2013 update Mapping government policies online - Govmonitor, a great new aussie site

For all the attention on government policies, the various announcements and documentation on political party sites, it can be very difficult to compare and contrast where different parties sit on different issues and, for governments, difficult to keep track of whether they are sticking to their election policies or amending them for pragmatic, political or other reasons.

While the capacity to provide quick and easy insights and access to party policy statements online is technically possible, it isn't often done. Even traditional media outlets tend to turn it into a shopping list or a tool for punishing parties rather than a tool for informing the public and improving policy discussions within and outside parties.

That's why prior to last election I participated in a Google doc project to map the policies of various parties, which prompted some very interesting conversations, but has not been maintained.

I suspect it is also part of the motive behind the latest attempt to 'crowdmap' the policies of political parties at govmonitor.org

The Govmonitor site (http://govmonitor.org)

This, however is a far more visual, accessible and interactive approach than the prior collaborative document idea, providing for easier searching and visual identification of what policies and positions parties support, don't support and haven't made a decision on yet.

The site offers a range of ways to view content, by party, by issue and by topic, with a full text search as well.

It also provides an easy way for people to contribute, adding party policies or positions on issues complete with evidential links and references supporting the party positions.

This is an excellent example of Gov 2.0 in action, providing information and education through evidence-backed crowd-sourcing to support people to identify the parties their views most correlate with.

It is also a great first step as a site, with the potential to expand to support robust issue-based discussions and allowing individuals to state their positions and connect them to like minded people. There's also quite broad international potential as the same approach can be applied to any level of politics anywhere in the world where citizens have a role in selecting their leaders.

Chris Doble has done a great job with this site and I hope it gains increasing attention and traction as we move closer to the next federal election.

2013 update Are Australia's web developers failing to deliver accessible websites?

In a recent story in ITNews, Accessibility checker surfaces errors, John Hibbert claimed that a new Mental Health website, www.mindhealthconnect.org.au, operated by the National Health Call Centre Network and funded by the Department of Health and Ageing, didn't meet the Australian Government's minimum web standards.

Based on a review using the ACheck tool for the minimum WCAG 2.0 'A' level of compliance, John reported that the checker:

highlighted two known problems, 245 "potential problems", 20 HTML validation errors and 115 cascading style sheet problems on the site.

I tend to always take the results of these tools with a grain of salt. Many of the reported validation errors and style sheet issues are often repeats of one single issue, or are not really issues at all, and the two known problems would take a couple of minutes to fix and do not pose direct accessibility risks at all.

However this article does highlight a concern I've had for several years - whether Web Developers, contracted to produce these sites for government, always have the appropriate skills and knowledge to develop accessible websites.

I've seen this type of issue repeated a number of times. A policy or program area, possibly with support from a central communication or IT area, goes out to tender for a website. Web Developers respond, get assessed and the successful tenderer goes about creating the site.

A few months later the site is complete with days to spare before the Ministerial launch - but fails accessibility testing by the agency.

"We didn't understand how important accessibility was to you" says the Web Developer. Note that I was in the room when these exact words were said to an agency by a reputable web developer regarding a website which was developed iteratively and we'd been giving them feedback about accessibility for a number of weeks.

So what happens next?

If accessibility was not explicit mentioned in the contract, the Web Developer asks for more cash to meet the requirement, even though it is a baseline requirement for all government websites across Australia, and says it won't be ready for launch.

If an accessibility level was explicitly agreed to in the contract, the Web Developer grudgingly assigns a junior developer to 'sort it out' - with a vague promise that it will be done in a few weeks or months.

The agency is left having to launch a website which doesn't meet the minimum and fix it as soon as possible afterwards - all because the Web Developer didn't recognise and act on the legal requirement for accessibility.

Of course there's many examples where Web Developers have done exceptional accessibility work for agencies, however I have seen and heard too many issues where professional Web Developers didn't understand the accessibility requirements of governments.

Delivering an inaccessible website to a government agency will cause that agency to break the law and expose it to enormous risks of legal damages. No vendor should ever put their client in this type of position knowingly, particularly where it is so easily avoidable.

My view is that any Web Developer that doesn't deliver a government website to at least the minimum accessible standards (unless otherwise explicitly agreed to by the agency in question) should not receive any payment until they have addressed all accessibility issues.

