New HTML5 elements and IE8 Last updated:29 October 2010

I was reading a discussion on a forum last week about HTML5, and thought I’d have a quick mess around with it to see what happened with various browsers. I created a test page using the new tags provided by HTML5 (<header>, <footer> etc.).

This page doesn’t display correctly in any browser that I tested (FF3, IE6/7/8, Chrome2, Safari 4, Opera 9.63 or even Opera 10 beta2), although FF, Chrome, Safari and Opera all display the <aside> element correctly for some reason.

I then applied display:block to the css for all the new elements and tried again here.

Slightly to my surprise the page looks fine in FF, Opera , Chrome and Safari. Slightly less surprisingly, it fails horribly in IE6 and IE7. The biggest surprise for me was that it looks just as bad in IE8. I suspect, on reflection, that the browsers that worked did so because of their treatment of unrecognised html elements, rather than their support for the new HTML5 elements.

HTML5_IE8HTML5_Opera

Here’s what it looks like in IE8 on the left (pretty much the same as in IE7 and IE6) and here’s how it should look in Opera 9 on the right. I have to say I expected it to be OK in IE8, given that’s the pretty much the newest browser I tested with. So much for HTML5 support in IE8…

Conclusion: being able to use even the simpler features of HTML5 looks a long way off…

Just start it… Last updated:29 October 2010

Sometimes the best thing to do with a project task is to start it…

Sounds a daft thing to say, but there’s sometimes a danger of over-planning and over-analysing a task, instead of actually getting on and doing it.

I can think of two personal examples. A few years ago I was managing a complex set of interfaces between two systems. One of our suppliers was reluctant to sit down and talk about it because they wanted more preparation. However, we gained more from a rough and ready two hour meeting with everyone round the table than we would have done from a couple of weeks planning. The supplier was generous enough to agree this as well.

My other example concerns a web tool which had been discussed for a long time at high level, to the extent that everyone thought it was a difficult task. Again in a couple of hours, by talking about some detail and the sort of data we wanted to display, we were able to take a big step towards understanding what we were trying to do, and realise it wasn’t quite as hard as we thought.

Sometimes you just need to start talking about something – it’s pretty certain that talking discussing some detail will focus thoughts and help get a clearer idea of what it is you’re trying to do.

Here’s a great quote on the subject:

“Whatever you can do or dream you can, begin it.
Boldness has genius, power and magic in it.”

Aside: This quote is generally, but incorrectly attributed to Goethe, but in fact was introduced as part of a “very free translation” from Faust by John Anster in 1835. Now you know…

Integrating WordPress into your website Last updated:29 October 2010

In the last couple of weeks I have incorporated WordPress blogs into two sites, applying the existing site style to the WordPress pages.  You’re looking at one of these right now…

I didn’t particularly want to replace all my existing pages with WordPress pages, I just wanted to use the WordPress functionality to power the blog element. Pretty common requirement I’d have thought. I found this article a useful starting point.

So, if you have an existing site with its own css file and you follow this tutorial exactly then you will end up with two similar css files, your original one and one just for WordPress files called style.css. So then you have to make changes in two places…I didn’t want this so additionally amended header.php to point at my “main” css file. The downside of this is that you can’t update your css file through the WordPress admin module, but that wasn’t at all an issue for me.

It’s pretty much guaranteed that this will break the display of at least some of the wordpress pages, as chances are you won’t have entries in your css file corresponding to the html markup that WordPress generates. The same things will happen of course, if you replace the contents of style.css with your own css.

For example, my main site uses #header and #footer as id names, just as WordPress does, so no changes required here, but I used #maintext instead of #content for the main page content. So you can either amend the html so the WordPress files use #maintext, or add #content to your css.

In this case, you can identify #content just by viewing the various .php files – but some classes and ids are directly generated “on the fly” and don’t appear.  I found the easiest way to do this was to view the actual pages using Firebug, and make additions to my css file accordingly.

This isn’t quite as painful as it sounds as WordPress is pretty consistent about the markup.

I’d guess it took a couple of hours in total to do, including actually amending the contents of sidebar.php. Worth it as all my styling sits in one file.

Project management documentation Last updated:17 June 2024

A couple of weeks ago I ended up talking for a few minutes about project management and what it isn’t, and thought I’d post it here as well.

There’s a strange sort of mystique about project management and what it involves. Clearly project management is about getting your documentation right – plans, risk registers, PIDs, issue logs , change requests, progress reports. No.  Most of the really good project managers I’ve had the privilege to work with have used these things. But all of the project managers I would judge as less successful have used these as well. There is an unfortunate tendency to mistake intense activity completing all these things for actually getting something done.

I can illustrate this with a personal example. Way back before I was a project manager I was on the receiving end of a project, as it were, when I used to manage a call centre. Every Friday I – and five colleagues doing the same job elsewhere – had to complete a document reporting progress and risks and issues. I used to complete the form religiously. Far more religiously than my colleagues, it turned out, since I was publicly congratulated for doing such a good job. All well and good. The problem was that I was making a right hash of actually completing the project…..

In recent years one of the least successful programmes I’ve worked on used the most documentation and adhered most rigidly to project management standards. It passed two heavyweight QMS audits without any major problems. And yet, ultimately, this particular programme failed. Another instance that sticks in my mind, on a different programme, was the use of very detailed documents to specify the interface between two different systems. Everything you could possibly need to know about this interface was contained in a single document. Great, eh? No. Our supplier had a highly intelligent lead programmer working for them who really operated on a higher intellectual plane than the rest of us. He was heavily involved in all the discussion about these interfaces, but even he couldn’t understand the resulting document – it was too complicated. And if he didn’t understand it, what hope did the rest of us have?

At the other end of the documentation spectrum I’ve also seen a project deliver an application to 6,000 people pretty successfully, which never had a risk register or an issue register. (Actually, it did have one right at the end of the project because it got audited, and one was “found”, ahem, but that doesn’t really count.)

So don’t be fooled into thinking that project management is just about filling in documents. The difference between good and bad projects isn’t documentation and all the hideous apparatus of the project management textbooks. It’s the people working on the project.

We all know this anyway, because good people in any sort of work make things variously easier, quicker, cheaper, and even, perhaps, more fun. It’s no different on projects. Good project managers find it helpful to use risk registers etc to help them organise a project. The key word here is help though.

Risk registers and project plans are not an acceptable substitute for sensible project management.

The lifespan of IE6 Last updated:29 October 2010

Read an interesting article this morning about the use of Internet Explorer. Specifically, the article suggests that now that IE8 is launched, users will migrate from IE7, but many who are still using IE6 will remain, to the point that IE6 will become more popular than IE7. Sound mad? Not really, because many corporate web applications were designed for IE6 when it was effectively the only browser available, and they won’t work with IE7. Larger companies tend to be intrinsically risk-averse anyway, upgrading a browser is low priority – my own experience certainly supports the argument.

A couple of sets of web stats highlight the issue. Looking at some stats from a large public sector website, 80% of visitors are using IE, of which 40% use IE6. This website will be frequently accessed by people at work. By contrast, one of the sites I run, which tends towards consumer usage, has only 60% IE users, of which only 15% use IE6

So the bad news is IE6 may live a lot longer than we might like…