of humour and archive. Out-of-context quotes are written down on a "quote board" (piece of paper taped to the wall) in the CSC office, and later transcribed into this section. Humour is the goal, public humiliation is the means. Certain worthy pictures are selected to go up as well. I hope to incorporate more of the CSC's corporeal photo collection into this space, but haven't yet had ready access to a colour scanner and the time to do it. Due to its size rather than popularity, the large MPEG movie found in this section often shows up amongst the top twenty bandwidth-chewers in the statistics.
The Statistics section contains statistical information for the web server, updated daily. This information is generated for the benefit of the www admin, as well as the members and clubs with web pages. Members and clubs can see which of their web pages are more popular, and perhaps adjust them accordingly. The www admin can do the same, but the other purpose of generating statistics is to catch bandwidth abusers. This is the purpose of the top three links on the statistics page. The system used is a bit unwieldy, as trends are not easy to spot without careful examination of the data. I have not yet figured out an acceptable way to improve this situation. The process of generating statistics is fairly long, due to inefficient software (a large perl script) and relatively slow hardware. This means that the amount of statistics generation must be kept to a minimum. The software can efficiently incorporate the results of previous runs, so the current setup is a rather obvious one.
The Members' Home Pages, and the Clubs Home Pages sections are actually the biggest sections of the CSC web site, but they are only marginally under the control of the www admin. All content found in these sections is the responsibility of the other people who maintain their individual pages. The first page of the members section is entirely automated, requiring manual intervention only in exceptional cases. The first page of the clubs section is manually maintained, but it changes very rarely. The remaining sections (Events, Lib, Misc, Pix, Skeletons) are used as organisational aids, and do not warrant discussion as sections.
Overall, there are several principles being adhered to. Foremost in my mind is the quality of the HTML. Wherever possible, it is written to strict compliance with the HTML 2.0 standard. If this version doesn't have a particular desirable feature, HTML 3.2 is used, again with strict compliance, and an eye to what effect it will have on viewers. This ensures that all web browsers can view all of the CSC web pages, without having to resort to different versions (eg. no-frames, text-only). As mentioned previously, the Flash section still needs a lot of work in this area. mathNEWS does not follow the strict HTML standards of the CSC, and I haven't had time to fix it up properly. Use of graphics is kept fairly minimal, and ALT attributes are used wherever appropriate, further ensuring accessibility. Another principle being followed is consistency of presentation. The use of server-side includes (SSI) has enabled all but a few of the pages have an identical footer. This means that a change in one file effects the change throughout the web site.
A secondary purpose of the web site is a learning experience for the administrator. The first time I was www admin, I had almost no idea what was involved, or how the underlying technologies worked. After my first term, I had a some idea of how web servers work, and how to use them. This experimental aspect of the site has resulted in numerous different ways of accomplishing basically the same thing. Specifically, several sections are divided by term, but no section does this in quite the same way as any other. This gives new administrators several different perspectives on how to use the technology at their fingertips.
The CSC web site fulfills its purposes well, and is fairly good at following the principles of good design. There are several elements of it that could stand improvement, most of which are a consequence of the limited time available to the volunteer www admin to fix them. A more general problem is the lack of feedback that it generates. I have solicited suggestions and/or criticisms numerous times, without success. Of course, this could be taken as a good thing. One could assume that "no news is good news", but I am wary of taking that view, as it is rather self serving. I wonder if, by taking on the role of www admin so often, I haven't in some way restricted innovation. Perhaps my perspective on the site is fundamentally biased, and I should step aside completely to let other people shape it differently. I have stepped aside for various terms in the past, but little has changed. It could be that a single term is not long enough for a new sensibility to establish itself, and a longer view must be taken. This will happen very soon anyway, when I graduate, and I can't help but hope that I will leave a robust environment, which allows experimentation without jeopardising the stability and usefulness of the information contained in it.