exo : blah

New version of my poorly named Blosxom plugin to let you access the title of an entry in it's individual archive page. This fixes problems with files containing . and -. Thanks to Antti Vähä-Sipilä for pointing out what my less than complete testing had missed.

Download storytitle 0.5.

After a fashion anyway. Poking through the weekend's Observer I came across an article on the decline of independent music shops which is saddening in itself. They're great places in which I've spent many happy hours. More saddening was this tale from the owner of such a shop.

'There was a guy came in last night,' he says, 'a young bloke of about 22. He said he wanted a couple of posters we had on display, and offered to buy them off me. I said, "I don't really like to charge for them - just buy an album, and I'll give them to you." He said, "Oh, I don't really buy albums any more. I download everything." What did I say to him? Oh, I gave him the posters. You've got to be friendly.'

What can you do in the face of such staggering ignorance? You walk into a shop and cheerfully inform the owner that you are undermining his whole business for the sake of a tenner or so. Just buy a fucking album. Given how much money you'll have saved from downloading stuff you think you could at least see you way to handing over the cost of an album. Think of it as a charitable donation if you like.

As a footnote I'd like to say it's not that I'm pro the RIAA and it's heavy handed tactics to cling onto it's bloated, creatively bankrupt business model. Just that I'm not sure how distributing music for free makes things better. After all, if you find the RIAA and their cronies that egregious there are ways to avoid buying from RIAA members.

No. 1:

The report identified domestic violence as underfunded.

From the back page of Saturday's Guardian.

After a fashion anyway. Rest assured that under the hood though it's all change. Gone is the home grown database access layer and in it's place a shiny new Class::DBI based one. Thanks to this it's all a lot cleaner, more extensible (hence the addition of the journal bit) and approaching a stage where I'd consider releasing it. Maybe if I can work up the enthusiasm to either write an install script or steal chunks of someone else's. First I need to smooth off a few rough edges though.

You've got to love it. In the bad old days altering the look of your website was, even assuming it was in some sort of Content Management System, a painful process of changing table and font tags, redoing graphics and generally throwing out vast chunks of the previous iterations HTML. Now, assuming your HTML was fairly clean, a bit of faffing with the stylesheet and you have a lovely new site design.

This is all a roundabout way of saying we've done some design tweaks here. Nothing to write home about really. It may have some rough edges in some browsers (well, anything that's not a recent Mozilla based effort to be honest) and it will look like arse in old browsers. Those in the latter boat might want to consider upgrading if they can. Of course if you use lynx and friends then it'll look much the same as ever.

In order to get round the fact that some people on the web seem to feel it's perfectly reasonable to stuff false referers in your logs in order to either increase their googlejuice or just as some kind of crazy low rent advertising you can now block these. Or rather you can block repeated ones from occurring. Simply fill in a list of IPs or URIs in @logspammers and if these either appear as the referring URI or the referring host then that entry won't be displayed. There are some in there to get you started (hello to whoever it is at the University of Bucharest that loves to fill my logs with this shit).

Oh, and and it knows about a few more search engines now.

Download apache_referers 2.1

Remote release cables for modern cameras are expensive. £40 is how much one for my Minolta costs new. Even on ebay, the saviour of cheapskates everywhere, it's looking like £20 is the going rate. The nice old fashioned mechanical ones you screwed into the top of the shutter release are about a fiver new. You can pay more for slightly posher ones but for occasional use £5 is a lot more attractive.

In more barely useful scripts news: pmail_recent will summarise your procmail logs. Get it from the misc code page.

Another update. Now you can list the date of the refereral as well as the host the request came from. This means the data passed to the template is completely changed so templates need to be updates. You'll also need to delete the file that old data was stored in.

You can also determine the maximum length of the target and refering URI as well as the host. Anything longer than that and the displayed version is truncated. As ever see the results on the referers page.

