threads bared

A neat idea from Sean of voisin.org: since joining the 9rules community is bringing in a group of new readers, why not provide links to some previous posts that I think they might like? OK, oldest first:

• terminal half
• 34 Eek!
• auto-alanisation
• refer unto others as you wish others to refer unto you
• the february of technology
• that friday feeling
• a new hope?
• cheesed off
• a grey day
• i know the feeling
• eternal sunburn of the witless mind
• re-progressive
• sonification
• many scars
• a common search
• how many webs do we weave?
• weblarney
• going south
• rebooted
• notebirthday
• schmap flap
• how fair the realm
• think agnostic – act atheist

I also have another blog I try to keep updating, one with a specific theme: found poetry.

new links

The move to a new site, still going on at the moment, is going to cause some disruption to other people, too: those who had set up direct links to items on this site, mostly images used on other sites. People have been taking some of my work and using it directly from their pages. When someone loaded their page, it included a direct call to my site for the image, which then appeared on their site.

I didn’t mind this, because the amount of traffic didn’t exceed the limits I was paying for, and it was interesting to see all the MySpace users with my pictures as their backgrounds. Well, that is still possible, but I don’t have the same “flat” image storage that I used to have, any more. If anyone wants to keep doing that, they will have to go to the image gallery, find the picture they want, and get the new link from there.

I haven’t yet figured out what to do with the “b3ta bodges”, the modified images and animations i put together for b3ta.com. They’re mostly animated GIFs, which Picasa Web will not allow me to upload, probably for reasons related to the patent on the LZW compression used there – even though there has been no need for concern for two years now. Lawyers…

I may up putting them on a static page, uploading each file to WordPress individually, but that will not be quick, and will not put the pictures at the same locations as before.

ancient history

I’m taking some time, slowly, to upload some older blog entries, so don’t be surprised to see posts from 2004 popping up. It’s a manual process, cutting-and-pasting HTML from flat files in to the WordPress HTML editor. Timeconsuming, but I don’t mind, since I can vet the posts as I go.

I could automate this with some kind of HTML -> XML Translation, but I’m leaving out some old entries that look out of place or embarrasing: yes, I’m not above a little personal revisionism.

rebuilding

Last night, after Matt from WordPress kindly fixed the Domain Mapping to point here, I bit the bullet and switched over, by taking out the DNS entries that pointed at my previous web hoster. Here’s a summary of what has changed and what hasn’t:

• Old blog entries are not yet transferred over: that will only happen after I can upgrade the old site to WordPress 2.1, the first version that will support exporting in the XML format used for importing into WordPress.com. (This is a new feature pioneered on WordPress.com, probably because the administrators didn’t want users of the hosted system to have to deal with the MySQL database format used there.)

As before, images used on this site fall roughly in to two categories:

• Incidental images added to normal posts and pages: these will use the upload process provided with WordPress
• Photographs and other large images: these go to the Picasa Web Gallery referenced to the right, from where they can be linked in to posts and pages, or viewed in place.

In both cases I will need to re-upload and relink them all, but I will be using the gallery more. I’ve found Picasa Web so useful, and quick, that I’ve already uploaded all the photographs I had online before, in a matter of minutes, yesterday. The captions are missing, I will fix those over time.

I’ve yet to see if ZIP files are supported for upload to WordPress directly. I suspect not, because a ZIP file can contain anything, and could therefore be used for piracy or other nefarious purposes. There’s a general restriction on file hosting through free services for those reasons, even for little sound files such as the samples I put up for download. If it doesn’t work I will see what I can do, or just do without them.

domain name sins

We’ve had a couple of days during which this site was unavailable, and though it was planned, it wasn’t supposed to go on this long. To cut a long story short, I made some changes to the Domain Name System (DNS) settings for this domain, to accommodate a potential new hosting service for this site. They’re offering it at no charge to the first few people who apply, but they haven’t come back with any more details, not even a yes or no. The outage could have been avoided had I done, at first, what I did to get the site back up. I’ll cover a few technical details now, for anyone interested and for my future reference.

When anyone tries to access stereoroid.com by name, that name goes through a DNS name resolution process that converts it to an IP address, so that the actual communication by IP can take place. I was asked to change the DNS servers registered against my domain, to point at the new hosters, who would then change their servers to accommodate my site. I made the change, which means that other computers, when asking “where is stereoroid.com?”, were told “go and ask these new servers”.

