Conquer the command line with a hands-on bash workshop!

2006-03-20: link roundup: WaSP redesign, iweb tag soup, internet security

  • Designing for: The Web Standards Project | And all that Malarkey: The one where Malarkey talks about the decisions that were taken during the redesign of The Web Standards Project web site. An impressive result, using design and key copy items to associate the site with serious social issues - a site designed to trigger subconscious reactions on the part of the user. Plus it looks nifty. I do find the body text too small and there are some problems with colour contrast*, however Malarky says alternate CSS will probably be added in future to address these issues.
  • iWeb, the new tag soup generator | 456 Berea Street. I'd comment, but then people would just get annoyed that I'm bashing Apple again ;) Seriously though, this is a pity and I hope it's resolved in future versions.
  • Virus watches mouse clicks | (16-03-2006). This doesn't come as any great surprise, since X/Y coordinates can be tracked; but it does disprove the claim (made by certain financial institutions) that a mouse-based login page is inherently more secure than traditional keyboard input. The reality is that a virus can pick up keystrokes and clicks, most banks are happy so long as their customers get a sense of security (and there's a disclaimer that the bank can use to avoid any problems).
  • Speaking of banks and security, the Bendigo Bank has just started advertising a security token system: Bendigo Bank e-banking. The TVC drives me nuts since they show the users firing up IE, which would be their primary security concern ;) It's an interesting move, probably made possible by the Bendigo's customer base being (apparently) loyal and willing to cough up the cash for the token generator. The entire system is still vulnerable to social engineering and so forth; plus the token generators are not exactly subtle - the bank's logo/URL is a bit of a giveaway as to what they are. The most likely attack I can think of would be a coworker installing keyloggers and nicking keyrings; so if you don't trust your coworkers the $99 generator with a pin number is a good idea.

* If you are curious to see for yourself, use Vision Australia's Colour Contrast Analyser to check the colour combinations. Since graphics are involved Firefox's colour check extension probably won't show the problem.

2006-03-14: going on safari: the search for version numbers

Q: When is a simple question not a simple question? A: When you need to get the answer out of Apple's website.

My simple question was this: what is the latest version of Safari? It sounds like a stupid question really, but bear with me here (and keep in mind I didn't happen to have a Mac handy).

My starting point: Friends who use Macs inform me that there are different versions according to which dot-point version of OSX a person is using. I know it's at least up to 1.2; and I've seen people talking about "Safari 2.0" so I'm pretty sure that exists.

So, needing an official source for the definite answer, I hit the Apple site. Being a geek, I make an educated guess at a URL. redirects to, which is all marketing fluff with one mention of 1.2 which I'm then told is out of date.

Having dealt with Apple Australia before, I try I discover this loads with broken images and doesn't appear to have version advice anyway.

Somewhere along the line I try ...well, at least I can confirm v1.2. But, like I said, our resident Machead has already assured me 1.2 is not the current version. Confusion reigns. Perhaps the only way to get the latest version is to install 1.2 and patch/update/whatever it's called on OSX. I still don't have the info I need, so onwards... lists versions of OSX but does not specify which version of Safari they contain. Even the update pages themselves are vague - eg. just says it includes fixes ... [for] Safari rendering of web pages. The detailed information page ( still doesn't mention versions.

At this point I give up on Apple and try a straight up Google search for "safari versions". The first two results we've already seen; the fourth is a beautiful moment in nomenclature: Safari Experts: About Web Browsers (yes, a company that does Safaris has a page about browsers).

I try the third result, 'Safari Developer FAQ' ( It appears to be far too detailed, but out of idle curiosity I happen to click on a question about Safari user-agent strings. Then I notice this: As the list of historical build version information for Safari and WebKit indicates, both version numbers may contain a minor version and possibly a sub-version number as well.

Could it be? I click historical build version information and discover a page with the title "Historical User Agent strings"; and the heading "Safari and WebKit Version Information". I hardly dare hope, yet here is detailed information on versions. For the record: the latest version(s) are 1.3.2 on OS 10.3.9 and 2.0.3 on OS 10.4.5.

That wasn't hard at all! :-]

So what's the moral to the story? Well, first off, dealing with Apple's website gives me a headache - not to mention you shouldn't bother with the website, just go straight to Google. Second, websites need to state information which may seem horribly obvious to the author; because that information may not be obvious or available to the user.

If Safari is managed via OS patches, that's fine - but Apple needs to put that information on the Safari product and download pages. It doesn't have to be front and centre, it just has to be mentioned somewhere; after all I did find the (inaccurate) 1.2 version info way down the bottom of the product page.