As a bonus the documentation's a chunk better. perldoc apache_referers or scroll to the end of the file to read the raw POD.

Download apache_referers 2.0

URIs for search engines get pretty long and they're not very easy to decipher. So, to make the referers page a bit easier to read I've tweaked apache_referers to provide the much easier on the eye "www.google.com search for things" instead of the URI. It should handle most search engines and I'll be adding new ones as I come across them.

Download apache_referers 1.6

email. It's a nice simple thing. It's all nice plain text. You send it to someone and regardless of their mail client they can read it. Or at least that's the theory. It's certainly the beauty of it.

Of course that simply isn't good enough for people. First they said that plain text wasn't expressive enough. They needed different colours and typefaces. They needed to embed images and have coloured backgrounds. They needed mail clients that could create and read emails formatted in HTML.

And then there was the problem that once they sent an email the recipients had a copy available to them for all eternity. Surely that can't be a good thing with something as easy and quick to send as an email. Surely something could be done? It could.

We've come a long way.

As has been noted in various places Unicode is not well understood by many programmers. I'd include myself amongst them. However, I'm a lot better informed than I was. I'm in the process of updating my RSS aggregator to handle Unicode better in versions of Perl that support it.

Useful things I have learned in this process are:

• Always check that all the software you are using is Unicode aware. Some versions of Gnome Terminal aren't and the Unicode aware version of xterm is a separate program.
• Make sure you flag in the appropriate way your data's encoding. In my case it was setting the Content-Type header of mails to text/plain; charset="utf-8".
• Mixing different encodings leads to surprising results. Life is much easier if you convert everything to a common encoding and then do any munging on it.

Anyway, it's all pretty much there although for the sake of simplicity I've not tried to make it understand anything other than plain text under version of Perl before 5.7.0. Maybe once I'm happy with the Unicode stuff under later versions I'll look at that.

And then another version that doesn't spew out a crap load of debugging information on STDERR. Sigh.

I must have misplaced the dog health section between them looking at the site and sending me the email:

From: XXXXXXXXXXXXXXXXXXXXXXXXXXXX
Subject: exo.org.uk
To: webmaster@exo.org.uk
Date: Fri, 3 Oct 2003 22:34:17 +0530

I am contacting you about cross linking. I am interested in exo.org.uk because
it looks like it's relevant to a site that I am the link manager for. The site
is about a new site with information on dog health, diseases, problems, and
health care.

I keep the web address confidential and will send it to you only if you give me
permission to do so. Just let me know if it's OK, and I'll send you the web
address for your review. If you approve of the site, then we'll exchange links.

Looking forward to your reply.
    
Sincerely,
XXXXXXXXXXXXXXXXX
XXXXXXXXX@XXXXXXXXXXX
http://XXXXXXXXXXXXXXXXXXXXXXXX
Experts in Quality Link Building
    
P.S. If for any reason you don't want me to contact you again, just email me and
let me know that.

Nice to know they're experts though.

The old version didn't really do sensible things to encode & in URIs with the result that you had odd things in the display of URIs. The new version does.

Updated script and template are available for download.

I got my first spurious Sobig bounce today. I feel so proud. In honour of this I'd like to point you to a plea from someone more afflicted by this than me.

All new and with more dependencies! Although also better dependencies and a much better structure internally. All the details and the download link are on the rss2mail page.

Gone is the old plain text and now we have a database, although as the database is SQLite it's not really that much of an issue. Because of this adding, deleting and editing feeds is much easier. RSS autodiscovery is now supported provided you have the necessary module installed. Diffs on updated entries are now provided in everyone's favourite Unified format.

It's still not likely to work on a non unix platform and the Unicode support is still shonky. The latter issue will be addressed in future releases.

There was no version 1 in case you're thinking you missed something.

