Well That’s Just Perfect…

Friday, October 31st, 2003

I guess this is what happens when you’re too stingy to pay for all the variants on your domain name: www.jujusoft.net

Seems a little unreasonable since when I chose jujusoft as a name none of the variants were being used and Google showed up zero results. Still, maybe they’ll turn out to be a much more successful revenue generating jujusoft and I can sell my half of the name to them for a lot of money ;)

Crossroads

Wednesday, October 29th, 2003

I preferred the PC before it was ubiquitous.

It was more fun buying a computer when the majority of people weren’t interested in them. When most people [including me] had not heard of the internet.

Computer programming as a hobby was fun. Writing code to draw circles and other geometric shapes was fun. Progressing from yucky old BASIC [with line numbers] to comparitively hi tech QuickBasic [with subroutines] was fun. Dabbling with the awesome power of machine language was fun. Even working out how to access more than 64K of video memory at a time was fun. Writing your own assembler based on code snippets from magazines was fun. It was all fun.

Feeling like you are among the first to do something is fun. Messing around with the guts of a new technology is fun. Reverse engineering is fun.

Inventing the wheel is fun. Even when you find out later that someone beat you to it by a million years.

Being expected to learn .NET, J2EE, XSLT, PHP and the thousand other technologies and methodologies that are in current circulation is not fun. It is overwhelming and disheartening. It makes you realize that any idea you have has either a) already been done by someone smarter than yourself, or b) is a waste of time because there are too many new ideas floating around already, and people are more interested in solutions than ideas.

Basically, now that people actually understand what computers are for, it is just so much harder to get excited about. These days, no one is going to write a movie about a personal computer becoming self-aware , because it would seem every bit as stupid as a movie about a microwave oven coming to life. For the same reason that a movie like Wargames could not be made today, I can not enjoy programming as I once did. There is too much information now, and too little mystery.

What used to be "the great unknown" to me is now simply "all that stuff I don’t know".

I still do things for the fun/challenge of it [eg create my own scripting laguage or attempting to write in Perl whilst stubbornly refusing to read help documentation], but instead of feeling proud of such achievements, as once I could have, I am now almost sheepish. You see, these things are a frivolous waste of my time and skills, and these days if I really want to program "for the fun of it" then I am supposed to be assimilating my technological distinctiveness into that collective that is known as the Open Source movement.

The idea of such collaboration, and the various concepts such as "peer review" that come with it are unsettling to one such as myself, still stuck in the mindset of computer as puzzle/challenge/toy, rather than universal tool. I was never so concerned with writing better software so much as seeing what I could make these computer thingies do.

And so I find myself at this awkward point, where I am coming to believe that the future is indeed Open Source, but at the same time that I will not be a part of it.

Look, it’s me!

Sunday, October 26th, 2003

This is a site I’ve visited before, and I still don’t know what to make of it. It is so over the top it appears to be satire, but it is so consistent it almost seems authentic as well.

While you’re there be sure to check out the Creation Science Fair

[Thanks to Ned Batchelder for reminding me of this site… there as in many places people discuss to what degree the site is a parody ]

Amazon adds full text search

Friday, October 24th, 2003

I’ve seen this announced in several places already, but in case you haven’t heard, is now supporting , ie you can actually search for books by content as well as author, title, etc. Seems to me a fantastic thing if you can remember a great piece of dialog but not where it comes from, or perhaps a particular character’s name. Or you can just think of a particular phrase and see how frequently it is used. For instance the phrase "slid down the wall" returns 184 hits, including this one:

". . . life from him and sent him knocko backward against the wall between the two rest- rooms, to the left of the manager’s office. He slid down the wall, leaving a webbing of gore upon it, for . . ."

Ewwww! Of course, doing a Google search on the same phrase returns a substantially higher number of matches (more than 7000), but there’s something special about knowing that a text actually exists in print.

Security issue?

I think the full text search itself is free to anyone, but to view the pages found you have to be a member with valid credit card details entered. I was about to enter mine when I noticed that the page asking for my details was not encrypted (ie the connection was not a secure one). I don’t know how much difference this really makes these days but I for one will never type my credit card number unless I can see my little padlock icon.

