Left Brain Right Brain

Whereas the previous design was brash, over-heavy and looked bad in certain situations, hopefully this design is the antithesis of that.

What I hoped to do with this design is bring my content to the fore and improve usability but at the same time create an aesthetic that promoted a feeling of calmness and willingness to continue as well as evoking trust.

The previous design was basically trying too hard. I forgot that the primary role of a blog was not to use every CSS trick in the book and ‘force’ a design on people but rather to make it easy for users to read the content.

To that end the body text is the darkest colour on the screen, I’ve set the ‘skip’ link to be viewable at all times rather than hidden via CSS. Links are unobtrusive rather than disrupting the flow of the content and external links are all shown and grouped better rather than hiding some away.

The ‘suggest’ feature whilst a nifty trick was hardly used so I’ve dropped that – it made me uncomfortable that it invalidated the markup anyway – and behind the scenes I’ve started introducing more categories in order to make finding things easier. Lastly, I’ve also tidied up the comments and reduced them to something less hard on the eyes – I think (hope!) I’ve made better use of Gravatars this time too.

So, thats it. I still couldn’t hang on for the CSS Reboot but I’m happier with this design than I was before. For now anyway ;o)

I’ve done a very difficult thing this week – I’ve admitted to myself that this design is poor.

When I set out tor edesign from the pink design I had last time I had two goals: first, to improve the readability and second to have something nice to look at. Its fair to say that I’ve compromised both of those goals.

This design is overdone, swamps the middle of the page and is way, way too heavy on the eyes. It doesn’t scan well on shorter entries and the lack of a footer makes the whole page unbalanced. I concentrated so much on grandstanding that I forgot a few of the fundemental rules of design. I designed solely for me.

In a way I’m glad I made this live before the CSS Reboot because it certainly wouldn’t grace that competition. I’m not fishing for compliments at all here but clarity after the fact is all too easy and clarity when designing for onesself is never easy to come by.

I have to view this as a learning process and so I’ll be redesigning this blog as much as I can over the next few days. In a way, it’ll tie in nicely to the fact that I’ve just upgraded WordPress to 1.5. Expect something radically different to this design that concentrates on the needs of the user when reading the page, rather than showing all the nifty CSS tricks I’ve learnt.

By the way, if anyone’s having trouble commenting please mail me kevleitchATgmailDOTcom. As I say, I’ve just upgraded and I know at least one person has had trouble adding comments.

Well, I finally got around to upgrading this blog to WordPress 1.5 after a brief unsuccessful dalliance with Textpattern.

A few things are broken depending on how well a particular plugin does or doesn’t work with the upgrade and category listings are screwed but other than that its was pretty smooth and easy.

OK, so I’m really impatient. I signed up for the CSSReboot and here I am launching very, very early. Sorry Adam.

Anyway, some notes from the design process as well as my aims for this design.

Continuing the theme of strong warm colours (the previous design was pink for those that never saw it), I decided to go for a very warm deep red with this colour scheme. My job requires me to design very pared-down sites so I like to be a tad exuberant with my personal designs.

Again I have to ensure in all commercial designs I do that the site is totally cross browser compatible (cough IE cough) so thats what I do. However, I’m getting so fed up with dropping in hacks to cater for some browsers (cough IE cough) that I’m growing increasingly uncaring about how my personal sites look in browsers that can’t adhere to standards. Hence I’ve made an effort to ensure the site works for web standards and then put in generic hacks for IE. I suspect the design might jitter slightly in IE5.* for both Win/Mac and I have no idea how it looks in Camino at all. I took at screenie from iCapture for Safari and it looks OK. Not spot on but workable.

This design sees a major semantic overhaul. I’m finally reasonably happy with the underlying code of the site and think I’m now in a situation where a redesign means a CSS overhaul only. Its also the first time I’ve structured my CSS files in such a rigid way (including using conditional statements for IE based CSS) and I think its definitely paid dividends for me in terms of time and cleanliness of code.

My DTD says I’m going for XHTML 1.0 Strict and baring one design element the design meets that criteria. The one area it fails on is the Search suggest tool (type into the search box above you to see what this does). It breaks mainly because for the Javascript hooks to work the form requires a name attribute as well as an id. I’m in two minds as to whether to keep this feature or not. If it gets used a lot I’ll keep it. If it doesn’t I’ll drop it. I’m trying to be more accomodating of users with perceptual/cognitive disabilities and this seemed a very helpful and intuitive addition to my site search options. We’ll see I guess.

As well as the Search Suggest tool, I’ve also added a menu switcher to differentiate between internal and external links.

As regards accessibility, the design mets current Priority One (A) checkpoints. It would meet AAA if it werent for the requirement for liquid layouts in Priority 2 (AA). However, as I’ve said in the past, the current WCAG standard is not great at meeting the needs of uses with a learning or perceptual disability and hence I’ve made my design work for a cross section of users rather than for an increasingly archaic accessibility standard. Here’s hoping WCAG 2.0 is better.

I’ve used the UK Governments accesskey convention for all accesskeys on this site.

I toyed with the idea of content negotiation for quite some time but in the end reluctantly decided to not implement it. Using it means I need 100% accurate code and in a site that allows markup in comments I can’t be sure that will always be possible so for now, I’ll stick reluctantly to tag soup.