I think we’re getting close to a winner here. I’ve put in a two-tone background for the main area, and while that still needs a bit of tweaking here and there, I’m pretty satisfied with the effect. It brings the yellow through into the main area just a bit, and gives the background pattern a bit more body, without screaming too much about it. I do actually want to reduce the yellow slightly from the example shown here, but not by a terrible lot.

Right now I’m actually looking for a filter, or a trick that’ll allow me to delete a certain percentage of somewhat randomly spaced yellow pixels from the top layer of the background image, without simply lowering the opacity for the whole layer, which just doesn’t have quite the same effect. I’d like a quarter, or possibly even a third, of it to be gone. But that’s mainly down to tweaking, the basic design is there now – I think.

One final point: if you compare with the earlier screenies you’ll notice I’ve changed the colour of the top menu-layer to match the background for the main area. The only other tweak I’m currently looking at is the link colour for the main area, which isn’t quite working for me now that it’s more or less blue on blue. Even so, I’ve got a feeling the end of the redesign process may well be in sight. Or at least the end of this round of the redesign process. I’ve sort of given up on the idea of ever being truly satisfied with this site’s look for more than a few months at a time.

Since I have a little time to spare from NaNoWriMo now, having reached 50k, I’m slowly bringing back some other activities in my spare time. One of them is playing around with the design for this site. I’ve got a new layout pretty much worked out, though it has one downside: it doesn’t scale right now.

The only way I’ve found of doing a good scaling design is to use tables for structure. This has been an official W3 no-no for a while now, but the fact of the matter is that tables, because all the cells are linked, are far more forgiving when it comes to scaling than divs will ever be.

Be that as it may, I’ve decided to play nice with the new design (for the time being at least) which means I’m doing it all in divs. That may mean fixed width, but I’ve actually managed to get it to a point where I’m not too unhappy about the way it looks, even if it doesn’t scale all that well.

But while I’m pretty happy with the new layout (and actually have been for a month) there are two things nagging at me. Firstly there’s the fact that the background acts weird when scaling down below about 1000px browser width. This is to do with the fact that I want a non-scrolling background and the only way to have that work in MSIE6 is to lock it to the body element (which of f*cks up the positioning when the window is smaller than the body width).

Now I could probably live with bad scaling, at least for the time begin if there weren’t also another problem: I can’t quite decide on the colour scheme. I’ve been contemplating a switch from black on a light background to white on a dark background for a while now, and have managed to get it to the point where it looks okay.

But the fact remains that most rendering packages tend to get confused when you do this. Rather than making the surrounding pixels lighter to antialias the font they make some of the pixels in the letter darker, which means a light font on a dark background is pretty much guaranteed never to look quite as clear cut as a dark font on a light background.

So I’ve taken some time yesterday and today to experiment with lighter backgrounds and black text again, and I think I may be close to getting a winner. But even so I would like a few more opinions. So as you’ll have noticed the thumbnails linking to the two screenies by now please have a look at both of them and let me know what you think.

Or: A Tale Of Limited Bandwidth

I’m not sure whether I’ve blogged about this before or not, but I do some volunteering for a local chapter of the political party I belong to. One of the things I do there is system administration, which currently involves a switch from Windows to Linux. Which is why I am, right now, downloading hundreds of megabytes of uploads over a 512 kbit DSL line.

Now there’s nothing wrong with 512 kbit lines as such, I remember dialup well enough for even this to feel pretty snappy in some respects. But when you’re pulling several hundred megabytes over a line like this you do notice that the rest of the internet starts to feel slow. So as I’m on a Linux system, with Lynx installed, I decided to do a bit of browsing that way.

For those who don’t know it (though I think I’ve mentioned it before), Lynx is a console (text) based browser which, as it is text based, omits images as a matter of course. And on a somewhat stressed line getting rid of the images (90% or more of the data for most pages) speeds up browsing quite a bit. Even so it makes writing this blog entry a rather different from usual experience.