To: robot@mirago.co.uk
Date: Fri, 15 Aug 2003 04:09:55 +0100

Hi,

I've noticed that your robot has recently crawled my site and for every
page it's visited it's claimed to have been referred there from a page
on your site. This is incorrect behaviour. If you read the relevant
RFC[0] you'll see that section 14.36 states:

14.36 Referer

   The Referer[sic] request-header field allows the client to specify,
   for the server's benefit, the address (URI) of the resource from
   which the Request-URI was obtained (the "referrer", although the
   header field is misspelled.) The Referer request-header allows a
   server to generate lists of back-links to resources for interest,
   logging, optimized caching, etc. It also allows obsolete or mistyped
   links to be traced for maintenance. The Referer field MUST NOT be
   sent if the Request-URI was obtained from a source that does not have
   its own URI, such as input from the user keyboard.

As there is no link to any page on my site from the URI in question (
http://www.miragorobot.com/scripts/mrinfo.asp ) it would seem to me
that your robot is not following the RFC.

If you wish to provide people with information about your bot then can
I suggest that you use the User-Agent header as detailed in section
14.43 of the RFC.

thanks

Struan

[0] http://www.rfc-editor.org/rfc/rfc2616.txt

No reply as yet.

Given that pretty much every other robot that comes across this website seem to be able to do the right thing you have to wonder why they can't. Of course, assuming they respect it, I can always use the robot exclusion thing to deny them access. Doesn't make it less annoying.

People seem to like friendster. However at least a few people have been disappointed not to be provided with a little graph of their 'personal network'. At least one person's written some code to do this. Sadly it's a windows application and that's considered a bit of a faux pas round these parts.

Fortunately we can fix that.

Presenting the very alpha fr_graph. It's full of holes but it does provide a pretty basic graph of your friendster network. You run it so:

fr_graph.pl -email foo@example.com -pass secret > graph
neato -Tgif -o graph.gif graph

It's got no real docs and the odd unfinished feature but my interest in the problem was only enough to get this far.

Oh, and as it sleeps between website requests like any good robot should it takes a very long time to collect if you have more than a few friends.

People who don't set up their DNS records and web server so that both www.example.com and example.com work. Take the UK retailer Argos. They have a website at, as you might expect, http://www.argos.co.uk/. However, at http://argos.co.uk/ there is nothing. Well, to be exact there isn't an argos.co.uk. There is no DNS record for argos.co.uk. It's not a hard thing to do. Heck, adding something like:

argos.co.uk.           A           129.35.77.19

into their zone file would probably fix it.

And yes, it's laziness on my part to not type the www but I'm not alone in this and it's far more cumulative time for me and people like me than for the DNS admins for argos.

rss2mail is now at version 0.12. Updates include compliance with the RSS aggregator tests, a much smaller memory footprint, and a few minor bug fixes.

If I could just draw your attention to this.

It's not the most impressive thing on CPAN by a long way but it should save someone the time of figuring out the Yahoo login process. Comments/feedback etc welcome.

New things available round here are apache_referers and apache_log_info which is available from the misc code section.

The former is used to produce the referers page for this site. It's a Perl and Template Toolkit kind of thing.

apache_log_info eases the process of simple log file analysis like listing all the requests that have garnered 404 responses.

Both of them suffer from re-invented wheel syndrome but sometimes it's easier to write something that does what you want than to find it. They're being stuck up here as part of my policy of releasing anything I find useful in the belief that someone else will find it useful too.

feedback, bugs and patches for both is welcome.

In one of my occasional pokes through this site's log files I noticed lines like this:

127.0.0.1 - - [04/Jul/2003:08:07:47 +0100] "GET /_vti_bin/owssvr.dll?UL=1&ACT=4&BUILD=3124&STRMVER=4&CAPREQ=0 HTTP/1.1" 404 520 "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; Hotbar 4.1.8.0)"
127.0.0.1 - - [04/Jul/2003:08:07:47 +0100] "GET /MSOffice/cltreq.asp?UL=1&ACT=4&BUILD=3124&STRMVER=4&CAPREQ=0 HTTP/1.1" 404 525 "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; Hotbar 4.1.8.0)"

Oh, I thought, that looks like you're typical random script kiddie probe from some IIS exploit. But then immediately after is a request for a page on my site from the same IP address, which is odd. So, off to google we go and come up with an answer. Ah, it's a Microsoft Office feature.