Unfortunately, the new servers have not been told to expect this, which is out of my hands, so the DNS requests were being met with a resounding “Huh?”. What I’ve done now, and what I should have done in the first place, was to put the DNS servers of my current web hosters in to the list for my domain, as backup for the new ones. I didn’t think of this earlier because I thought I could only register two such servers, but it turns out I can do many more. After the new servers go “Huh?”, the request now goes to the old ones, who say “yep, he’s here”.

So, we’re back up, and I can bide my time while the guys on the other side to get round to making the changes they need to, or at least letting me know if I’m included in their tests. If I am, the site will look very different at first, and I’ll need to do more work on it and transfer the data. I’m doing it, though, so that I can hand over most of the technical management behind this site to people who are much, much more expert at the software I am using now: the people who wrote it. There will be some loss of flexibility, but I can live with that. It will also mean major changes to the photo gallery, but I can live with that too.

schmap flap

Schmap have an interesting business model: they take publicly-available material and fashion it in to travel guides, and use this in the marketing process. I only heard of them because they borrowed a photo of mine from Flickr, one that I had marked as available for commercial use under the Creative Commons license, and used it in their Dublin guide. This is something like what I had in mind, so no complaints on that score, as long as they provide proper attribution (which they have). So, grab a copy of the Dublin guide, and see some more of my work.That’s all – enjoy!

divides

I take the previous post back: after playing around with DIVs for a bit longer, I found that the problem wasn’t so much in the use of DIVs in general, but in the way they had been implemented in the WordPress Theme I use. I didn’t create that one from scratch, but modified the “Original WordPress” theme. It had a lot of unnecessary tweaks, such as squeezing the text together, which might have given it a different look but had a negative impact on readability.

I’ve also changed the primary font to Palatino Linotype, but that may not last.The logo over the picture is achieved by using another DIV with the picture as its background, and you can expect the picture to change over time. I’ve also resurrected one of my favourite little PHP tricks: the script to calculate a new light colour on each page load, this time applied to the Sidebar background.

Apart from the known compatibility issues with using tables for layout, I also found it had a negative effect on the way the pages were being displayed. The reader would be stuck with a blank page until all the table content had loaded; because I did not fix the width of the left-hand column ,the browser had to do some calculations on the content before displaying it. I definitely do not want to fix the widths, since I have seen how badly that works on other sites, especially when viewing them on small screens.

The DIV method is a pain to manage if you’re not used to it, but the way it works is fairly straightforward in practice. The main body of the text on the page is in a DIV without “absolute” parameters, just some tweaks to the margins, so the browser does not have to work too hard to lay it out as normal.The right-hand margin is much larger, to allow space for the sidebar, which is given an absolute position at the top right, and a fixed width. If the browser does not understand this, it displays the Sidebar at the bottom.

I can see this happening on limited Internet Explorer browser on my old iPaq (which has survived its near-death experience). It’s not all good news, though, since the margin is recognized and takes up half the tiny screen. More experimentation to follow, possibly involving a script that leaves out some formatting user agent parameters. It’s all go here…

theme scheme

Expect the look of this page to vary a bit in the next few weeks. I’ve finally gathered up the willpower to strip this site’s WordPress theme down to its very basics, with the aim of rebuilding it in my own way.

To start with, I’ve re-engineered the main page template to look similar to the old one, but without the “absolute div” positioning of the sidebar that had me worried about compatibility with older or limited browsers; it’s a recent development of style sheet support that I’m not terribly keen on.

To get the similar look, however, I’ve had to use an older method that I don’t like, either: the use of Tables for page layout. It also has problems with older browsers, though careful use can reduce the impact of this. Switching themes is easy, so I can return to the previous theme if I feel like it.

In the long term, I think I’ll have to find a better way, but I can take as long as I like. What I will definitely avoid, however, is any kind of “Web 2.0″ bull, especially the kind that relies on client-side scripting to work. There are JavaScript components on my site, such as the Flickr Zeitgeist pictures, but they’re optional. It’s going to take a lot of research, I think.

talk talk

I thought I should just do the whole SQL update thing in one lump, not by record and it appears to have worked, but it’s not something I will try too often, because it wipes any data that changed since the backup was taken – comments, etc. I only did it this time because I made several offline fixes to blog entries in the database, spelling & grammar etc., too many to repeat online. Other changes I plan will be to styles and templates, and those are flat files I can simply back up and upload without changing the content. The important thing is to have the mechanism, and an accurate offline copy of the site that I can tweak.

