I don’t think this change was motivated by the key numbers that Lance mentioned – although he is right that they will be watching these closely to ensure they are aware of any negative side-effects. I think this was about putting a foundation in place for future changes. We’ll all have to wait for these to start to show up over the coming months.

Also, I agree that job satisfaction amongst the dev + design teams is a good criteria. This was certainly a big part of the motivation behind the migration to .NET that I have written about here previously, and wouldn’t be surprised if that was again the case with these changes.

We can only speculate at the motivations and all of yours are along the lines of what I would’ve asked. It’d be interesting to have stats for page size pre and post change. CSS based layout is often a smaller page size but a much larger stylesheet size. However as the stylesheet is loaded less, it’s a better deal in total. When done properly, the flexibility of CSS in terms of layout is excellent as you’d well know.

To say I’m surprised at the fixed width layout is an understatement but perhaps there’s some jiggery pokery going on in the background that’ll help future adjustments to this.

But haven’t we got to a point in screen sizes now where maximisinig is virtually worthless? Apple has been aware of this for a long time but PC users still seem to love maximising, me included. At what point does screen resolution actually hinder your website layout. 1024? 1280, 1600, 2000? I’m running 2×19″ monitors and everything still feels okay. The next monitor I’d likely purchase would be around 23-24″. Would I really want a site then stretching that full width? I don’t have the answer but would guess no. Something for browser manufacturers to be thinking about too.

No doubt as Lance concludes, the motivation was for better income. However, don’t let job satisfaction be excluded from your list of criteria as the reason for the HTML change. If you get the best HTML’ers in the country in 2009 and tell them they have to code in tables they’re not likely to stick around.

As a side note your email notification doesn’t seem to be working.

i.e. Failed redesigns: Use web standards or don’t bother redesigning ;)

I like the idea that if you’re going to do it, do it right. But then I am a perfectionist, for all the good and bad that does me!

And I think producing great html/css/js is only possible if all your developers know how to produce great html/css/js (i.e. they have to care), and I get the impression that it’s still pretty hard to hire/train that into a team (or to justify doing so). Not that I’ve tried.

@Adam
I like the way you think :) On the bizarre id’s… I wouldn’t be surprised if they were the spawn of ASP.NET. But then I haven’t worked with ASP.NET for a few years, maybe that sort of thing isn’t an issue anymore?

From Hacker News:
Do sites using tables for latyouts have an advantage over W3C compliant sites?

Interesting debate, and sort of the same point that I was trying to make in my previous comment.

From 456 Berea St:
Resolution vs. Browser Size vs. Fixed or Adaptive Width

Check out some of the articles linked here, they all point out that the key thing is not screen size (as reported by most analytics tools( but browser window size – i.e. the width of the actual area where pages are rendered (which seems to be mostly speculation). I know I for one don’t have my browser maximised on the large screen I use.

Those are the key things, but I don’t think they are the motive. Of course they will be watching to make sure that none of those numbers tank, but I don’t expect that will happen.

I think the motive was to setup the design to better accommodate future design changes (I don’t know exactly how you measure whether or not that was “achieved”). So, I guess we look forward to those?

The success or otherwise of the redesign will be unknown to us outsiders (though if we see it revert we will have our suspicions).
Internally it will quickly be reflected in the daily stats, the stats for each step in the core processes and, something we can see, in the number of listings at any one time.

I have yet to have a good look at it. Perhaps the blandness is to blame.

My question is not whether the HTML is right or wrong (how do you tell that?) My question is whether it’s really an improvement over what came before.

I understand the cost of making this change, I just don’t understand the benefits.

Some common reasons that might be used to justify this sort of change (some of which I’ve used myself, to be fair):

Speed (as a result of smaller page sizes and better caching) – How much smaller are the pages following the change?

Accessibility – Was it tested before and after? What improvements, specifically?
Easier to read – What difference does this make? Presumably it would be easier to make changes to the code if it’s easier to read. Any way to measure this?
Separation of data and layout – Again, this sounds more like a means of achieving a benefit than a benefit in and of itself. What does separation afford? And can we really claim they are separate if the mark-up is full of redundant div tags and browser hacks
Support on different devices (e.g. mobile phones) – Perhaps? Was this something that was broken with the old mark-up?
Search Engine Optimisation – As I’ve previously mentioned here the most important visitor to your site is blind, so this seems a good reason. But, again, how do you measure this? Increased pages indexed by Google? Better click-thru rates?
Future Proofing – See Adam’s comments re: HTML5
Better browser support – Really? The code seems to be full of hacks to support the existing set of browsers that people currently use

Are there other reasons I’m missing?