If I understand it this means that everyone who has activated the discussion bar, which is at least off by default, in IE will now be sending requests to every website they visit on the off chance it has some form of Microsoft discussion software on it. This sort of thing really irritates me. There's a perfectly good HTML tag to allow this sort of thing. It's the link tag. Surely it'd be better to specify some sort of use of the link tag to indicate the presence of the software and save on the 2 redundant requests per server? Or advertise in the HTTP headers that the service is available.

It's just seems like a really lazy solution to the problem.

Everyone says the only way to really experience the centre of Australia is to drive through it. In this spirit we're off to Alice Springs tomorrow to meet friends and then drive to Adelaide. Pictures and the like on return.

If you were to have a look in the code section of this website you will note the arrival of rss2mail (or rss_to_mail, the focus group results aren't in yet). It does exactly what it says on the tin: RSS feeds are transformed through the wonders of Perl into email. Plain text email, pure ASCII goodness all the way.

It requires a string of Perl modules including the badly named (suggestions welcome) Foo2Mail. It's still got a few rough edges and could really do with an easier configuration system (i.e one that doesn't involve editing the configuration file), some tests, and some sort of automated installation but it works for me. Comments, patches, feature requests and marketing strategies to code at exo dot org dot uk.

While it's good that a book can still receive this sort of treatment it does all seem more than a little ridiculous.

And I'm not saying that books are better than other forms of media, more that it's reassuring that books can still generate this level of mass hysteria in our jump cut age of sound bites. 768 pages does seem long for a children's book though, for any novel come to that.

sigh

One of the nicest things about the bicycle is the ease of daily maintenance. Everything that you need to do to a bike to keep it in good running order can pretty much be accomplished in 1/2 an hour with only a cloth, a toothbrush and some oil. Some degreaser helps too. And then at the end of it you have a clean bicycle. And unless you cycle a lot or it's been wet you can probably get away with only doing it once a month.

Given how easy this is the number of people with badly maintained bikes amazes me. What amazes me most is that they must be missing out on one of the main joys of cycling: the silence. Once it's well oiled and clean a bicycle makes very little noise. Some slight noise from the chain, although there really should be pretty much none, and the noise of the tyres on the ground and that's it.

Footnote: Pedros website.

The one thing I have learned in 8 months or so in Australia is that Australian's seem to be pathologically indisposed to closed windows. Or indeed to any substantial attempt to insulate the inside from the outside.

Tim Bray has various thoughts on counting RSS feed subscribers and why this will come to pass and how best to do it.

This is an old discussion that's largely been done with regard to user stats for web pages. There are now recognised ways of identifying unique users and so on. RSS Aggregators are slightly different beasts from web browsers but I really see no reason why the existing metrics aren't good enough, especially as the people that actually care about user figures already use them.

I'm sure I've mentioned this before but one of the things I do with my copious free time is polish my RSS to email aggregator. As part of this I, clearly, have to fetch the RSS from webservers. This, you would think, should be a straight forward thing as HTTP is a nice, simple, well defined protocol. You would be more or less right.

The most recent tweak was to make sure that permanent redirects were handled correctly, i.e. after it receives one the aggregator stores the new URI for that RSS feed and uses it in future. This was all fine until today when it told me:

    Failed to fetch from http://www.example.com/index.rss:
    URL must be absolute [400]

Huh? It's at this point that a brief explanation of how HTTP redirects work. If your browser requests a URI that the web server knows is somewhere else it will return an response code of 300 and something. It also returns a Location header that contains the URI that you are being redirected to. The browser will then go and request the new URI. All this happens behind the scenes and you're probably not even aware it's happened. If you are writing a client you have to write code to deal with this.

