Rowan Simpson

There are, in my experience, two types of websites:

1. Websites which are slow; and
2. Websites which are noticably slow.

It’s important to understand which of these categories applies to your site.

If the people using your site tell you that they think it’s slow then you are definitely in the second category.

What you can do about this?

• Measure your current performance
• Identify possible improvements
• Understand best practices
• Realise that you’ll never be finished

Also, you can make sure that you include time in your work plans to make small performance improvements whenever you make changes to the site. This is important because (despite developers expectations to the contrary) it is unlikely that the day will ever come where you’ll be able to stop working on new features or bug fixes in order to just focus on performance.

Making your site faster needs to be part of what you constantly do, rather than something that you hope to have time to work on at some point in the future.

Phil from Xero pointed me at this article about XMPP (a.k.a. Jabba):

XMPP (a.k.a. Jabba) is the future for cloud services

What do you think? Has anybody done anything interesting with this that you’re using or know of?

It would be interesting to hear about how you find it.

I’ve been doing a bit of design work over the last week or so, which has been fun. I’m not a great graphic designer, but I can do enough to be dangerous. I enjoy having the freedom to experiment with different ideas without having to bother somebody else to do the heavy lifting.

This is the first time I’ve really got stuck into this sort of work since I switched to my Mac, so I’ve had to get used to a bunch of new tools.

For HTML and CSS I’ve been using Coda, which I chose basically on the strength of its logo (see right), but which has turned out to be a nice tool. I especially like the built-in reference books, which are great for somebody like me who doesn’t spend all of their time coding and needs regular help.

http://www.panic.com/coda/

For image editing I’m using Pixelmator to replace my old favourite on Windows, Paint.net. It’s taking me a while to get used to this, and I’m not sure I’ll continue with it beyond the free demo period. I am feeling quite slow using it, and it’s missing a few of the things I use a lot - for example, it doesn’t support vectors at all so there is no rectangle tool, instead you need to select the required area and then fill in the selection which seems cumbersome when you’re doing it a lot. I’m also pretty slow with it generally, but I have a feeling that will improve once I get on top of the keyboard shortcuts etc. It’s very pretty, but I’m not sure this is the right tool for the job I have in mind.
http://www.pixelmator.com/

I’ve had a quick play with Inkscape, but it uses X11 and doesn’t really feel like an OS X application.

What other alternatives can you Mac users recommend?

(In looking around I did find a really cute tool called Scribbles. If you’re into graphic design I definitely recommend having a play with this, just to check out the smart interface, including a whole new approach to layers which seems really intuitive. It’s designed for kids, but I suspect that kids of all ages could have a lot of fun with it.)

I’ve also found a couple of other useful tool-lets which I thought I’d share (for my own future reference if for no other better reason):

At Webstock Dan Cederholm talked about basing a colour palette on a photograph from nature. This tool from De Graeve makes it easy - simply upload your image and it will pick out a range of complementary colours from the image. It even gives you the hex codes:

http://www.degraeve.com/color-palette/

They also have a simple fav icon editor:

http://www.degraeve.com/favicon/

Robin Wood has a really useful introduction to the OS X Colour Picker, which is actually much more powerful than it first appears.

The magnifying glass and swatch drawer were both new to me.

And, what’s more the picker can be extended with third-party add-ons.

I found this from Waffle Software, which adds a simple hex picker tab (see screenshot right):

phttp://wafflesoftware.net/hexpicker/

Fun, fun!

What other tools are people out there using for this sort of thing? I’d be interested to get your suggestions.

:-)

I’ve mentioned this Joshua Porter post to a number of different people over the last few weeks. It should be compulsory reading for any designer working on an interactive web site …

“You can’t appreciate a web site in the same way you appreciate a logo or a poster. When a logo works, it makes you think certain things. Makes you think about the company, their influence, their reach. It’s about branding. The IBM logo suggests a solidity, the rock that is Big Blue. At this point, after you’ve thought these things, you’re done. There is nothing else to do. Maybe you’ll consider their products in the future.”

“When a web site works, on the other hand, you’re using it to do something. You might be looking for your next favorite book on Amazon, or searching for a critical piece of information on Google. You’re using the web site…interacting with it, having an experience that, contrary to logos, involves you. You are inputting information, asking questions, getting answers.”

“So, as a web designer, there is no analog to ‘look at this logo and see how it stands for a company’. That’s relatively easy for graphic designers because we can quickly appreciate the way a logo graphically depicts some attribute of the company: ’solid, blue, Big Blue, trustworthy’. Even if we don’t like the company or if its never done anything good for us, we can make this judgment of the design of the logo.”

“But in web design, we can’t pass such sophisticated judgement on a design without having an actual experience with the web application itself. Without actually experiencing the value first-hand, we can’t look at a web site and say ‘hey, that web site is well designed because it represents the company well’. This is the primary disconnect when talking about judging great web design. You’ve got to experience it in a real way to know if it is great.”

If you want to read more I recommend you start with his “Five Principles to Design By” on his about page.

One of the fundamental decisions for a web designer to make when working on a new site, or updating an existing one, is what canvas size to allow themselves.

At the risk of showing my age, when I first started working on the web one of the key questions was how the site would look for those using a monitor with a resolution of 640×480 (for the younger readers: this wasn’t THAT long ago!)

Thankfully, time and screen sizes moved on.

When we updated the design of Trade Me late last year we increased the standard resolution that we supported up to 1024×768, although with code in place to ensure that the site still works well at 800×600.

If you’re interested load up the home page and try re-sizing your browser window - you’ll notice that, thanks to the magic of JavaScript, at 800×600 the logo and advertisment in the top-right corner jump up above the tabs and the font size on the category links in the body of the page is smaller.

By comparison, Xero, which has slightly more modern browser requirements, uses a fixed-width design which more-or-less fills the screen at 1024×768.

Why is this stuff important?

Well, as it turns out, people are not so quick to increase their screen size or resolution as web designers like to think.

Whenever I speak to a tech crowd I normally ask for a show of hands from anybody who is using a low-res screen (lately that has been 1024×768 or less). It’s unusual for anybody to raise their hand. But, out in the real world there are still plenty of people browsing at that resolution.

In fact many thousands of them…

Using the Neilsen Online stats for September ‘07 of the ~3.5 million unique visitors to Trade Me during the month around 15% of them were using a resolution of 1024×768 (that’s 515,000 visitors) and even more incredibly just over 4% were using a resolution of 800×600 (that’s 145,000 visitors). Many of these people will be using a monitor that is capable of a higher resolution, I’m sure, if only they could work out how to increase it!

At the other end of the spectrum there are a number of different very large resolutions now in use. I’m typing this on a screen that is bigger than could have even been conceived of in the late 1990s, which of course makes life difficult for sites like Trade Me that use a liquid design.

Where it really gets interesting is when you look a little further down the list of browser resolutions. For example, the fifth most common resolution, with 1.82% market share is 214×138. That’s pretty small!

After I mentioned these numbers at the WDANZ conference I spoke at in Auckland a few months back Harvey Kane from RagePank picked up the analysis and came up with some even more interesting findings.

He made the observation that what really matters is the number of horizontal pixels available.

He looked at the Trade Me data and found:

That is a lot of people using little screens! Presumably many of them handheld/mobile devices?

So, there might be a role moving forward for those of us who remember designing for tiny resolutions after all! :-)

The lesson from this…

Design for your audience, not just for yourself, and always remember that they are probably nothing like you.

Here is a nice (old) post from Sam Ng at Optimal Usability about Ben Goodger’s presentation at Webstock last year:

“Ben Goodger, lead engineer for the Firefox browser, obviously believes in the power of the user interface and credits their ‘less is more’ philosophy as one of the key reasons for the browser’s success. As part of this philosophy, they made sure that the interface was clear, removing words or interface elements wherever they could to increase clarity. They also had fewer and more useful options, only including a configuration option if 15% or more of users were likely to change it. And they worked hard on using smart defaults, like turning the pop-up blocker on. All these small changes created a ‘cumulative feeling of quality’.”

From: http://www.optimalusability.com/post.php?postid=31

Designing for the 80% majority (or 85% in this case) is a great idea.

Sam - this is good stuff, what’s happened to your blog?

Last week the DominionPost had a story about Peter O’Hara leaving Fairfax. He was formerly the Editor-in-Chief. I only met Peter a few times during my time at Trade Me, but was impressed each time. He will be missed, I’m sure.

I went looking for a link to the article this afternoon on Stuff.

A search for “Fairfax” on Stuff returned only one result: “Fairfax Magazines Terms & Conditions“.

Not very helpful!

A search for “Fairfax site:stuff.co.nz” on Google on the other hand returned about 415,000 results (in 0.32 seconds no less!)

An on the first page, this article: “Fairfax Media’s editor-in-chief leaves“.