And it’s not just different, it’s also slightly nostalgic. I have no real memories of text only internet, but I do remember text only (DOS) computing, albeit without anything like the internet around at the time. Working with a black background and (mainly) light grey text does bring back memories. For that matter some elements in the interface, especially a bar with yellow text on a blue background at the bottom of the screen, make me nostalgic for Word Perfect 5.1.

I’ll end this entry here. I mainly started it to kill time while waiting for the downloads to complete, and from the look of it they’re nearly ready. Still it’s been a fun exercise in retro computing, and one I may well repeat at home.

While talking about importing from Blogger to WordPress with a friend I decided to go ahead and take the plunge myself (I’d been thinking about it for a while). So I’ve imported my old Blogger Writing Blog into this one. It’s going to take some time got get all 65 posts properly recategorised, but I’m making progress on that, and at least it means all my blogging is now in the same place.

Update: I’ve finished shoving everything I just imported into the “Old Writing Blog” category. I’ll probably go through later to add some more category details, but it’ll do for now. I’ve deleted exactly one old post, to be precise the “blog moved” post when I closed up that one and moved everything over here.

Yes I know, I haven’t actually finished the re-design yet. For that matter I’m not at all certain the current code will survive, but I have got a testing version of the new layout up and running… more or less. Right now it’s got a hack to display right in IE6 and probably still needs one to work in IE7, but that can wait.

You can have a look at the testing version of the site now right here. The only thing not working in the testing version is the blog. And the front page content is currently identical to the My NaNo History page because I had to have a good amount of text in a hurry and that happened to be easily available. I guess I’m really going to have to figure out what the hell I want to do with the front page when I put this design up for real.

I’m also by no means certain about the current colour scheme, or for that matter anything beyond the basic layout. But having said that, it is a pretty fair test run for the basic layout. Oh and one more IE6 note: the on hover sub menu switching doesn’t work in IE6. I’m currently considering whether or not it’s going to be worth the effort of hacking in some javascript to get that to work. I have no particular sympathy for anyone using IE6 voluntarily, but many people are forced to use it at work (I should know, I’m one of them).

I’ve been pondering the site redesign in recent days, and particularly the problems having a menu over two layers would cause. If I want to move the menu to the top of the site, which is what I’m currently thinking about doing, it simply can’t accommodate all the entries it has now.

One option would be to do drop-down submenues.¹ But it’s difficult to do those well and most of the times I’ve tried to do them I’ve ended up pulling them out again because they didn’t quite work. That consideration more or less left me stuck with a menu down the side of the page the way it is now.

Then I realised the answer’s been staring me in the face almost daily. To be precise every time I check the Arch Linux forums.² My homepage and blog, and the headers from the present menu could form the top layer of such a menu, with the other entries grouped on the lower layer, and displayed based on context.

The one dilemma that leaves me is whether I should go for a simple static implementation like Arch’s, or a more complicated one which changes the second layer on mouse over as well. I’ve done that before as well, but in that case I’ve never felt quite happy about the result, so I’m hesitant to try it again. On the other hand it would allow a greater amount of One Click navigation.

I haven’t even got a preliminary design for this yet, since this type of structure change would definitely mean redoing the entire site’s design. But the realisation that I have been looking at the perfect way to do this almost daily is certainly worth an entry here (as well as hitting myself gently in the head later on).
_______
¹ The Dragon Page offers a good example of these
² Located here

Or: How Monitors Can Ruin Your Designs

Yes, the long awaited monitor rant is here. Reinforced by the fact that I checked my two recent attempt while at work. And on my monitor there I could sort of see why someone might not like them. They looked… different from the way they look on my own monitor at home. I could probably change that monitor to match what I have at home, but that would merely solve my frustration not the basic problem.