Now, if you read the HTTP Spec you'll see that the Location header is an absolute URI. That is, it should include the full URI to the new resource. It's at this point that you realise the cause of the above error: the Location header had a relative URI. A quick poke into the LWP::UserAgent source and the relevant code cut and pasted and all is well.

The biggest problem with standards is not that there as so many to chose from, it's that when people pick one they often don't actually read it very carefully. I know I've been, and no doubt will be, guilty of it.

I both dread and enjoy writing specs for software. The dread bit is the having to write down in plain english concepts that are far easier to demonstrate in code; it's taking phrases such as "and then we stuff it into a database" that make sense to you and actually having to detail what this will mean. The fun side is that when you're, or at least I'm, fleshing out the initial draft quite a few useful ideas usually occur.

Whenever I have to write a spec I generally start out by just spewing everything that comes to mind onto paper or into my text editor of choice. Once I've done that I try and organise it into some sort of structure and translate it all into something meaningful. The first bit is fun and interesting, the second is dull but it's what turns it into something useful as it forces you to think about it how things tie together. Sometimes it lets you see things you've missed, sometimes things that shouldn't be there. It makes you look at the thing as a whole and forces you to think about why things should be done. Or at least it should.

The problem with the second bit is trying to produce something that's useful to anyone who has to implement it but meaningful to anyone non technical, and by non technical I mean non programmers. The latter generally have no interest in exactly how things are to be done but more in what is to be done, how long it will take and so forth. The former are more interested in why they have to do it that way.

That last comment might seem snide but it seems to me that the why you have to do something a certain way is the key. With regard to the code being the photo section one of the requirements is that it has to be written in Perl because it's the language I know and like best and this is supposed to be for fun. It is though, a rather arbitrary requirement. There is no reason why you couldn't do it in Python, PHP, Java or various other languages. Once you know that's why the requirement is there though it's clear that "but if we used language X" is not a valid point. A more meaningful example is that I want the site to be templatable. Now, in order to achieve this I could just whack some perl that prints out XHTML into a file and then just have the code pull that in and process it as perl at the relevant point. The problem with this is that the reason I want to use templates is to enable me to provide a well defined interface between the code and what I display. Evaled files do not do this although they does meet the requirements. Without understanding why templates are to be used you can't make a good decision on how to implement them.

Of course, you also have to pay attention to the what.

With the photo code the spec is aimed at me so it leans more to the well organised brain dump end of things than an easily digestible essay on my vision for a photo gallery application. I am, after all, doing this for fun.

After a pointlessly long delay while I worked up the enthusiasm to get the film developed and scanned some photos of Tasmania are making their way onto the site. More will follow.

As a side note my small sample of four photo developing places in Sydney have only reaffirmed my conclusion that good film processing is not cheap. In the end I went to a cheaper place simply as I had 9 rolls to get developed and scanned and I can apply the GIMP to get them up to a decent enough standard. Plus I can always rescan them when I get back to the UK. Anyway, some of the images might be a bit suspect due to my not so elite photo manipulation skills.

Most galling was the place that made the most fuss about how great their scanner was then scanned the film I handed them at 75dpi rendering the scans useless for anything but the web. You'd have thought that 300dpi was the minimum these days.

When I wrote yesterday about content negotiation I knew there was one small fly in the ointment: clients that didn't send an Accept header didn't get the file I wanted. If there's no Accept header then Apache falls back to the less useful trick of selecting the smallest of the matching files. This meant that for this weblog the RSS feed would be selected over the html version. This seemed to me to not be the expected result.

In order to fix this I first looked at type maps. These are files that allow you to tell Apache what values to use for the matching files, as well as telling it what the matching files are for a URI. I had two problems with them though: you have to provide a type map file for each URI; I couldn't get them to work. The second was the real clincher.

