mrry (Happy New Year)
Blog Reality Check 11/May/2003

I've not known what to think about for the last couple of weeks, and now I'm hit with the rude awakening that is the Summer exam diet - beginning in earnest one week on Saturday (24th June).

Perhaps I should have exams in mind, but - what's that in the near-distance? - there are three assessed exercises to be done for Computing.

Not too stressful really: I've written my functional program, and had it accepted. I've written, but not made pretty (yet), my cheque clearing program, utilising a generic table class. (Contrary to the linked article, I enjoyed writing it, but then I am a fan of object-oriented programming, and its antecedents.)

Last, and least, for my Information Management module I have to create a "website", based on a database, and evaluate it. You'd think that that would be enjoyable, given that what you are reading right now is a website, based on a database. However, due to time- or teaching- or marking-constraints, or whatever, we are forced to put up with a neutered client-side/Microsoft Access implementation. Access is a good package for creating small-scale database applications, but it's worse than useless for creating websites. Pages are either static, or incorporate an unwieldy ActiveX control. Said control is very difficult to manipulate.

The punchline of the joke is that we are to formally evaluate our design. We must round up a few subjects, and get them to carry out five(!) small tasks. Given that our "websites" only comprise five or six pages, coming up with a task sheet is no small order. Given that we have little-to-no design control over the Access-generated pages, how can an evaluation possibly reflect on the designer?

With all the pretensions about this being a formal exercise, why not teach us (a smattering of) PHP and MySQL in the first exercise, set up a small web-server, and let us try our hand at true server-side scripting? Say we only had to write two pages, fully XHTML and CSS compliant, one of which is a PHP application? Or at least offer us that option?

If I had been able to do this, I would have been left with something usable, for a project I have in mind. But having used Access, I'll have to start again from scratch.

Apologies if I'm sounding a little ranty today, but I wanted to get that off my chest, and this is the most likely forum for people to agree with me.




Gary said:
It's not that I'm against OOP, I quite like the idea. But I think that generic OOP is taking it a bit far: everything is a bit too hazy and abstracted.

The architectural astronauts essay that I linked to makes some good points, as does an interview I read a few years back featuring the lead designer of Unreal. Basically he encouraged his team to mix all elements of the code (physics and graphics) as they appeared in-game. So code for, say, a switch is directly tied into the physics behind the switch and the geometry used to draw it (quite unusual in game design). The point is to keep everything that should be together, together and not let layers of abstraction get in the way of productivity.

Anyway, from your post, I can see that you too are a 2nd year CS student at Glasgow. By any chance, do you know of any other people in our course who have blogs? I've tracked down 3 people so far.





Please enter the number 9524 in the box below:

CommentsTell a friend about this page

Your Name

Your E-Mail

Your friend's E-Mail


< # Scottish Blogs ? >
Technorati Profile
Listed on BlogShares

Subscribe to the mrry RSS feed
More about RSS.
Trackback URL for this article: