Sitsofe's diary for February 2001

28 February, 2001

Group photo was taken by Mark. I'll put up the best one in the Photos Section when I've got a bit more time to sort through all the pictures that have been taken with arthur's digital camera.

Today I decided that it would be a good idea to make the change to using queues that contained both a head and a tail. This turned out to be a huge structural change to the program with massive repercussions in the pieces of code that did the actual encoding/decoding. The code was complicated already and hunting down bugs turned into a mammoth task with some major queue rewrites.

I've started thinking that we won't have time to implement and run all the tests that we should for this program. I've spent so much time getting the existing code working that I haven't had the time to write to write tests to prove that it's valid. Whilst I do test my code to check to myself that it's doing what it should be, I have not written them down :(

In the later part of the evening to work how big the drives in the DEC external storage units were and whether they even worked. It turned out they did work, but need a long time to get up to speed and for the computer to send a spin up signal to them to start working after this long start up. I also took the opportunity to replace the IPX that we were using on iridium with the Sparc 2 that was spare. This meant it was yet another long night.

27 February, 2001

I've borrowed arthur's digital camera but unfortunately the whole team was not together so now photo was taken. Lots of work was done on the group project, starting to become very tired. We really should have finished all the coding by now but this hasn't quite happened.

25 February, 2001

Major rewrite of group project code. Decoding is finally finished and now encoding uses the a queue to read in a line making it work in the same way to decoding. Also had a variable renaming session to make what was going on a bit more obvious.

24 February, 2001

I've been fiddling about with JBuilder again. It seems to be working a little faster under KDE 2 than under Gnome so maybe there are some window manager issues.

I've also noticed that Netcraft have noticed that sucs went down and the uptime has dropped accordingly.

I'm beginning to wonder about this XML thing. Could it be worth looking at just in case it turns out to make life easier? What if it means I'm more likely to update this diary?

23 February, 2001

David (Brooks) has been made aware of the greatness of vi(m). Although he claims he prefers to use Nedit. I think it is only a matter of time before vi becomes his Linux editor of choice. Actually this is no bad thing - allow me to elaborate:

Vi is available on many different platforms and on almost every box that is running some sort of *nix (it seems to be considered an essential utility). In fact, after I installed QNX the other day I noticed that it too came with a version of vi (elvis). So once again I had no trouble editing text files and no new learning curve to progress up. This to me is possibly what makes vi better to learn (at least initially) than emacs if you are working on a variety of circumstances - vi will always be available in some shape or form whereas emacs may not (especially on systems with limited hard drive space).

Vi is small and works from the command prompt. The reason why David was using vi in the first place was because I was too lazy to tell him how to get a remote X working. As such he only had telnet which is not enough for nedit (or at least not the versions I've seen). However vi doesn't need X to run (although if you do have X you may want to look at gvim). As such it can be run across telnet where there is no gui but it still works fine. Gareth (James) routinely uses vi to edit lots of his coursework on sucs across the net from home.

Finally my favourite thing refers more to vim than vi and that's the number of languages it can syntax highlight by default. It's an awful lot. It can also do intelligent indentation on c programs and a reasonable stab at other non-c syntax languages. Add to that the fact that gvim can turn it's highlighting into HTML and I'm ready to quote Dave's favourite catch phrase - "beautiful".

Poor old gw (the sucs router) hung itself today during a heavy match of Xblast. I suspect that this was due to all the packets that were being sent to and from cmckenna's machine on the guest network. After almost emergency restarting Platinum, we brought gw back up.

Roddy also brought his machine in for a reinstall and unfortunately managed to break his 500Mb SCSI drive by giving it an incorrect Sun disklabel and formatting it. I don't think Jason (who had offered to take some of the old Suns off our hands) was too impressed by this.

22 February, 2001

Booted into Windows and hunted about to see if anything needed updating (which of course it did). Grabbed new copies of my mouse driver, Winamp, distributed.net client and cdrw firmware.

Spent part of the night wandering why I couldn't see the length method inside the class definition for arrays in Java. It turned out that this was because arrays in Java aren't really objects...