With “Fairfax” in the headline and repeated several times in the article you’d expect this to have been in the site search results.

How is it that Google can index Stuff better than Stuff can index themselves? Stuff only has to index Stuff. Google has to index the whole web. And still they do a better job.

Including search functionality on your site is very important, but unless it works you may as well just refer people to Google.

Neilsen//NetRatings recently announced that they will be changing the way they rank sites they track:

Tyranny of the Page View nearly over?, from Read/Write Web

Who cares? Advertisers use this data to help them understand the traffic patterns of the sites they are advertising on, so anybody who relies on advertising as part of their business model should be paying attention.

And lots of consumer sites are designed, consciously or otherwise, to take advantage of the existing measure. For example, ever wonder why the NZ Herald site regularly splits articles over multiple pages?

At the moment the most popular way of comparing sites in NZ is by unique browsers - i.e. the number of distinct people that visit a site over the course of a day, week or month.

By this measure Trade Me is the clear leader:

Ranking based on total time on site (unique browsers x frequency of visits x average session duration) doesn’t really change the order much, but if you look at the actual numbers you can see just how dominant Trade Me is relative to the other big NZ sites:

Data from Neilsen//NetRatings for June 2007

Note that NZ Dating (the 28th biggest site by page views) and FindSomeone (the 34th) both make the top 10 when ranked by time. TVNZ drops from 7th to 24th.

And, yes, that’s over 366 million minutes we collectively spent on Trade Me during June, or just under 700 years!

Interesting.

Before we all get too carried away though, Joel on MarkerBlog makes an excellent point:

Neilsen//NetRatings quits smoking in favour of harder drugs

Nice!

The videos from the latest Webstock Mini have been posted.

If you’re looking for a bit of light entertainment check out the Web 2.0 debate (a.k.a debatr, a.k.a. LOLDebate), where we resolved once and for all the question of fizz vs. substance. Or, something like that.

Mike Brown’s conspiracy theory alone was worth the price of admission: “Rod Drury. 8 letters. 3 r’s. You do the maths!”

Phil’s slides were featured on the SlideShare home page a few weeks back and are worth checking out (if you’re interested in the substance).

Mike’s subsequent tribute to the “moral superiority, intellectual acuity and all-round web prowess” of the victorious negative team is also worth a read.

Peter Griffin’s column had a nice summary:

“I think the debate came out how it should have, despite the “fizzers” presenting a more compelling and humorous argument than those with substance.

Above all the inane chatter on Twitter, the annoying music blaring at you from MySpace pages and the flying penises in Second Life, there’s something powerful going on in these new web communities.

Whether they will all live on remains a moot point, but one thing is for sure, the new makeup of the internet is seriously changing our approach to information use and social interaction. Whatever price you put on that, such transformation in a few short years has been nothing but substantial.”

PS In other Webstock related news … the NZ team to compete in the FullCodePress Trans-Tasman Challenge was announced today. Thanks to everybody who put their name forward and congratulations to the chosen few. The nations expectations sit on your shoulders. Jeff from Xero has more details.

After a bit of faffing around I finally got Parallels up and running earlier in the week.

So, with access to IE7 again for the first time in a while I decided to try again to buy the Blackbox M-14 headphones I’ve had my eye on (see my previous post for the back story).

I hit their homepage, already anticipating the enhanced quietness that was now only a few clicks away …

Sigh!

So, I installed Flash 8 and, held my breath as I tried again to get through the check-out pages.

Woo hoo, it worked.

But then, this:

Enough!

The next day I walked into the Bose store (which is dangerously close to the Xero office!) and walked out with a pair of QC3s. It was too easy. Offline shopping just works so well.

I don’t want to completely bag the guys working on this site. I got a comment back from Leon at Oktobor to say that they were onto the problems I pointed out last time and were working to make it better. It wasn’t just words - they have re-labeled the buttons on the check-out form to make them more intuitive. And, I got an email from them later in the day to say they had been having some technical problems and I should try again, which at least shows that they are now tracking abandoned orders. But, by then it was too late.

Are you taking advantage of the web platform and constantly making small improvements to your software?

Or have you got an excuse for why this is too hard (your application is too complex, you can’t easily deploy changes without interrupting users, you can’t afford the testing costs associated with each release, you prefer to focus on the bigger headline stuff, etc, etc)?

Here’s what Google achieves …

“… half a dozen major or minor changes are introduced in Google’s search engine every week, and each change can affect the ranking of many sites — although most are barely noticed by the average user.”