Last night I tried to “clear” an old videotape with a couple of movies I recorded to watch later. First up: The Outlaw, directed by Howard Hawks and produced by Howard Hughes, from 1943. A largely forgettable cowboy tale, of how Billy The Kid (Jack Buetel) walks in to the wrong town, offends Sherrif Pat Garrett (Thomas Mitchell), a former outlaw himself, but is defended by Doc Holliday (Walter Huston) for no apparent reason. After tensions escalate, Billy is shot, but Holliday carries him to the home of his mistress, who nurses him back to health and starts an affair with him – despite the fact that he killed her brother, and she was Holliday’s mistress.

This mistress is largely responsible for the film’s notable status today: Rio, played by Jane Russell, displays a monumental cleavage, allegedly with the assistance of specialized hardware designed by Hughes. It had the censors in a tizzy, and didn’t see the screen until 3 years later, in a cut version. Apart from Russell’s assets, however, the film is tame, and I didn’t finish it, because I had a hopefully better one on the same tape.

The Conversation (1974) was Francis Ford Coppola’s “pet project”, the one insisted on making before he would agree to The Godfather Part II. It was getting late, and I thought I might drop off to sleep, but no chance of that; it’s a psychological drama of the first order. Looking for reviews afterwards, I found Leonard Maltin calling it “one of the greatest films of the 70′s”, and the link above is to Roger Ebert’s second review of it, this time for his Great Movies list. Me, I’ve long been an admirer of Gene Hackman’s acting, and this film might just be his finest hour.

experiment sequel

Now this is dangerous, for me at least: an attempt to update this site in a totally offline fashion. The steps involved are simple enough in theory, but carry the potential for disaster.

I now have a replica of the live site on my notebook computer, a task complicated by the fact that WordPress, on which this blog is based, uses a MySQL database backend. You can’t simply copy that data over from point to point; instead, you have to export (dump) the data from the database to a text file, and reimport it at the other end. The import is a one-shot process, but before you do that you usually have to delete all the current data from the database.

Or do I? Actually, now I think about it, perhaps I can do something with the single record that will be created when I save this blog entry, by carving it out of the SQL backup. Hmmm…

entering and breaking

I’ve only once had someone contact me about a broken link on this site, but that doesn’t mean there haven’t been any. Today I took a little time to go through the referer logs, not just to pick out the weird referers, but mainly to analyze who or what is using this site. I also fixed a few links that had been left broken in the recent changes.

The majority of hits on stereoroid.com are from search engines. Not all of them are marked as such, and some supply false or missing browser details (e.g. from the UNIX wget command), but you can easily tell them from the access patterns and timings. The number of direct cracking attempts are small; two in the month of March, attempting to access standard mailing scripts by name. Not gonna work here, folks.

Comments are another story: disabled for the moment, until I can investigate further, perhaps defeat the scripts by renaming the files and links. I can see only one attempt at automated comment spam this month, but I can’t say the same for Referer spam: I can see many hits with the same referer, but since I don’t have the referer logs viewable publicly, it’s pointless – no links up for search engines to jump on.

I’m not the only one being hit by this same spammer, who I’m not going to link to, but search for andrewsaluk spam for many similar reports from others. Sounds like a amateur script kiddie using some crude “search engine optimization” (SEO) methods. If the culprit really is someone named “Andrew Saluk”, he’s doubly a moron for using his real name. I’ve just modified my htaccess file to block any referers with that name, as suggested at candygenius:

RewriteCond %{HTTP_REFERER} andrewsaluk
RewriteRule .* - [L,F]


Let’s see what effect that has. From Whois I can tell that the Domain is registered to a Florida address, but the server is in St. Petersburg, Russia. It appears to be an online Texas Hold-Em site: abandon hope, all ye who enter credit card details, unless you actually intend to bankroll the Russian Mafia. Other reports suggest it’s hosted in China; perhaps it’s multiple servers, or moving around.

