drupal

...for a few days. As will responding to peoples' comments.

A couple of months ago I bought myself a new great big server and shamefully I have not set it up yet. Since it is priced in dollars and the pound is falling I suppose I ought to get on with it so I can cancel the existing one before it next needs paying!

For anyone interested it will be Debian Etch, running Xen virtualization, to give me a Zimbra virtual server for email and collaboration, a web server for my various projects and then back end servers for databases and user data.

I really need to redesign the blog, and in the process move it to my new domain jockcoats.me and upgrade to Drupal 6.

Additionally, a number of projects have been languishing waiting for this shift to the new bigger server:

OX3Online - a project to produce a community portal for the Headington area of Oxford

LiberalALTERnative.org to accompany the book on economic liberalism I am co-editing with members of the Lib Dems ALTER executive

OxfordBloggers.net - an aggregator a little like LibDemBlogs to link together as many bloggers writing in or about Oxford

OSEF.org.uk - a new site for Oxfordshire Social Enterprise Forum which we intend to relaunch in November's enterprise week

...and my latest wheeze...

f5c.org - "Freedom's Fifth Column" to provide a space in which libertarians (especially those hiding within existing non-libertarian parties) can write, pseudonymously if necessary, to try and show how libertarian and anarchist ideology can work through most existing parties to achieve our freedoms.

Lots to do! But don't worry, the suspension of blogging is only in order to give me a few days while I am off work this week to get the server up and running - these other projects have to work alongside my own writing...:)

As you will see, I've redesigned - again!  Following my Christmas make-over one or two readers felt the whole thing was too complicated.  And in the meantime I discovered how to make shiny buttons that should expand properly with the content.  So here's the result, nearly.

As with last time, I'm going live with it at the "80% complete" point as getting every little bit right will take quie a bit more tinkering.

As a Mac user I've done little testing of it in Windows with Internet Exploder, but I have noticed running it in my Parallels Windows session a few weirdnesses/differences from the Mac/Firefox version.  I will try to address these over the next few days, but if you have particular problems with aspects of it, let me know in the comments, by email or using the contact form link at the top of the page.

I'm, off now to the big smoke for the evening - I hate going to London, but the company and the thought of my first ever visit to the National Liberal Club have outweighed by London specific agoraphobia for now!

Just a couple of weeks after getting back into full swing blogging frenetically after redesigning the site it's likely to be another slowish week ahead. Mark Wadsworth thinks the new design is a bit confusing, and, whilst it appears to have achieved some of its aims in keeping more readers on the site to look at other stories, I tend to agree it's not been ideal. But I learned a lot in the process and so am starting again and will hopefully take a lot less time to rebuild, this time with more bling. And I've got a busy week ahead, with lots of conspiracies beginning to take off for me this week:  read more »

Though this is not a techno-blog, I do do a fair amount of work behind the scenes myself in php, css and html to get it looking like it does (not a winner I notice!). But one thing has bugged me for some time...

The observant amongst you may have noticed that it is developed using the Drupal content management system. This has lots of add-on modules contributed by nice people all over the world. One of these is supposed to process "trackbacks". You'll have seen these on others' blogs - where if someone on another site posts something about a story you've written, you get a magical little line saying "Such and such a site or blog refers to this post".

Well with the Drupal trackback module this only appear to work if people have explicitly "pinged" the trackback URL for the post of mine they are referring to. And nowadays it seems that very few people actually do this. When I post on my blog it does ping each site I've linked to in the posting, I think (it certainly does at least the first site I refer to) without me having to explicitly tell it to.

So others see when I have referred to them and can put a link back in to my blog if they want. But I don't see automatically when they do, so they don't get a link back to their post underneath mine. But I know I am being linked to - I've been mentioned in the Lib Dem Voice "Golden Dozen" a few times, and in the Brit Blog Round-up and so on.

The trackback module claims to be able to discover a site that is referring to mine just from someone clicking on the link in another blog's article that refers to a post of mine. But it doesn't appear to work. So, while you've all been to conference, I've been trying my hand at rewriting that module to enable my blog to discover when someone refers to my blog just from click-throughs. And that's why I've been "blogging lite" this week.

I've not actually got very far yet. Just setting up a debugging environment took long enough. And now that I sit down to think about how to do it, I find it much harder than I first thought. I can see why the module authors have not implemented it yet! For example - how do I distinguish between a referrer that is a search engine results page, a "real" article that refers to me, or just a "blogroll" type link in a sidebar. Incidentally - I notice that Lib Dem Voice picks up these sidebar links, as I've seen my pages listed as referring to some of their articles when actually they seem to have picked up the feed from the Lib Dem Blogs aggregator I put in my sidebar.

However, if you've stumbled on this because you are also looking for this feature in the Drupal Trackback module, you might be interested in having a look at the logic I think I have finally settled on in the following specification to achieve this:

Note: I am going to do this as a Drupal cron job in the trackback module using entries in the Drupal access_log database table. This is because the processing of each one to check whether it is a real "referral" or a search engine page or a link outside an article like a blogroll will take a little processing time for each page load if I do it when each page is requested by a user agent. So I guess I'm doing something similar to what Technorati does when it indexes your site when it has received a ping. Except it won't be triggered by a ping, but a referrer record in the access_log table. So here's the logic in crude "pseudo code".

1. START processing:
2. When cron runs,
3. For each row in the access_log table that accesses a Drupal node and has a referrer URL since the last time cron ran
4. Try to fetch an RSS type feed url from the referrer site.
5. If the site doesn't provide a feed url, then it's likely it's not a blog or news type site and we can stop processing.
6. If it does provide a feed url, fetch the feed and parse it.
7. For each article in that feed check whether it contains a reference to the Drupal node referred to in the access_log record.
8. If it does not contain a reference to said Drupal node we can stop processing.
9. If it does, extract an excerpt and title from the article in the feed and save the whole lot into the trackbacks_received table. It should then appear under the node on my site when it is viewed.
10. STOP processing.

As a slight aside, I'm wondering how to check comments on others' posts as well. I'm not quite clear whether all types of feed have a way of discovering the comment feed, if one is available, for each article in the main feed. If so it can probably be done.