I suspect this is merely an oversight due to the newness of the service, but they really should fix it. [I sent them an email about it, because I like to pretend there are people at the other end who actually read such correspondence]

Wednesday, October 22nd, 2003

Truth as Satire

The Onion: Muscleman Put In Charge Of World’s Fifth Largest Economy

Arbitrary style change #1057

As usual I grew to loathe the way these pages were looking, so I changed the style again. This time I’ve decided to try the banner image as a feature, and generally tried to make the thing more approachable (ie more like other people’s blogs) The previous look just had too many framed boxes in it. Anyway, it’s either that or write some software…

Astrology Revisited

Sunday, October 19th, 2003

A while back in this post I mentioned an online Astrology forum I had participated in. Reading it again now I am surprised at how carefully I constructed my comments, given that I have such an incredibly strong aversion to Astrology and the psuedo-sciences in general.

I stand by most of what I said, and had to smile when I saw a little quiz I’d added:

1) Which of the following was most responsible for dragging western civilization out of the dark ages?

a) Astrology
b) Religion
c) Politics
d) Science

2) Which of the following is the most self critical?

a) Astrology
b) Religion
c) Politics
d) Science

2a) Which of the following eagerly anticipates revolutionary reform?

a) Astrology
b) Religion
c) Politics
d) Science

3) Which of the following benefits least from people’s fears and insecurities?

a) Astrology
b) Religion
c) Politics
d) Science

But does that constitute sarcasm? Is it just a cheap shot? Personally I think it makes a valid point in a succint and entertaining fashion.

Idle programmer wastes time on Perl comments script

Believe it or not, if you disagree with me you can actually say so in my first ever public comments/feedback section: Astrology: Is it so bad? [UPDATE: link removed since moving to intepid.com]

“Astroturf”

Thursday, October 16th, 2003

If you don’t read many political blogs/punditry, you may not have come across this term.

A perfect example of Astroturf is in the flood of letters which recently appeared in newspapers all over the US, purporting to be from US soldiers in Iraq. The soldiers talk about all the positive things that the US has achieved in that troubled region, and basically attempt to give an honest account of the state of affairs from the ordinary soldier’s point of view.

The problem is, all these letters are in fact the exact same letter, written by just one man, Lt. Col. Dominic Caraccilo, and mailed at least 500 times under different names and signatures. Often the letters were run in the home town newspapers of the soldiers whose names were falsely attributed. More on this particular story here…

I think it’s important to be aware of astroturf for 2 reasons:

  1. Public/popular opinion can be skewed, ironically by misrepresenting that very same public/popular opinion.
  2. When astroturf occurs, it is actually relatively easy [and fun] for anyone with access to the internet to uncover.

In most cases [of this abuse] a Google search on a distinctive phrase in a well written letter to the editor will often turn up a surprising number of nearly identical letters published in different newspapers and attributed to different people.

Up until now much of the Astroturf uncovered by observant media watchers has been related to the outstanding leadership qualities of President George W Bush.

Perl: You don’t have to understand it to use it!

Monday, October 13th, 2003

Spent a glorious hour or 3 modifying the generic search script I just installed [see previous post] and managed to add the following groovy features:

  1. Handles quoted strings now, so you can search for a phrase.
  2. Returns extracts from each page with search term hilighted for easy reference.
  3. Embedded HTML is now more W3C compliant and references a style sheet at the server root to make customizing appearance easier
  4. Filters HTML tags out of documents before searching (and ignores the document HEAD) [this may be irresponsible on my part, as it involves tag stripping every document before searching, which may make a little more work for the server than is considered polite]
  5. Supports an include list as well as an exclude list (for deciding which pages should be searched) Basically it seems pretty dangerous to set up a site search if there’s any "non-public" or just generally messy data lying around on your site. Unlike a search engine, a cgi based search will scan every file on your site, whether it’s linked or not. The easiest way to keep the results clean in my particular case is to whitelist individual log entries only.

It will be interesting to see if my host kills my new script. That’s what happened last time I made substantial changes to my formmail script.

I know all that sounds pretty normal as far as searches go, but strangely none of these things were supported in the default script.