Unless there is a clear reason, which can be measured afterwards, then my suspicion is raised.

@Adam:

I’m pleased to hear that HTML5 is catching on. God forbid that developers would be interested in using a standard that is actually widely supported by browsers that normal people use! ;-)

<title id=”SiteTitle_Title”>

One thing I didn’t like was the de-emphasis of the Motoring/Property/Auction sections. The graphics for those used to be quite easy to scan for, but now they’re replaced by those generic coloured boxes, which seem a bit abstract and require a bit of cognitive effort (Don’t Make Me Think).

Code wise there’s some curious bits: “” for instance. I can’t imagine what the ID’s used for?

There’s still a lot of JS in the HEAD as well. I wonder how performance tuned it’s been.

And XHTML? Bah! It’s all HTML5 now, baby. Must be time for a redesign… so behind the times. We’ve all just had the epiphany that HTML was as semantic as XHTML after all.

Gutted that they have animated ads. Cue ad blocker.

What’s wrong with that HTML? Looks like good inheritance based CSS to me. The HTML comments are probably due to IE6 borking whitespace and are a necessary evil. The fact they’re empty otherwise would probably mean they’re for presentation, likely as corner boxes. We’ll get those with CSS3, so about 6 years time. In the mean time that’s a whole lot better than tables with rounded corners surely! Semantically, it makes sense, they mean nothing.

http://rowan.wordpress.com/files/2009/07/kevo.gif

Only lasted on the site for about an hour before we had to remove it due to lack of sense of humour. :-)

Animated ad’s effectiveness is inversely proportional to our ability to ignore them. Unfortunately their irritating-ness is proportional to their effectiveness…

Is that true? It seems so on the surface, but I’d be interested to see some numbers.

PS what are these “DVDs” you talk of? ;-)

When these started I installed an ad-blocking plugin to firefox. Now I use a different browser, and although I haven’t installed a new plugin – I don’t need to.

Over 90% of the time I spend on the same sites which I essentially boycott if they have animated ads. I.e. I read BBC news rather than stuff.

Unfortunately Trademe has a monopoly in NZ (I think – will now do further research).

Animated ad’s effectiveness is inversely proportional to our ability to ignore them. Unfortunately their irritating-ness is proportional to their effectiveness…

I think it comes down to a simple call – is our site essential enough that we can piss off our users and they will still come back. In the case of trademe, with a near monopoly – the answer is certainly yes.

Site owners should love their users and thank them for making the site a reality by visiting it, rather than overly trying to monetise them _at users expense_

Rant over – that feels better, animated ad’s bug me as much as that anti-piracy techno video at the start of DVDs these days

One thing i don’t feel was good, was the transition, they changed the site so much, that when I logged in the following day, I thought I was on the wrong site, and went and clicked on my bookmark again.

It can’t be a great idea to alienate your user audience like that.

I also use kiwibank which has had a new redesign and for a week or two before the transition they showed an image of what it will look like when they change, and it was an easy transition and felt comfortable when it happened. Obviously a bank is a last place you want people to feel insecure about being on the wrong site.

kiwibank has also gone with a fixed width design, but they have retained some elements that go full width, and overall I am really impressed with the design and usability enhancements they have come up with

For me personally I think that emotion is an important part of any design (talking here about both graphic and UX design). A great design produces a tailored positive emotional response in the user and a bad design produces either no emotion or (worse) a negative emotional response. These emotions contribute to both the users first impressions and their overall experience.

When your switching from an existing design like TradeMe has then you also have to deal with the emotion that existing users feel for the old design and their reaction to loosing it. When a significant portion of your revenue comes from those existing users then your really should respect that and offer real value to those users along with any change you make. As you say, the benefits appear to be mostly to TradeMe themselves and not their users.

Ignoring the fact that their was a previous design, my immediate response to the new design is that it’s not particularly attractive, it lacks character and is a little rough around the edges. That and annoying things like the fixed width design plus the obnoxious ads means that I’m not emotionally engaged in a positive way. Of course that’s my subjective opinion.

Is it just me or does the new header design seem to have the same influences as the stuff.co.nz redesign?

I’m a big fan of good semantic HTML, but it looks like TradeMe has caught a case of divitis and classitis… which imho is as bad as being a tableholic ;)