Apple is a repeat offender on this one. They seem to assume at all times that you already have detailed knowledge about the product they're talking about (and, in the case of hardware, that you already own at least one). They assume you know their exact terminology for things; for example you don't "update the name attribute" or "change the volume label" on an ipod, in fact you're looking for the article "naming your ipod". As if it's a kitten.

The main Safari product page does not actually state the latest version number, despite talking about Tiger. Nor does it mention the fact that it's an entire version ahead on OS 10.4 compared with OS 10.3. If you don't happen to know your Tigers from your Panthers, there's no reason you'd suspect that people with OS 10.3 can't just upgrade to the latest Safari - but that's the deal, apparently.

So next time you're writing some documentation, remember to state the obvious. You might keep someone from needing a couple of paracetamol and a lie down.

2006-03-13: link roundup: betas, templates, passing acid, superheroes, uk accessibility

Tags: , , , , , .

2006-03-03: when web 2.0 attacks!

Plazes is pretty cool, but if you get the latitude and logitude wrong... suddenly Brisbane airport is somewhere in New South Wales: plazes.beta Plaze:Brisbane Airport/Australia/4007/Brisbane/Banksia Place.

Plus there are some limits to localisation - Australia does not have 'zip' codes, for example; we have postcodes. But then it probably doesn't matter - since most of our television is American, the average Joe probably does know that a zip code is like our postcodes. Plus I'm pretty sure they had to forward the phone number 911 through to 000 (our real emergency number). But I digress.

The real point to be observed is that systems are only as good as the information people put into them. Enthusiasm doesn't replace accuracy, nor does a person's willingness to enter information imply they'll proofread it first.

Still, it's fair to say that I am surprised how well community sites can work. You just have to be a bit cautious: is that Wikipedia entry accurate? ...or is the page in the middle of an edit war and someone just rewrote it with an extreme bias?

Web 2.0: remember, more signal means more noise too.

2006-03-02: IE7: good signs?

I've been completely unimpressed by all the buzz about IE7, with its promises, dictates and preview-preview-alpha-betas. But when both Molly and Malarky get excited enough to post big things about the next beta*, maybe it's time for some cautious hope :)

I wouldn't say I'm excited, but this would have to be the first time I've not felt a sense of dread about IE7.

* See: Microsoft IE7 Progress: Sneak Preview of MIX06 Release and The IE7 MIX 06 release | And all that Malarkey.

CATS online resource launched

CATS: Home: The CATS Project has established a framework for good practice that provides information and resources to assist universities to create equitable access for students with disability and to comply with the Disability Discrimination Act and the Disability Standards for Education.

There are a huge number of accessibility/equity challenges involved in education. The CATS resource not only covers the what and why of accessibility, it goes beyond discussion and into the practicalities. For example, Making information systems accessible recommends that you should Consult with students with disability... and it then links on to a guide on how to actually go about it (Accommodations: Consulting with students with disability about their needs).

Well done to all involved!

2006-03-01: the standards of standards

I've been thinking about the Presentation Zen post Nobody's Perfect. Thankfully not due to some presentation going horribly wrong, but instead because it reminds us that we can't get a perfect score every time. Standards advocates would do well to keep this idea in mind.

Because we shoot for the highest standard, sometimes we are too inflexible - unwilling to compromise or accept that something less than perfect may still be an excellent step forward. We can also fall into the trap of trying to say no instead of trying to say yes. Worst of all we might swap high standards for a high horse (or possibly a shetland pony).

People may think standards guys like me are critical of other peoples' work (we pretty much have to be), but the truth is we're probably much tougher on our own work. If we're not careful we can end up focussed on what we haven't achieved, without giving ourselves any credit for what we have achieved.

making hard work of it

I fell into the "but it's not perfect..!" trap a couple of weeks ago. I had a bad week: good work being hijacked with bad work; copping personal attacks for professional duties; getting a reputation for being "anti" because I don't change my opinion of a bad product just because some time has passed.

Everyone has those bad days where you think about chucking it all in; but ultimately I came back to the realisation that I didn't choose the web standards path because I thought it was easy, I chose it because I thought it was right. That's right sportsfans, I'm a true believer.

What I hadn't done was take my own advice and stay sane in the middle of it all. So I took stock, reminded myself that we have actually made quite a bit of progress and it's just my impatience (and a frustrating week) that's getting me down.

so are high standards really a problem?

In general, high standards are a good thing - after all, if you aim low then you'll never hit a high target. The trick is to aim high but still within the bounds of reality; and not to aim so high we can't accept the odd compromise.

Sometimes we just have to cut ourselves some slack about where we aim and whether we fall short.

We're not perfect, the industry is not perfect and we're not going to attain worldwide web standards perfection tomorrow (probably not ever, actually). We have to accept the imperfections of our industry just as we have to accept our own flaws.

Blog Archive