That’s the benchmark.

Yesterday, as I’m sure you’ve heard, Apple announced a version of Safari for Windows and also revealed that Safari will be the platform for app development on the iPhone.

Josh Catone on Read/Write Web has a good summary.

Understandably there has been a mixed reaction to these announcements.

Here are my random thoughts:

Safari is right on the brink of becoming a browser that web developers need to care about. This might be enough to make that happen. That would take the tally to four (including IE6, IE7, Firefox). It’s like the 90s all over again, which will come as a bit of a shock to developers who grew up in the Internet Explorer dominated era.

Some people have questioned who in their right mind would run Safari on Windows? That’s an easy one to answer: all of the Mac fanboys who are stuck using Windows PCs at work will (cough Amnon, cough Tim) and, I suppose, all of the Windows developers creating apps for the iPhone.

I agree with Jason from 37signals. Creating a separation between the platform (the physical phone and the browser) and the apps which third-party developers create is a smart move.

What do you think?

In the last 12 months Safari has grown from 1.1% market share to 2.1%. This is a simple result of selling more Macs. According to this recent Bloomberg report Apple now has 7.7% market share in desktops and just under 10% in laptops, which is a lot more than I would have guessed. So, what does the future hold for Safari?

And, related to this: what will happen to IE7? As I noted yesterday, it’s grown quickly to ~30% in the first half of this year, but seems to have stalled there.

Let’s hear your predictions for the next 12 months.

We’ll come back to them in a year and see who was closest.

Sam sent through the latest Trade Me browser stats (for May ‘07):

It’s interesting to compare these to previous months: February ‘07 and December ‘06.

After growing from nothing to 30% market share in the first few months of the year IE 7 has now totally stalled. It seems that everybody who is going to get the new version via Windows Update already has.

IE 6 is hanging in there at around 51%. Presumably all of these people have either disabled Windows Update, work for somebody who has disabled Windows Update or are using an illegitimate copy of Windows.

Perhaps, as Robert McLaw suggested in a recent post about compromised web servers, Microsoft’s policy of not patching pirated copies of Windows is actually causing them more problems than it is solving?

In other browser news, the first beta of Netscape 9 was released last week. I was surprised to find there still was a Netscape, to be honest. I’m not entirely sure why they are bothering.

Yellow Pages have put up a beta of their new website:

www.yellow.co.nz

As previously noted they seem to have dropped the “pages” part of their name.

As far as I can tell, apart from a bit of lipstick, the site is the same old site we all don’t love:

It’s still just an online version of the printed directory. Most listings contain only a phone number and an address, not even a description of the business. You actually get more information from the hard-copy where at least they allow graphics etc (although you can view the ad from the printed directory by clicking on the ‘view ad’ link).

There is so much more useful information that businesses could include in their online listing. Imagine, for example, if they allowed restaurants to list their menu.

They don’t cater for browse dominant users. To view the category hierarchy I need to click a link and then I end up in a horrible control which lists every leaf category alphabetically.

Their category hierarchy continues to be a complete dogs breakfast. There are too many leaf categories and consequently the structure is too deep and too fragmented. The structure is also unbalanced. ‘Business Services’ is a top-level category – covering everything from ‘Accountants’ to ‘Security Guards’. ‘Funeral Arrangements’ is also a top-level category, although it has only seven sub-categories which hare all leaf categories.

I can’t list a new ad online. Hello?

The search is still way more complex than it needs to be. Google allows me to find any page on the whole internet using a single text box. That should be the benchmark. Really, who is going to search for businesses “within 100 kms”?

They’ve added “My Address Book” functionality, but don’t allow me to add comments. I’m completely disconnected from all of the other people searching on the site. Where is the “other people searching for this found these listings useful” functionality?

Oh well. Maybe in the next version?

In the comments to my recent post about XHTML Scott Mayo asks an interesting question:

“How many complaints you have had about the usability of your website by blind, site impaired or other-wise impaired users. Surely as NZ’s site with the broadest coverage you would have a lot of exposure to such feedback?”

I have to admit that I haven’t had any direct feedback, or any first-hand experience using Trade Me with a screen reader.

I’d be interested to hear from anybody who does.

At TechEd last year we used Trade Me in a demo of the speech recognition features built into Windows Vista, and it worked great.

Amongst all of the other things to consider when creating a new site or page, it can sometimes be hard to get excited about accessibility.

But don’t forget that the single most important visitor to your site is effectively blind (a.k.a GoogleBot).