That being a failure I did what every seasoned Apache user does at times like these. I used mod_rewrite. A quick wander through the mod_rewrite documentation and you get the following rewrite rules.

    RewriteCond %{HTTP_ACCEPT} !.+
    RewriteRule (^/.*/[^/\.]+$)  http://exo.org.uk$1.shtml [R,L]

As part of the slow process of improving this website I've been pottering through the recommendations in this W3C document which details various non-optimal website practices. One of the things they discuss is Content Negotiation.

It's not an easy concept to explain so the easiest way is to use and example.

If you have a URI of http://exo.org.uk/colophon (note the lack of a trailing /) then under content negotiation the web server will look at all the files colophon.* and serve the contents of the most suitable one. It uses several criteria to decide which one to serve but most likely it will look at the Accept HTTP header that the client sent. This header lists the MIME types of content that the client can process in order of preference. The server will send the contents of the file whose type matches the most preferred type. If none of the available files match then it will send a default option.

In our example if we have a colophon.txt, colophon.pdf and colophon.doc and the Accept header lists text/plain, application/pdf, text/html we will get the contents of colophon.txt.

This is a bit of a simplification and is based on how Apache handles Content Negotiation. Your web server of choice may deal with it differently. I should point out that you need to make sure the web server knows that a .html file has a MIME type of text/html and so on, although for the common ones Apache generally does know. The relevant Apache documentation has all the information.

Suffice to say that as of now most URIs on this site will not have file extensions.

All the pages on this website are marked up in XHTML and as such I like to try and do the right thing, like using ' instead of a plain '. I'm actually not sure why I did this as a plain ' is actually fine in XML but I did anyway. Chalk it up to Cargo Cult Markup and slap me on the wrist.

The problem is that ' isn't valid in HTML which means that any browser that doesn't understand XHTML tends to render ' as just that instead of '. I'd just assumed (another slap on the wrist) that it was valid HTML. To be honest I'd probably not have cared about it too much, heck I'd probably not have noticed, if not for the fact that Internet Explorer 6, which you'd imagine understands XHTML also renders ' as that. At first I thought it must just be an IE bug, and I still think it is a bug, but then I had a look with a few other browsers and they did the same thing.

So, one quick shell script later and this website is free of ' and only uses '. Sometimes the web sucks.

Being the nice Christian country that it is Easter is something of a deal here. This means public holidays, days of work and oddness of shop opening hours. All this is fine although the shutness of shops here on Good Friday is extreme: much more is shut on Good Friday than Easter Sunday. The policy relating to alcohol sales on Good Friday though is just perverse. In New South Wales, and I can't vouch for any other state here, if you want to go to a public house and consume alcohol in public you are allowed to do so. If, on the other hand, you wish to buy some alcohol and take it home, or indeed anywhere other than a public house, you cannot. This is the law.

Now I can understand, if not sympathise, with the idea that consumption of alcohol on such an important day in the Christian calendar is to be discouraged. That it is to be discouraged only outwith the confines of pubs is just bizarre. Is there a tacit assumption that the weight of public opprobrium will mean those frequenting the many fine pubs and bars of New South Wales will confine themselves to fruit juice and lemonade? Or is it that drinking in a pub is more acceptable than, say, buying a bottle of wine to have over dinner with a few friends?

Regardless it seems to be a particularly arbitrary line to draw. But then so is the one drawn between drinking at half past 10 at night and half past 11.

Well, the new stylee photo section is up and running. For those (and I'm kidding myself here) that have bookmarked any of it then you should update the bookmarks, although you will be redirected seamlessly to the correct location in the new scheme if you use existing bookmarks.

Now I just need to spec up the existing code, fix the few niggling issues that it has, document it, write some tests, some installation instructions and then it'll be ready for release.

And after that comes version 3. Which will be getting a spec in advance as well as some sort of timeframe.

