S.KRAUSE

I do not suffer fools gladly.

Few do.

In Kansas they are debating the merits of evolution and intelligent design again; the world is full of morons. Let me list two of them: attorney John Calvert, managing director of the Intelligent Design Network, and Kansas School Board chairman Steve Abrams. Intelligent Design is neither intelligent nor really about design—discuss.

Fake/Celebrity Blogs

The following are my favorite fake and/or celebrity blogs, though I admit that I like them more in principle than in actuality—I do not actively read any of them. But I like the concept(s).

• Stephen Hawking’s Blog
• The Darthside: Memoirs of a Monster
• Terri Schiavo’s Blog
• The Pope Blog (not “fake,” only unofficial)

The Day in Review

At 10:07 I left my apartment; outside it was cold and damp, uninviting, grey, and clearly the wrong month. Before stepping out the door I paused to read a notice posted on a board: the resident of apartment 305 has had his or her electricity disconnected due to failure to pay. The bus was prompt, quick, and not crowded.

In Van Hise I graded; I planned on showing a film during class (“Was tun, wenn’s brennt?”). Sarah’s laptop died; this was perhaps a sign of things to come for me. At 1:12 things started going downhill. I took the stairs, but had to go to the second floor to pick up the TV cart and DVD player. Along the way I got stuck behind two students walking side-by-side who decided to walk down the middle of the stairs, making it impossible for anyone to pass on either side; in addition, they were going at a very leisurely pace during the passing period. Morons. Once there, I discovered that the twits working for LSS had lost my reservation and assigned my cart to someone else for the same time, even though I had specifically reserved cart 10 because it had a multi-system VCR, which was necessary for connecting the region-free DVD player. Long story. They transferred things around and gave me the proper cart, but failed to give me the DVD player, so I arrived on the 5th floor with a TV, multi-system VCR, Laserdisc player, but no DVD player. Fucktards.

We did not watch our film.

We listened to two presentations (on exploding toads and on mini-nukes, aka bunker-busters) and then I gave feedback on the presentations. Today was a loss; Wednesday, which I need to re-organize since the film is not an option, will hopefully be another matter.

RSS

I had no interest or need for RSS ... I still do not, to an extent, for there are few things that I check where an RSS reader would really improve my reading experience. I prefer a more anarchic, loose browsing pattern over a structured one. Some friends claim that if it is not provided by RSS (and it is blog/diaryesque) it will not be read by them. I am not a huge reader of such sites, but after a certain amount of convincing by these friends I decided to put together my own feed/channel.

Being the independent sort of fellow that I am, I left the task of learning how to create and RSS file/feed/channel to google: rss tutorial. I found this RSS Tutorial for Content Publishers and Webmasters relatively concise yet informative and instructional. It exlained what RSS is as well as its various versions, and provided several useful links to other sites or resources. In order to create my own feed I downloaded several samples; the one created by Slashdot was a bit messy and large, so I instead modeled mine on mns’s. I basically took the structure of his file, inserted my own contents, and removed anything that did not apply to my project. I then put it up for a few friends to test, and when they reported that it worked, then I was happy at last.

The file itself is available on my website, and I try to update it whenever I add a new entry here. I also downloaded and installed wnews, a console-based rss reader. It does not like extra HTML in its RSS feeds—nor should it, for it is bad style to include extra markup in the feed, but a lot of blog software seems to do so anyway. As for the future, perhaps when/if I restructure how I run my site, I will code an rss-generating module. Until then, I will stick with my minimalistic by-hand method.

Web Content and Structure

Several issues face a site such as mine: size, organization, and the relationship between dynamic and static content. A small site does not need to concern itself with matters of organization, but mine has not been small for a number of years. I then organized my site according to a few guiding principles and as a result I ended up with two large categories—sections and projects—, each of which had numerous sub-divisions. All the sections had the same design; the projects were independent and could have their own style. However, over time only the writing and photo sections have been growing, and now I need to focus more on them. By separating presentation and content to a degree (by moving to SSI and CSS) I made dynamic content possible on my previously static site, but I am interested in adding more dynamic features, both so as to provide a richer experience for visitors as well as to make my administrative work easier.

In the beginning (1996) I had a series of rather plain pages, both my own and those related to course projects. Once I began to care about organization I switched to frames to separate content from navigation, but once I realized what a terrible solution frames were, I dropped them in favor of a table-based layout, which was, honestly, only a temporary solution, for tables should be used for tabular content, not for layout.2. History of solutions to now. Even when I had tables I implemented CSS and ditched all <font> tags and such, but later I moved to SSI (Server Side Includes), which allowed me to provide site-wide abstraction and layout; <div> tags solved the issue of tables. In order to add a comment system I found it easiest to provide some pages in PHP, and this added another layer of technology to my site.

This all, however, has its drawbacks. Plain html, shtml, and php pages are now mixed in one large site. I have SSI templates as well as SSI cgi-scripts, in addition to database-driven PHP pages. There is site-wide layout, but the subsections and projects are not evenly balanced, and some of the projects are still archaic in their markup. As the number of pages grows the ability to keep track of all changes by hand is strained. In the past I did site rewrites, but to do so now, in the age of Google and relatively high rankings for some of my pages would mean confusion for many visitors. My dynmaic content, specifically my comment engine, is a dirty hack, albeit one to which I gave a decent amount of thought and care. It is limited in functionality and currently not very extensible. I would like to add a number of interactive (sub)systems, but scalability and interaction between systems is a concern.

An off-the-shelf product is one possible solution, be it blogging software, a CMS, or a DMS. The more minimalistic blogging packages, such as Simple PHP Blog (recommended by mns) or PyBlosxom, championed by Ryan. More full-featured CMS packages, such as scoop or zope, would be overkill, both in terms of features as well as scope and resources. And since I a more interested in webpages as content, a DMS is not really for me. As is always the situation with pre-coded solutions, the solution rarely if ever fits the problem set exactly, and thus it has unneeded features as well as gaps in the list of desires features. That is perhaps one reason why so many people end up coding their own.

I will focus on abstracting a better templating solution for my specific needs, and probably borrow some features from other projects along the way. In addition to searching for solutions, it is often useful to return to the initial question to see whether it is in need of reformulation. In my case, I should not focus necessarily on dynamic content; for me and for most folks, comments are about the only dynamic content necessary, though occasionally having the whole site be a representation of a database has its uses. Templating, too, can be done in a more static fashion; one can create pages at the time of request, or one can “compile” them earlier.

On that note I would like to mention a few interesting projects I have found along the way:

• Bash Blogger: blog software written in bash (or “a small, bash-powered, weblog engine” as the author likes to call it)
• NanoBlogger: another bash-powered “web engine”
• HTDB: “an elegant, database-aware, extendable scripting and programming framework in which the drudgery is removed from the task of building dynamic web applications.”—mns recommends it, and I like the idea, but I am not yet willing to convert my pages.
• What about putting together a complete CMS in bash (or another Unix shell)?
  • webserver: see bash-httpd, for example
  • embded in the webserver for the scripting engine: although no one has (yet) done it, one could write mod_bash
  • SQL engine: Dana French has sh and ksh-based SQL projects; someone else wrote a mostly-usable SQL engine in sed.

This last big is a foolish project, and some would argue that it should not be suffered at all, but much like bad puns they are ideas that I cannot resist.

It snowed this morning. It snowed yesterday afternoon. It rained on Saturday. It is May. On the LUG mailing list they have been discussing file-based CMSs. And a bash-based CMS is foolish?

—May 2 2005

---