The Dark Fortress

Redesign: Setting design change goals

18 February 2020 | 8th Edition

It's always useful to set down your priority goals that your redesign project should attempt to achieve.

Here these are largely based around improvements to functionality, usability, speed, and the simplifying of back-end maintenance.

The visual design needs sharpening and simplifying too. The first casualty to this has been the removal of the large distinctive sword and wings motif that proudly sat at the top right of every page. It might get replaced by something else, it might not. Things will change further. I'm fairly certain that the DA green background (in use since 2003) will turn to something a bit more friendly on the eyes.

There will also be a bit of tinkering with font weights and sizes too. The main headlines for instance, will become more prominent. And getting a proper hierarchy of heading types sorted out greatly helps comprehension of longer read items.

So visual design aside, the main goals design can be summarised as follows:

  • Reduce page loading times. Probably easily achieved through a combination of items further down this list — getting rid of the bloat and crud.
  • Reduce the use of third-party tools. In my experience over-reliance on a crucial external feature will lead to trouble if that third-party goes bust, or changes their product, or simply starts charging too much. Additionally these services often rely on embedded script and/or links to external servers to function. Bloat that needs to go. As a result one statistics service has already been terminated.
  • Cut down the size of the css file. All bytes count. Presently weighing in at 50Kb this isn't large at all, but by judicious culling of the ad-hoc one-off classes the whole thing becomes leaner, cleaner, and faster and easier to maintain. One method would be to make more use browser defaults wherever possible — rather than fighting them run with them.
  • Reduce, or actually do away with all javascript dependencies to make page features function. In an ideal world everything should be made to function using sensible quantities css alone. Failing that a graceful degradation to browser defaults. This might be a longer-term target — it might even be impossible — but still a worthwhile one.
  • Reduce social media and 'sharing' links. I'm not convinced of their worth and as these usually required javascript of some sort to function and/or operated through a third party service. Frankly it's another reason to wave them goodbye.
  • Do away with commenting. It's not that comments are bad per se, it's because I don't like the systems of delivery. As a result Disqus has already been deleted from all pages. Unfortunately not being WordPress based I can't take advantage of its nifty baked-in commenting system. It might be a neat exercise write one myself.
  • To become more accessible. I write this not really knowing what that means. But more more attention should be paid to ARIA descriptors than is currently the case for sure. Oh and getting rid of that green background.

No doubt I shall fixate on many of these items in more detail at a later date.

Ends