Not only are people getting to the photos section of this site by typing porn thumbs into search engines (mostly msn, although I think that says more about the engine than the users), they're using RISC OS to do it. I guess minority OS users have needs too.

While rambling in an email to rodcorp I had a realization that half the problem with the never ending write a photo gallery system project was that I was coding for myself. The reason for this is that it means I set no real deadline, no definite goals and no planning other than "it should display photos".

This was a mistake

The reason for this being a mistake is that without these things feature creep is more or less unlimited. Unplanned things like being able to attach journal entries to photos or categories suddenly get added in. You realize that the handy feature of listing the last 10 photos added is missing and doesn't really fit in anywhere obvious so refactoring to make it fit occurs. In short it turns into a never ending project.

Having realized this I can now try and do something about it. Principly I'm going to set some goals. The first is that regardless of the shonkyness I should move the existing photo system over to the new one by end of next week (or the 18th as it's called). I've already got a dev site with the new version so I figure that's possible. After that I'm going to try and write a plan of what to do next with some sort of timeline.

Then I'm going to make a big Gantt chart...

We all know that sometimes people get a little obsessed. We all know that genre's that attract an above average geek fanbase are especially prone to this. That an issue of X-Men would be subject to detailed analysis is then no surprise. That they could make the layout so appalling is impressive though.

Well, actually the same design with some tweaks to the CSS so that when you change the font size in the browser everything else changes size.

Only it doesn't work that well in IE 6. As far as I can tell I'm right and IE is wrong and I've not the inclination to work round it.

Oh, and I've not applied this to the photos section. It's not really relevant there for a start.

Thanks to some mendacious piece of software or other the windows machine in this room goes through fits of spreading pop-up adds across the screen with abandon. In what I can only hope is a piece of tongue in cheek marketing among the most common are for software to block pop-ups.

Even so, the logic of the marketing mind defies my comprehension.

Among my favourite public spaces are several of the stations on the Jubilee line extension of the London Underground. Most of them have an airy open feel which is an odd thing given that they are underground. And bits of them are just stunning.

Other things I like are architectural models. The reasons for this are either obvious to you or they're not.

Places I don't expect to find a combination of these two things are Sydney's former Customs House and yet there they were in an exhibition about the Jubilee line extension.

One of the things that interested me about the exhibition was the difference in how the stations where portrayed in the photos and in the video. The photos were of brightly lit, almost antiseptic places, while the video had the slightly grimy look that is usually seen in the "driving round late at night with a hand held video camera" school of music video. Many of the photos made the stations looks like brighter places than they are. The video the opposite. Is this a stylistic difference or to do with the nature of the different mediums? I don't know enough about video to speculate. Both brought something to the exhibition although I prefer the photos. They seem to better capture the spaces.

I've got a few pics of Jubilee line stuff in a Fotango public folder. The resolution and quality of the scans is down to Fotango. The lack of information about the pictures is down to me.

Nice place. You should go. Photos later.

For those that might be inclined to think I only post at night you should remember that while I am in Australia on Australian time my server is in Britain on British time.

I could change the server to Australian time but as it runs a few mailing lists that are mostly used by UK people it would cause more confusion than it would solve.

While I was wandering through the photobloggies nominees I came across a post on how to get your blog noticed which claims one of the key things is getting linked. More bizzarely that "linking content is the foundation of the blogging phenomenon." Which is another example of blog people stating the obvious like it's a newly found and amazing truth. Of course linking content is the foundation of blogging. It's the foundation of the whole Web. You remember that H in HTML and XHTML? It stands for Hypertext, the foldoc entry for which tells us that Hypertext is "A term coined by Ted Nelson around 1965 for a collection of documents (or 'nodes') containing cross-references or 'links'."

For god's sake, it's called the World Wide Web. Web. As in Web of links.

Oh, and a blogroll? Please. Just because it's a list of links to blogs you can't call it your bookmarks?