And the basic problem is this: no matter how you tweak a design, you can’t make it look the same for everyone everywhere. Another good example: my Samsung SyncMaster 940BW can actually cycle through a number of pre-set configurations as well as my own custom setting. I can use this to test what a design might look like on different monitors (probably should do that more often) though none of them happen to match the way my designs looked at work.

When I go over to Starstuff’s Rule-TwentyNine.com and cycle through my screen settings at least one of them makes the blue borders and sidebar background fade into the white background for the rest of the page.¹ That of course can be headed off by using something with a bigger contrast, but even then it would look radically different on each setting, and there are people using most (probably all) off these out there.

Which more or less forces you back to Black on White or White on Black if you want to have any real influence over the appearance of your site. Black and White still do fall prey to these settings (especially some of the more ludicrous ones), but the two of them – and to a lesser extend shades of grey – are the closest thing to consistent rendering available. In fact, no matter how wacky the monitor configuration people are likely to know at least what Black and White look like on their own monitor.

Now I’m a control freak when it comes to this sort of thing anyway, but I suspect this would be enough to drive me into ranting territory even if I weren’t. Trying to get things to look the same across browsers is a nightmare already, but at least it can (to an extend) by managed by testing and then coding around the problems. Trying to get things to look the same on different monitors is an exercise in futility.

One would almost wish someone would come up with a standard configuration to be enforced across monitors. Unfortunately that would almost certainly be fucked up, so I guess I’ll have to learn to live with it. But that doesn’t mean I have to like it.
_______
¹ The same thing goes for some of Wikipedia’s background colours.

I had been planning on writing the monitor rant tonight, but just didn’t feel like it (I’m not in a ranting mood, so to speak). Instead I found myself playing around with the colours of the design again, using The GIMP to take the light version of the background image through pretty much the entire colour spectrum. I had great hopes for the reddish tints, but funnily enough I found myself drawn to the green version instead.

Like the previous yellow version I’m considering this one as a possible replacement for the current theme, though I’m pretty definitely considering redoing the whole site now. The way I see it now I’d like to have a menu along the top, locked in view like the currently location bar, though I can’t see how to do it without having to change the menu structure again as well. Of course I could work with main items and dropdown submenues, but I tend to dislike the way these work out. They’re elegant enough in theory, but somehow never quite work in the real world.

As always, comments are welcome.

After doing the dark design earlier in the day I decided to give into temptation and design a lighter alternative as well. Again this is by no means the full redesign I’m considering, but I might actually consider putting this one up as a temporary replacement for the current look.

The one thing I’m not too sure about in this design is the yellow for the page title and copyright notice sections, but that’s certainly easy enough to tweak. And while the blue for the sidebars is the darker in my current design than it is on the site now I may well end up taking that back to the present level. Right now I’m in the process of trying to find the right balance for contrast.

Of course, as with the previous design, all comments are most certainly welcome on this one. One of the main reasons I like it is that it seems to make the text in the central section come out clearer. I may end up switching to that background image even if decide to leave the rest of it be for now.

I’ve been playing with the site design a little this morning. The thumbnail to the left, and the larger version it links to, are the preliminary result.¹ I’m not really seriously contemplating using this on the site (though that might change, it’s Sunday after all so I’m allowed to be capricious). But it’s a quick first look at a potential new colour scheme.

If I really do go in for a full redesign I’ll probably switch things around more than this. For one thing I’m no longer entirely satisfied with the way I’m doing the site menu on the left. It works well enough, but it leaves an awful lot of empty side bar beneath it a lot of the time. By no means a major disaster, but never the less something to look into.

For that matter, if I do go for a full redesign, I may well end up moving pages around quite a bit, although I have to admit that on the whole I’m pretty satisfied with the way things are now in that respect.
______
¹ In case anyone is interested, this is actually a screenshot made using a totally local version of the site on my own computer, not just a mockup. It shouldn’t take me more than 10 minutes to roll this out to the site if I wanted to do that, since it’s just CSS and one new image.