[insert witty title]

I have been planing on doing this for quite some time now, but it has finally happened. [iwt] is back. Over the past couple of months I really have missed writing on the internet every day, but really wanted a break after having done so for almost two years.

Now I am refreshed, ready to resume twatting about on the internet and aware that nobody reads this page any more!

Since I have a bit of spare time today I started by replacing the horrible header image. And then cleaned up the admin side of the blog, which got seriously messy over the past two years of installing every single plugin I could find, and never bothering to delete crap that I don’t use. In doing this I stumbled over the Search Stats. Every time somebody searches for something on the blog (the box to the top right of your screen) it keeps records. Here are 50 recent searches, I have bolded my favourite search terms:

smallville, lock, sex, layered liquids, bears, “i did not draw”, research paper, research paper, draw, draw, photoshopped pictures, photoshopped pictures, the shield, deformed cat, BigDog, tiger, science projects with house hold items, advert, muscular girls, yoghurt, what american state ends with three , trailer park boys, trailer park , corn , corn flour, knox cap, fishpondbubble07, shopper discounts, shopper , 24, cornstarch balls, heroes, asimov, easyjet, hero fountain, raptor jesus, blind cat, wikipedia, wikipedia, shopperdiscount, trinity, human weapon, plot, lethal weapon, urban exploring, time , Funny Farm, podcast, map, FOOD TEST, springer, hero’s fountain, mark steel lectures, mark steel lectures, mark steel, mark steel, tremendous kitten, liquid, liquid, layers, WHICH AMERICAN STATE HAS A NAME THAT ENDS IN 3 VOW

[iwt]: Your source for information on deformed cats, tremendous kittens and muscular girls.

…come to think of it I’m going to add that into rotation for the random taglines.

Browsing through old posts on the blog yesterday evening I noticed that one particular page was showing this advert:

The claim seems pretty unambiguous: I can achieve the ultimate qualification over the internet. Now, being a complete and utter nerd I couldn’t help but wonder exactly what the ultimate qualification would entail. After all for something to be ultimate it has to satisfy at least one of these definitions:

1. maximum; decisive; conclusive: the ultimate authority; the ultimate weapon.
2. highest; not subsidiary: ultimate goal in life.
3. basic; fundamental; representing a limit beyond which further progress, as in investigation or analysis, is impossible: the ultimate particle; ultimate principles.
4. final; total: the ultimate consequences; the ultimate cost of a project.
5. not to be improved upon or surpassed; greatest; unsurpassed: the ultimate vacation spot; the ultimate stupidity.

Really, the online degree must bestow upon the student some sort of profound philosophical knowledge about the universe, otherwise unreachable to mere mortals. It must be taught by God himself, who lets you know everything that he does, so that no further qualification is even possible.

In short, Liverpool University Online are claiming to turn you into a veritable Man-God.

…or they meant ‘ultimate’ only in the figurative sense.

Browsing through the comment moderation queue I noticed this message had been tagged as spam:

Name: ff7c30d5t19aba0a8e9c
URL: ff7c30d5t19aba0a8e9c.info
IP: 72.22.71.65
Title: ff7c30d5t19aba0a8e9c…
Body: ff7c30d5t19a…

Does anybody know what on earth this could be? The URL doesn’t point anywhere, so as spam it is completely useless. I have seen a few of them over the past few months.

My only theory so far is that the spam bot posts random strings on millions of blogs then google searches for them a week or so later. If he gets hits then he has found a site with no spam protection and continues to post.

Anybody have any better ideas?

There already exist PHP scripts that can import b2evolution databases into a Wordpress blog, but I personally did not find them to be a satisfactory solution. This is due to the fact that the version of the PHP script you require depends critically upon both the version of b2evolution you are running and also the version of Wordpress into which you are importing (the script has currently not been updated in nine months and is incompatible with all recent b2evolution installs). Additionally it is necessary to install b2evolution and Wordpress into the same database, which is not easy if you are, for example, importing your posts into an already existing blog, or moving site entirely. Finally I don’t really understand databases and as such don’t feel happy in running a script I found on the internet to make massive changes, since if something goes wrong I do not stand a chance at fixing it.