21 February, 2001

I am very unimpressed with the difficulty of getting a Java program running on the computers inside the Compsci Linux lab. Fiddling about with my PATH and setting CLASSPATH environment variables is not my idea of a good time, especially when I'm using (bleh) csh.

This whole episode has made me install the Sun JDK on Platinum. It's certainly a lot less hassle than doing it up at the Compsci lab.

Today may also have been the day that I encountered that "ALL YOUR BASE ARE BELONG TO US" meme (courtesy of Captain Cursor). I can't help thinking that I was a bit slow seeing this one coming but I also reckon that this sort of thing lends credence to the idea of memes (which I first heard about from a guest speaker at BGS in my penultimate year).

20 February, 2001

At sometime after six in the evening Gareth (James) pointed out that sucs was not accessible I tried to ssh in to fix the problem but couldn't get any further than gw. I popped down to the sucs room and discovered that the problem was that the reset switch on Silver had been accidently pressed and the subsequent fsck on the reboot was asking for the root password. Since no one in the room at the time knew the root password this was something of a problem. But what do ya know? I know the root password so after manually running fsck (and pressing y lots) the server came back up.

I have to admit that one of the main benefits of having stuff on sucs is that because it is used by people who can fix stuff it stays down a lot less than other places. In short the people are knowledgeable and also have a vested interest other than money in keeping it running.

19 February, 2001

Gave Alfie the Java JDK instead of the Java JRE which only allows you to run Java programs and not compile them. Learned that getting Java up and running in the Linux lab up at Compsci is throughly convoluted (no surprise there then).

17 February, 2001

Finished script to allow my computer to turn itself off when no one was using it. Realised I had lost the Crimson Skies receipt and as such I could only get Electronic Boutique to give me vouchers. The train I caught home was horrendously delayed and I missed all my interconnecting trains as a result.

16 February, 2001

Sadly didn't get to go home because of presentation at uni and generally poor planning.

15 February, 2001

I gather cmckenna managed to get the bnc cable that was thrown out of the window of debates B working. Many thanks to Chen and the Compsci department for lending us that mini hub.

14 February, 2001

Most of the evening was spent trying to get rid of stupid bugs in the group project encoding unit.

10 February, 2001

Spent all of the morning and most of the day in playing Crimson Skies. Finally changed out of my gown at around 6pm.

8 February, 2001

Bought Crimson Skies as an early birthday present to myself.

7 February, 2001

I've been spending large amounts of time trying to do my part of the group project.

4 February, 2001

I seem to have found one difference between 2.4.0 and 2.4.1 - 2.4.1 seem to be more unstable. Upon booting today it locked solid when trying to start X and the magic key sequence wasn't successful in preventing an fsck. Having said that a consequent boot seemed to work fine.

3 February, 2001

Updated my site with new The Counter javascript which now means that the front page validates. Decided to reverse the order of my diary so the newest day is added at the top. Got round to mailing Simply again because they still haven't replied.

2 February, 2001

Someone rattled the sucs room door today whilst I was sitting down out of view. Obviously the potential thief didn't know exactly how low spec the computers in sucs are ;) I also hoovered the room and gave lead a new monitor as its previous one had been producing colours that were far too dark to read. I owe cmckenna a debt of gratitude for taking the monitor from me when, after picking the monitor up, I received a load of static shocks. I have been trying to avoid opening computers and lifting monitors since.

I also compiled the 2.4.1 Linux kernel but I can't say I noticed a huge difference from 2.4.0. Oh and I grabbed a nightly build of Mozilla.

I can only express disappointment with the way that incremental revision (the idea of steadily looking over notes every day) has been treated by some of the people I revise with. I think I will give up on doing it in a group environment if it continues to be taken with such a blase attitude.

1 February, 2001

Finished putting what will end up as Gold (which is intended to be a computer purely for burning CDs) together from bits of 486 and removed the broken hard drive from pb (a case that can hold two full height hard drives). It turned out that pb's problem was that one of the hard drives in it had died and that drive turned out to hold data that the sparc needed to boot.

Picked up group project details. Need to work on it over the weekend.