About six months ago I found myself getting interested in playing poker, perhaps even Texas Hold-Em. The game has been the subject of major publicity recently, with high-stakes tournaments in Las Vegas and on board ships, and what appears to be a gentle learning curve through low-stakes games. The idea was that I could round up a few interested people here and play for chips or pennies. Sadly, two things have served to almost kill what interest I had:

1. the way the game is presented as predatory and antisocial, with a strict hierarchy of predators and prey. There are many people trying to get in, but gaining skill carries a price, and there are enough people skilled at reading your tells, waiting to take your money off you. Hardly the social fun I had in mind;
2. the rabid commercialism surrounding it, especially online. It seems like everyone has a scheme to make money off the game online, but what’s the point? If you can’t try to read your opponents twitches – the social side of the game I was interested in – it becomes just another card game, as random as Blackjack.

Do you trust a strange server to be programmed to play fair? Who’s to guarantee that one of your fellow players is not also running the server, and can see your cards, or control what you’re dealt? To me, the game only makes sense when you, and you alone, can watch the cards, and the people, in person. Why would you play it online – in the hope you’re going to win money? The money pulled in by the likes of “partypoker” should be enough to discourage anyone from getting involved.

Lastly; for some interesting reading on the relationships between SEOs and online gaming, including (allegedly) some established names, have a read of this.

buzzwords-a-go-go

Or: how my experience of information, including this site, has been influenced by recent trends in web interaction. I’ve been using buzzwords without a proper explanation what they do, so here goes:

RSS
If you follow the “rss” link in the right-hand sidebar of this page, you will be presented with a page full of hairy XML code. This is not designed to be read directly by humans like you (assuming), but if you insert the link to that XML into a program called an RSS Aggregator, it is converted into a readable “feed”. There are various RSS Aggregators out there, including some web-based ones, so you don’t have to install software on your computer if that’s a problem: see the following blogspace page for some background and lists of readers.

The advantages of this approach are many, but the main one to me is attention conservation: if I change something on this page (like this new blog entry), or another feed changes, it is flagged as new, so you can read it in your Aggregator. Until that happens you don’t need to visit the site hosting the feed to check for updates. So, when I do this with multiple feeds – and I have dozens set up in FeedReader myself – I have access to updated information from many website sources, as it arrives, neatly categorized.

Depending on how you use RSS, you may find you prefer a “full text” feed, like this site provides: I let WordPress include the full text of each Blog entry. This works well if you read content offline, as I sometimes do on the iPaq (a process I’m currently re-examining). Other sites assume you have an internet connection available as you read, so they include only a headline, perhaps a short summary, and a link to a live site where you can read the rest. This is a choice made by the website owner, you have no control over it, but I find it affects what I read, or not. I always want a full-text feed – why not, for the little extra bandwidth involved?

Podcasting
For the last couple of months I’ve been enjoying Podcasts from various sources: but what is a Podcast? It is simply an audio file created by someone, and made available for download, usually in MP3 format.

What’s new about that? Hasn’t that been done for years? Yes, but what is new is the delivery mechanism: it piggybacks on to a RSS feed, as described above, with all the same advantages. It means that, if you want it, the MP3 file will be automatically downloaded and saved to your local disk, using an “enclosure” extension to the RSS standard.

Note that none of this actually requires an Apple iPod: Podcasting’s current popularity can be attributed to the success of the iPod, but it’s not required. I listen to Podcasts on my iPaq when commuting, and on my laptop at home, as I would with any MP3 file. iPods apparently have a mechanism by which MP3s dropped in a specified directory are automatically copied to the iPod: I just include the Podcast MP3s in the data I back up to take to and from work, and the iPaq reads them straight from the card. My PC Podcast client of choice is iPodder v2, which also comes in a Mac version, and the main Podcast directory is iPodder.org. That site also has a an explanation of the process: “what is podcasting?”.

The latest Podcasts are logged at audio.weblogs.com, but a nice place to start is with Adam Curry’s Daily Source Code, a semi-professional Podcast by a former radio DJ and MTV VJ, who produces an almost-daily show that comes close to radio, but with a more human side – bloopers, hardware problems, swearwords, his daughter wandering in and out – great fun. Because of the time-sensitive nature of RSS and Podcasting – new Podcasts are brought to your attention like RSS feed entries – there is a motivation for Podcasters to produce fresh new content regularly – which is part of the fun.

Since Podcasting uses RSS, why do we have separate RSS Aggregators and Podcast clients? Simplicity, probably: I tried out Egress for the iPaq, which did both at the same time, so we know it’s possible. I can see more such convergence coming in the future.

Next time I will talk a little about Tagging: a subject I am barely getting my head around. I know how it works, but the why is still a little unclear at this point. Later.