Here I present a method that should work on all versions of Wordpress and all relatively recent b2evolution installs. You will not need to mess around with databases since we use an alternative approach. We are going to skin b2evolution in such a way that if you look at the source of a webpage it appears in a simple format that Wordpress can import, the Movable Type ASCII export format. With no further ado here are the three easy steps you need to perform to move your posts and comments from b2evolution to Wordpress:

1. Apply a new skin to b2evolution

BACKUP YOUR CURRENT SKIN. I’m not responsible if you forget to do this and then change your mind about moving over. Copy the following files over your current theme:

replace _main.php with the contents of _main.php.txt
replace _feedback.php with the contents of _feedback.php.txt

If you now view the blog in a browser it will look like a bit of a mess, but if you view the source of the page (click View > Page Source in your browser) it is in Movable Type ascii export format, which you can import directly into a new (or old) Wordpress install!

2. Download your old blog to text files

I found that the easiest way to grab a whole lot of posts at once was to view the archive page for a whole year, with the display of comments enabled. For example if your blog resides at http://www.mydomain.com/blog/ then the page that will show you all of the posts and comments for the year 2007 is at:

http://www.mydomain.com/blog/index.php?m=2007&c=1

View the source of this page (View>Page Source in your browser) then copy and paste it into a text file. Repeat this step for every year of posts you need to import.

3. Import the text files to Your New Blog

Wordpress

New versions of Wordpress can import directly from the admin control panel, just click “Manage>Import>MovableType and TypePad”. Upload your text files and click import. Older versions of Wordpress already have well written instructions on this page.

Typepad

I have heard that this works, but do not know how to do it…

Bugfixes For Older Software Versions

b2evolution 0.9.0

“author()” isn’t recognized by this version (or at least, the blog I am working with), but if it is replaced with “Author->prefered_name()”, everything works out fine! (thanks Layne)

Questions, comments and criticisms are, as always, welcome below.

I decided that the new blog now contains enough functionality to go live. It’s not quite finished, particularly in the aesthetics department, but it is functional and it means that I don’t have to support two different blogs at once!

Pros:

• Pretty new design (I spent most of the evening working on this and am very happy with it)
• Popular Posts page, makes automatic league tables for the most commonly read/linked to/commented on posts
• Share Me button is really fancy (see below each post)
• Comments on the miniblog

Cons:

• Importing the b2evolution posts led to the loss of all but the primary category for each post. The categories to the right are therefore no longer very accurate and need correcting
• The template isn’t finished yet! (titles on archive pages are wrong, links are wrong colour, no blogroll, categories and linkblog entries look messy, spacing wrong in writing beneath each post)
• Using wrong RSS feed!
• My oh so hilarious random headers are gone for a while

After my last set of search engine ‘optimizations’ Google has pretty much stopped sending visitors my way. I’m pretty sure this is because it can’t find the individual post permalinks (I stopped the google bot from crawling the archive pages since it seems to find them more relevant than individual posts). Luckily there is a way around this problem in the form of sitemaps. A sitemap is an xml file that tells Google what pages are on your site and how important each one is.

Unfortunately I can’t get the automatic sitemap generator to spider this site because all of the pages are dynamically generated, and I can’t find a way to get a list of post URLs from b2evolution. I got around this by following JEG’s lead and using Python to make a little script that trawls the archives and dumps all of the permalinks to a text file urllist.txt. This text file is then fed to Google’s sitemap generator, which sends an XML file to Google to let them know know what pages I want it to index. This process is performed by the server every midnight, and fingers crossed it should guarantee that Google finally indexes us correctly!

Just in case anybody is interested, here is the code that creates urllist.txt. It’s a bit clunky because it’s the first thing I’ve written in Python!

from mechanize import Browser

———postlist_file.write(link.absolute_url+’ changefreq=never\n’)


(swapped tabs for — since html doesn’t seem to like preserving whitespace)

—–

Bit more housekeeping. Firstly author names are now displayed prominently at the top of each post. Hopefully a couple more people will begin posting here, and now it’s clear who wrote what.

Secondly looking through my statistics I noticed that Google had indexed the site incorrectly! When Google finds multiple copies of the same text on a site (for example on this site the same post appears on its own page, on the daily archive page, on the monthly archive page, on the yearly archive page and on the category pages) it takes a guess at which copy is ‘most important’ and shows that in search engine results.

The problem I had with this site was that Google would see a post at a url like this:

sciblog.php/2007/04/27/weightless_wonder

and also its archived copies at

sciblog.php/2007/04/27
sciblog.php/2007/04
sciblog.php/2007

the ones with the ‘most important’ looking URLs were unfortunately the archive pages, and as such the vast majority of Google visits to this page land in the archives. This is obviously somewhat less than ideal as it is quite hard for people to find what they’re looking for when confronted by a single page trying to load everything we wrote in the previous year!

