The site has scarcely been updated since the end of June. I managed to upload several posts from India but none of the accompanying photographs – and indeed none of the re-edited prose. Early this morning I reloaded the amended versions of the Goa pieces, so they make at least some sense now.

After we returned from Goa we (almost) immediately bought a house. This completely surprised everyone involved, and inevitably caused immediate chaos. Plans needed changing; finances needed juggling; furniture needed moving; walls and ceilings needed painting; and strange shops and bus stops needed finding.

One month later and version two of my thesis is still a set of unpublishable notes, while the ideas I had in Goa about this site (and hence about the memi that it supposedly sketches) are still unimplemented.

I have four main problems with the current incarnation of this site (insofar as it is a demonstration of some of the main concerns that lay behind the idea of the memi). The first of these may seem trivial but isn’t.

The permalink system is not to my liking, and is very unhelpful when trying to locate past articles in the Admin section. It uses the approved date & name system like this: /2007/04/04/selling-the-sixties/ but since my stuff isn’t sorted by date this has limited utility. I am currently thinking that a system using the article id & name would work better: /288/selling-the-sixties/ since the id number would serve a logical purpose.

There are several problems with changing this though. Firstly anyone who has linked to an article will now have a broken link. This maybe okay once, but it does mean I need to be clear about why I am doing this – and put up a clear 404 error page explaining how to find what you were looking for. Secondly, all the internal links will break, and the only way to fix this would be to go through the database record by record altering them by hand.

Of course, I also need to avoid delay. If I am going to change the permalink system then it has to be done soon. The longer that I leave it the harder and more time-consuming it will be; and the more people it will annoy.

The second major problem is the category and menu system. I think it needs reworking, but I will detail the reasons for this, and the proposed solution, in another entry.

The third problem is the future. A memi, a personal / portable data bank ought logically to include future events, in a calendar form, as well as archiving past events, writings, thoughts, and external material. Ideally a calendar entry ought to degrade smoothly into a historical note. I can find no obvious plug-in that will achieve this, although there is one – Event Calendar – that might get close.

This raises issues that elide into the fourth problem: where are the walls of the garden?

In a networked world it does not make sense to imagine a completely individual memi, in the way that Bush originally conceived the memex. So, where should m photos be stored: on my server or on Flickr, with some neat links and embedding? How should my site relate, or link, to my Facebook profile?

The issue about how (and at what level) an individual repository should merge with social networking sites is fraught with complexity. Sadly, this is a complexity that I definitely need to begin untangling.