This is exactly what I’m talking about when I say I don’t get why people use these old scripts [The script I have been modifying is dated 18 April 1996]. As a dyed-in-the-wool C programmer, I can barely even read Perl and yet still I can make sensible changes and improvements in just a few hours [and only that slow because every time I test I have to upload the thing to my Web Host as I don’t run Perl or a Web server locally].

So go on, use this new super improved search facility to find out how often I use the word bottom! [UPDATE: Link removed since moving to intepid.com]

Search added, just for the hell of it

Sunday, October 12th, 2003

In reality I could have added it myself using cgi scripts, but I probably would have ballsed it up ;)

My ISP finally got around to letting me add search capability to my site, so there it is in the sidebar for a trial run. Go on, try searching for a few rude words…

Precious Perl

One thing that I really find weird about the mysterious world of Perl scripting: On one hand Perl programmers are very proud of their art, including its idiosyncrasies, and yet on the other hand you get ancient buggy scripts being used for long periods of time because people hardly ever write a better one.

There’s this collection of scripts at Matt’s Script Archive, which by the author’s own admission are not terribly well written or secure. On this page he recommends using drop-in replacements developed by some other dudes, available here.

But why are these guys so precious about replicating the interface to some 7 year old code? It’s not as though they are a joy to install anyway, and configuring is a total bitch given all the embedded (badly formatted) HTML code spread througout the Perl scripts themselves!

A truly decent formmail [or guestbook or search interface] would automatically support human readable template pages uploaded by the user, so that no editing of the perl files is required. If I want to change the way my search page looks, this is just one of many chunkettes of data mingled with code that I have to edit:

sub PrintBodyHTML
{
    local($filename, $title) = @_;

    print <<__BODYHTML__;
  • $title - $filename
    __BODYHTML__
  • }

    Go on novice user, get stuck into that! I guess this is maybe what PHP is designed to rescue us from?

    Unfinished Thing: JujuDraw

    Friday, October 10th, 2003

    Like most of my software, there is no new brand new technology to patent here. What I do have is the start of a promising no-fuss sketchpad. Every stroke is stored (and editable to some extent) and the user always sees the drawing as it will be finally rendered… ie there is no sketchy "preview" version of a stroke while drawing. Simple export to SVG currently supported (click on image for svgz file, you will need an SVG viewer installed ).

    Like many projects, its current strength is in its simplicity, and feature creep could make it suck. The outlook for JujuDraw is uncertain.

    Thursday, October 9th, 2003

    Congratulations California!

    Sure Arnold has no political experience, but as an actor he is about as professional a liar as can be had, and that’s bound to come in handy! Although, governing a state is probably a little more like directing than acting, what with all those pesky bureaucrats to oversee… lucky for you guys Arnold has had experience there too, directing the 1992 smash-hit Christmas in Connecticut! Your future is obviously in good hands…

    Going with the flow

    Thursday, October 9th, 2003

    One thing that keeps becoming clear as I screw around with color is that it is hard to abandon white as a background color.

    Not only is white the most paper-like, and therefore most compatible with printing, but sticking with it means you can standardize your images to look good on a white background.

    Once again I have spent several hours of my time screwing around with the formatting of this blog, rather than doing anything actually useful or constructive.

    Hmmm… that’s a pretty predictable thing to point out, and I’m sure that thousands of bloggers say it everyday.

    What’s problematic is that I think I like the layout less now, although the whole idea is to make it more flexible for variable width browser windows [it seems such a waste to force people to put up with wasted space and or horizontal scrollbars] Just taking the old template and making it variable width didn’t seem to work very well.

    But now it’s just looking more and more like everyone else’s generic blog template. It’s almost as though I’m trying to pretend I’m using Blogger or Movable Type when I’m not. If I were a graphic designer I swear I would come up with something more original but at this stage I can’t help being influenced by the way other people’s blogs look.

    I have also added "proper" monthly archive links, so you can check out what I was doing in November 2002 if you like. This stuff still isn’t as automated as I’d like it to be, but it’s getting better. The entries in each monthly archive are stored earliest first, which is a bit inconsistent with the main page but it’s nice to think that someone could comfortably read things in chronological order for a change.