Despite the small number of photos on this site I really enjoy taking photos. What I don't enjoy is the process of turning small pieces of polyester, gelatin and sliver halide into digital files. What I dislike even more though are digital cameras. Or at least the ones I can afford, especially as I'm a gentleman of leisure.

On the other hand I came across a link to the Photobloggies and some of the sites there made me think I should get back to putting up more photographs. Of course this is slightly confused by the in process development of the Perl scripts that are responsible for managing the photos on this site, which has kind of taken over from the actual taking of photos.

I guess this is a long way of saying that I'm going to try and post more photos again and make more of a point of linking to them from here, at least until I add a journal component to the photo scripts here. Or at least I will when I get back from Tasmania in a few weeks.

I'm not by nature a dog person but this photo, I couldn't resist.

So, the nice people at the swsx festival have released their list of finalists in their website competition, which is nice.

What isn't is that clicking on a random 5 or 6 I found only one that didn't require flash. In the experimental section only http://mc.clintock.com/ does not require flash, and to my mind is actually the best one in the category.

Now I know flash is all very nice and lets you do all sorts of spangly things with it but it always seems like such a lazy solution to me. It screams that the people involved are all from a multimedia background and see the web as a handy distribution mechanism for their self contained little apps.

It seems to me that googlism, or the trackback system for blogs is much more experimental and 'of the web' and hence more worthy of praise, especially as, from what I could see, most of the ones in the experimental category had been done before in some way.

However, most galling was http://www.nikolaskoenig.com/, a photography site that requires flash 6. So, off I go to install the plugin and do I find a site that does all sorts of things plain old HTML is incapable of? No. With the exception of a fade in transition for photos it's all doable in HTML, or at least DHTML. And I think with IE only CSS extensions you could probably do the transitions as well[0]. Just sheer laziness.

Oooh, it makes me mad. All in all it makes me think that the 5k competition is (was?[1]) the only web design competition worth anything any more.

Or am I just old fashioned?

• [0] Not that this sort of thing is to be encouraged but it's a damn site better than the flash solution.
• [1] see http://www.ntk.net/

The slow process of shifting this site over to slightly more thought out CSS has started. As a result some of it may look odd for a while.

Once I can make it work in things other than mozilla I'll be moving to a stylesheet that uses relative sizes for everything which should make the whole site more pleasing to the eyes of those not using the same default font size as me.

20 minutes faffing with awk, sed and touch later I read the scp man page and realise the -p option does exactly what I want.

In case any of you ever read this I'd just like to say hello to the few people a week who find themselves here after typing "porn thumbnails" into their search engine of choice. I'm sorry to tell you that it is merely a quirk of algorithm that has led you here and suggest you move on to the next site suggested. Of course you should feel free to look around first.

While everyone worries about how fast and how much Lufthansa's in flight internet access is going to be no one seems to be asking the most important question: will I be able to plug my laptop in to a power source? Given that unless you are blessed with some form of recent Apple laptop you'll be lucky to have a battery life of more than two hours, especially with the insane power drain of Wi-Fi, having power sockets will be much more important. Of course they'll probably only be available in business class. For all I know they may already be.

And while it's very nice that it'll be available throughout the flight you have to wonder how many people in the less than capacious confines of economy will use it. Having tried it, the use of a laptop in economy seating is not much fun.

For they have saved us from the evil do-ers that wish to stifle creativity by making sure that copyright only applies for a limited time. Thank god for Jack Valenti and those that follow him for they have made sure that Disney and friends have no worries about releasing such original and creative works as Snow White and the Seven Dwarfs or Steamboat Mickey into the public domain. That secure in that knowledge they can continue to innovate and create original new works like Treasure Planet and Hook.

And we should thank those people who long ago realised that the most import right in America is that of funding your Congressman to the hilt in order that he can devote all his energies to thinking long and hard about the issues that matter in the world today.

But most of all, Thank Fuck for Democracy and the right to make a difference with your vote.