Tag Archives: documentation

Tag clouds, calenders, and more

Snap Preview Anywhere

Last night, I finally disabled the Snap Preview Anywhere feature (now know as Snap Shots) on my blog (I actually used the Snap Preview Anywhere WordPress plugin). It seemed like a good idea when I first installed it. Every external link on my blog had a little icon alongside it that you could hover over to get a thumbnail preview of the site you’d go to if you clicked the link. At first, however, it didn’t work – I just got a message saying that there was no preview available, for any link. That could have been the plugin not working properly because, after a WordPress update, some of the link previews started working, but only those that pointed to high-profile sites like Wikipedia and Flickr. Which is a bit crap because the majority of my links were still just saying that the preview wasn’t available.

Looking at the Snap website now, the description only really says that such high profile sites will be previewed. The original descriptions that I read when I first installed it on my blog suggested that the previews would work on any site as long as Snap Preview Anywhere knew about the link – and that they’d go preview new links within a short time. Maybe that was too much of a commitment for them to honour. It’s a free service so I’m not too bothered but it seemed like a good idea that’s not really happened.

Tag cloud

I’ve been thinking that although my blog/website is pretty and red, it is actually a slightly old-fashioned design now. So I was trying to work out what would be better without being dull. One of the things I wanted to do was make it a bit more Web2.0 – not that I particularly want loads of flashy drag-and-drop – more that I’d like to have fewer static-looking lists of links. Having a tag cloud is one of the things that I wanted. Partly because the categories that I had for my blog posts were pretty bad, and they have a fairly static, hierarchical feel about them. Although it’s a difficult mind-switch to make, I’d like to get away from thinking in hierarchies.

I once attended a task analysis tutorial given by Dan Diaper and one thing he said that stuck in my mind (bearing in mind my job as a technical writer) is that although we impose hierarchies on many many things (in order to categorise them), the real world isn’t hierarchical – that is, as soon as something belongs to more than one category, the hierarchy breaks down, and this happens frequently.

One thing that new, tagging-based sites like Flickr have shown is that we need to be more free with how we classify things for retrieval. On Facebook, for instance, although you post your photos in albums, you and your friends can tag the people in the photos so that it’s possible to retrieve all the photos containing a given person from across multiple albums. You aren’t restricted to a strict hierarchy of albums and sub-albums to locate an individual photo.

A similar example is some Adobe software that my Mum uses to organise her photos. She tags each photo with things like the names of the people in the photo, the location of the photo, the event taking place, and so on. The software can order the photos along a timeline so that you can pick a particular date in time and see what photos were taken on or around that day. You can also sort by tag; for example, show all the photos of pets, which is great because the software dynamically creates an album starting with the family dog when she was young, the pony she had from when she was about 15 years old until I was 15 years old, the dog she and Dad had for 17 years from before I was born, the various cats through the ages, up to their present dog and our two-year-old cats. Tony’s recently found a similar piece of Open Source software, F-spot, which aims to do something similar and which I’d like to try using to organise our photos.

So, anyway, information is all about retrieval. It’s no good me using logical but useless categories (like ‘Personal’ and ‘Opinion’) to organise my blog posts (they’re more useful for me than for other people). The trouble is that I find that when I’m thinking hierarchically, it’s difficult to decide which category or categories a post fits in to. Also, the site design that I use just lists the categories down the side of the page so the more categories I have, the longer and more boring the list. Tagging is a bit more natural an exercise, I think.

The SimpleTagging WordPress plugin provides what I want. I imported my existing categories (then decided they were terrible and deleted them all anyway), and then went through my existing blog posts and created and renamed tags to describe them. The tag cloud is a visualisation of the frequency with which each tag is used. The more popular tags (ie the topics on which I have written most) are shown in a larger, brighter font in the cloud.

What does the tag cloud show the reader? Well, apart from looking pretty, I think it gives the reader an idea of what the blog is about and, by extension, what I’m interested in writing about. I was a little bit disconcerted (though not terribly surprised) to discover that my interests seem to mostly revolve around cool or geeky stuff, though it’s nice to see that HCI (Human-Computer Interaction) is top at the moment.

On a bigger scale (though not here), the tag cloud can show what’s popular across a group of bloggers. For instance, a lot of IBMers have internal blogs at work. On there, a tag cloud can show what topics are of most interest to the IBMers who blog across the world.