I have fixed this by removing dates from the URLs of individual posts (see this hack, or rather my modification in the comments), so individual posts now reside here:

sciblog.php/weightless_wonder

and also placed a robots.txt file in the base of this site containing:

User-agent: *
Disallow: /blog/sciblog.php/2006/
Disallow: /blog/sciblog.php/2007/

Disallow: /blog/sciblog.php/2008/
Disallow: /blog/lifeblog.php/2006/
Disallow: /blog/lifeblog.php/2007/
Disallow: /blog/lifeblog.php/2008/

This should stop all robots spidering the archive pages and instead force that they notice the individual posts!

Now if I can just get the webserver to hide the .php part of every URL I’ll finally be happy!

—–

After writing on this blog almost every day for an entire year I’m feeling a bit burnt out and completely lacking in the motivation and enjoyment I used to get from looking after this site.

Therefore I’ll be taking a couple of weeks off. See you when I feel like writing again.

p.s. if anybody feels like making a guest post get in contact with me

(insertwittytitle . at . gmail . dot . com)

Back home now. No time for a proper post, see miniblog for a few updates —>

—–

I have just spent a little while hacking the blog software to behave in the way I want. The net result is that one of my favourite features from yesteryear has returned: The Contemporaria (now called Miniblog). This is a list of interesting links that I find around the web, and provides a way of highlighting interesting links/images without having to go to the trouble of writing a whole post about them. The archives from the old contemporaria are available here and here

The main reason that the old version was discontinued was that it was a surprising amount of work to update and archive each link:

Type up address and description, along with the date as an html list entry -> Edit blog template -> Add extra row to contemporaria list html -> remove last entry -> find and edit contemporaria archive post -> add link to top of post.

OK so it’s only a minute or so each time, but the point of the contemporaria is to share links that catch the eye and may be interesting to others. They’re like the equivalent of a throwaway comment or snappy aside in real conversation! The new one works like this:

Paste address and description. Press button.

Hopefully it’s simple enough that I’ll actually bother to use it! If anybody has any links that may be suitable let me know!

TODO: Link to contemporaria archive; get dates in proper (i.e. non-American) format, and without year.

Gemma has been working hard today so I have had a whole load of free time. I decided to spend it creating a frontpage for the site, http://www.geometricrate.com , porting in a bit of the old content and adding a bit of extra functionality to the blog. The frontpage needs a bit of polish but I’m sure it’ll be looking nice in no time!

—–

Just a short post for now. Last night I did a bit of housekeeping around the site and collated together a few old posts into collections that some people may be interested in reading:

Amusing crimes, crimes that people deserve to get away with.

Terrible reviews, reviews that made me smile. Collected mainly from Amazon and eBay.

When I have a bit more free time I do intend to continue with these post series.

—–

Sorry about the downtime for the past few hours. The machine the site is hosted on has recently been heavily upgraded and for one reason or another this site reacted badly to the move. The conflict was fixed incredibly quickly by my hosts who ironed out the kinks in the database within twenty minutes of me letting them know something was wrong.

late edit: I have additionally placed a placeholder page on http://www.geometricrate.com as I don’t know when I will actually have time to put some proper content there

—–

The great spam battle continues, today I have:

• Disabled trackbacks since a spambot sent a trackback to every single post on this blog. It’s going to take ages to moderate them all
• Removed comment moderation. You now have to answer a simple question to post

—–

Only took a few hours to get this whole thing set up, I’m sure I have missed something so if parts of the site don’t work, let me know what they are!

I’m currently aware that:

1. In IE there is a big gap between the first title and the text (no bloody idea why this is. My work PC running firefox also shows this error) only seems to appear on old posts. Ignore for now.
2. The title and tagline are above the title picture, instead of inside it. fixed
3. Blogs have horrible names (/index.php?blog=2, /index.php?blog=3). Fix these to more sensible things (/iwt.php, /science.php) blogs are now at index.php (UnifiedBlog), sciblog.php (SciBlog) and lifeblog.php (LifeBlog)
4. The geometricrate.com homepage needs to use b2evolution as its CMS. geometricrate frontpage is now a blog of its own. I’ll work on getting it looking like a more static site soon

5. Anti-spam measures on the comments. Comment moderation is now on
6. The background is buggered up on the work computers fixed. the background png was not wide enough for high-resolution screens.