Design for accessibility and you’ll often get search engine optimisation for no additional charge!

Pete has posted his annual review of NetGuide nominated sites.

Interesting reading!

I notice Russell is claiming bragging rights for having the only site which is fully HTML and CSS-compliant.

I was a bit disappointed to read this comment though:

“One thing I did notice is the number of sites now using XHTML, but still using tables for layout. I’m looking at you Trade Me. It seems so frustratingly stupid, why go to all the trouble of moving to XHTML and not use it semantically?”

From: Validation the 2007 NetGuide Awards

This time 12 months ago Trade Me didn’t even have a DOCTYPE.

That was embarrassing!

Moving to XHTML (as part of the migration to .NET) was a big job and shouldn’t be underestimated. We’ve removed a massive amount of non-semantic mark-up as part of that process. But we’ve also been pragmatic about it. Where it was significantly easier to use HTML tables for layout we’ve used them. The net result is that our pages are now mostly valid and much smaller than they used to be, but still with a lot of room for improvement.

There are a lot of people who are very passionate about web standards. That’s a good thing. But sometimes I think they approach their evangelism with a little too much vigour.

Give people some credit for the improvements they make.

Remember that they are often hard won.

Don’t confuse better for best.

P.S. It was a good night at the NetGuide awards for Trade Me. We picked up the award for ‘Best Trading Site‘ as well as ‘Best Motoring Site‘ and ‘Best Real Estate Site‘. Full credit to everybody who has contributed to those successes and thanks too to everybody who voted. And congratulations to SmileCity for picking up the ‘Site of the Year’ award. :-)

http://www.drivl.com/posts/view/700 (warning: piss take!)

From Dan Cederholm’s SimpleBits.com

:-)

Ferrit is taking a kicking at the moment, from all directions.

Lance has several recent posts:

• Online retail: Ferrit sinks
• Ferrit - it’s worse than I thought
• Ferrit - $36m flushed

(remember Lance, three posts in a row is a rant, four is a crusade!)

Over the weekend Sam predicted that the site would be toast within 30 days of the new Telecom CEO being appointed (Dominion Post, Sat 17 Feb).

Even the mainstream media have caught on:

• Ferrit’s Xmas gives ammo to doubters (InfoTech Weekly, Mon 19 Feb)

If the numbers that Lance quotes are correct then it must be a pretty depressing place to be working at the moment.

It’s a bit premature to perform a post-mortum on a still-warm corpse, but …

Whenever they talk about their product they always seem to focus on the size of the opportunity and how much money they think they can make:

Ralph Brayham thinks the New Zealand on-line retail market is under developed, saying that in the US 5% of shopping is done over the Internet, with this number reaching 10% in the UK, but only $200 million is spent on-line in New Zealand.

From: http://www.geekzone.co.nz/content.asp?contentid=6549

I’d be much more optimistic about their prospects if they were instead saying something like:

We think that it’s too hard for kiwis to find somewhere to buy the things they want on-line. We believe we can solve this problem by providing a kick-ass search engine and shopping cart all in one place, and making it easy.

That would at least indicate that they were approaching things from a users perspective and trying to solve a problem.

Of course, even if they were saying this sort of thing they’d be wrong.

Here is a quick experiment using three of the products featured on the Ferrit homepage this morning:

• Shuttle SN25P (a desktop PC)
• Bvlgari - Blv Eau De Parfum Spray 40ml/1.3oz
• The Memory Keepers Daughter (a novel)

How hard is it to find somewhere to buy these products online from a NZ supplier?

Using Google (a well known and trusted brand, no?) …

Search for “Shuttle SN25P”, and the fourth result is The Computer Lounge (based in Auckland with an online store). The third result is PriceSpy.co.nz, a competitor to Ferrit that has been around for a while.

Search for “Bvlgari - Blv Eau De Parfum Spray 40ml/1.3oz”, and there are options on the first page of results from both Zillion and FragranceDirect.co.nz.

Search for “The Memory Keepers Daughter”, and Dymocks.co.nz is the third result (after a couple of reviews).

Of course, you can also find all three of these products currently for sale on Trade Me:

• Shuttle SN25P Athlon64 CPU, Gforce 6600GT, 1 Gb
• BVLGARI BLV (W) ~ 75ml EDP SPRAY ~~ 100% AUTHENTIC
• THE MEMORY KEEPER’S DAUGHTER By Kim Edwards

So, it doesn’t seem like this is very broken to me.