What my tag cloud doesn’t show is the popularity of the posts with readers. In some ways, it might be interesting to show the relative number of comments on each post, or something like that. Though on a low-traffic blog like mine, that probably wouldn’t show enough to be that interesting.

Another interesting use of clouds is on the Nabaztag support website where the popularity of search terms is shown in a cloud so that you can just click a term in the cloud to retrieve the information instead of entering a search term. The more people who have searched on that term, the more prominently the term is displayed in the cloud. I guess the idea is that whatever problem you have, someone else has already had it and searched for a solution so you may as well benefit from their experience.

Anyway, take a look at my tag cloud, to the top-right of this post.

Archive calender

Before: an increasingly long list of months of blog archive.

After: a neat little calender in which you can click a date to view that day’s post (if there was one).

No plugin required for this one; just paste this bit of php in the right place in the WordPress theme and WordPress does the rest: <?php get_calendar(); ?>

Good eh?

Facebook badge

Finally, the last addition to my blog website today was my own personal Facebook badge – the photo of me and my Facebook status to the top-left of this post. Tony had made one of these and he showed me how. Just go to your Facebook profile page and scroll right to the bottom where there’s a link to create your Facebook badge. You get to customise the badge a bit, according to what information you want it to show. Facebook then generates the bit of HTML code that you need to paste into your blog page. And that’s it.

WebSphere Message Broker – verifying without the Toolkit

Well, as my first post in a while, I’m going to plug JT’s developerWorks article called Verifying WebSphere Message Broker V6 without using the toolkit. It’s an ace idea for an article.

If you happen to install WebSphere Message Broker, the product documentation typically instructs you to verify your installation using some of the samples (I know; I wrote some of it). But this method assumes that you have installed the Message Brokers Toolkit, the GUI development environment. While I think this is a fair assumption if you are coming new to the product to try it out or just have a play, if you’re more hardcore, you maybe just want to install the runtime (ie the bit that isn’t the Message Brokers Toolkit and that does the actual broker work).

JT’s article basically steps you through how to verify your installation without using the GUI.

I’m also happy to plug his article because he cites my IBM Redbooks publication WebSphere Message Broker Basics, as does (I just found) the Wikipedia article about the product.

Not that I’m easily flattered or anything… 😀

Task modelling

This morning, I attended a demo and discussion on the latest and greatest version of Task Modeler, a software application developed at IBM Warwick to help user interface designers and technical writers to model users’ goals using electronic sticky notes.*

The earliest incarnation (aptly known as ‘V1’) was a Java application that had the sole purpose of supporting human factors people (a.k.a. user-centred designers) to represent the hierarchical breakdown of tasks that a user performs when trying to achieve a goal. The best-loved/loathed example of Hierarchical Task Analysis (HTA) is that of making a cup of tea and the steps that you must perform to achieve that goal. A nice but slightly more complex example of HTA that I found online is in the design docs of the Dance-O-Matic.

So, anyway, at some point during the subsequent V2/V3, it was discovered that it might be useful for structuring the navigation tree in hypertext-style software documentation (a la much of IBM’s software documentation now). Bringing us neatly to V4 (the aforementioned ‘latest and greatest’) which is closely tied in with IBM’s Open Source XML documentation format, DITA. You can, for example, use Task Modeler to develop ditamaps (navigation trees) and relationship tables (something clever to do with managing links between topics) in DITA. You can also, of course, still use Task Modeler for doing HTA work if you’re more interested in HCI (human-computer interaction), user interface design, and human factors.

Despite having successfully installed Task Modeler V4 on my Thinkpad in time for this morning’s session, I have not yet had chance to play with it. I will say, though, having seen a demo, that the new Eclipse interface and funky icons that it has acquired are really rather pretty and it all looks much nicer to interact with than V1. When I’ve done some playing, I’ll hopefully be able to report more.

If you’re into human factors work or technical writing (especially with DITA), you can now download Task Modeler from IBM’s Alphaworks website.


* At this point, I must admit that I have had no hand in developing Task Modeler. I do, however, work as a Technical Writer for IBM United Kingdom Ltd at the Hursley Software Laboratories so I’m not an entirely independent observer but I should point out that the views in this post, in this whole website in fact, are entirely mine and are in no way intended to represent IBM.