They should also lose their right to bid for other government business until they can prove they have fully trained their staff on accessible web design.

These may be harsh and strong measures, and I doubt they will be considered due to contractual and practical issues.

However if a vendor contracted to sell a government agency a car that turned out to not be street legal or rent them a building that turned out to not meet the building code, government would walk away without paying and ask for damages, plus be very cautious about working with that vendor again.

Why should it be any different with illegal websites?

2013 hot Is the Australian Government really slow to update staff to modern web browsers?

One of the concerns I faced when working in government, and that I know many other people faced as well, was the currency of the web browser(s) available for use by staff.

Some agencies still used Microsoft Internet Explorer 6, a ten year old browser that isn't supported by many major websites and online services and that even Microsoft admits is insecure and out-of-date. It is now used by only 0.8% of Australian web users.

Statcounter research - Web browsers used in April 2012
|in Australia and Oceania

Others prohibit access to Firefox or Chrome - which, according to some reports, together now hold a larger share of web browsing by Australians than Internet Explorer, and are also considered by many to be more standards compliant.

In fact Chrome v21 (at 21.1%) is reportedly the most used web browser by version in Australia, followed by Internet Explorer 9 (19.6%) and Firefox 11 (16%).

Why is the selection of browser so important?

There's a few reasons that spring to my mind.

Because the browser selected can limit the ability of staff at agencies to use the internet productively. To source information, monitor conversations online, use modern web services and even access advanced intranet features.

Because it costs more to develop for older, standards non-compliant web browsers - with Internet Explorer 6 compliance often adding 20% to the cost and development time of web sites and intranets.

Because it constrains testing of websites. While some web teams have special dispensation to access every browser for test purposes, in other agencies staff are forced to rely on their personal devices, or simply can't test for modern browsers.

Because there is an imperative on government to not use software more than two versions old - a particular issue for agencies still using Internet Explorer 6 when the current version is 9.


I can understand agencies who are 'trapped in the past'. There's often more important priorities for IT and management - critical systems that need to be managed, budget and resourcing concerns. However if you could improve the productivity and happiness of all your staff with a simple software upgrade which also improves your security, well...

There's also sometimes technical issues. While web browsers are free, upgrading an entire department isn't. There are dependencies - particularly with SAP, which stubbornly only supported Internet Explorer 6 until recent versions. It costs money to upgrade SAP and to manage this and a browser upgrade across thousands of computers, including any communication and training support required. Agencies, with other priorities, may put off this work as long as they can.

All this aside - how are Australian Government agencies actually doing in terms of how modern their web browsers are. Are the majority still stuck on Internet Explorer 6 or a similar old and insecure web browser?

As part of my FOI request on social media in March, I asked agencies which web browsers they used, as it impacts on which social media tools they can use.

The exact question was:

Which web browsers are currently mandated and/or supported for use by your agency's staff when using agency supplied IT equipment as specified below?
(Please tick applicable web browsers or supply by email a copy of the documentation on your Standard Operating Environment detailing this information)

While some agencies may regard this as confidential, please note the web browser type and version can, in most cases, be detected by any website visited by your staff.

Aside from two agencies who told me that this was "commercial-in-confidence" information they would not release, most agencies were very willing to provide this information.

I've aggregated the results in the chart below based on the 65 legitimate survey responses I received (the easiest information to analyse). Other (non-survey) responses haven't been included due to the analyse time required.

Web browsers officially mandated by Australian Government agencies
for use by their staff - sample from 65 agencies.

Looking at this response, many agencies supported multiple web browsers - generally Internet Explorer and one other.

Few remained on Internet Explorer 6 or 7, and most sat one version behind the most recent released web browsers - such as on Internet Explorer 8.

I did make one error. I forgot to include Blackberry's browser as an option for mobile phones. This is used as standard across all Blackberry mobile devices, so can be considered a standard.

So overall, how did Australia Government agencies do?

Very well in my view - and better than I had anticipated.

While a few agencies (including some very large ones) still lag back on Internet Explorer 6 or 7, most are using acceptably modern web browsers, even providing a choice in many cases - which helps compensate for some of the minor niggles in some browser versions.

Note
You can now view (and analyse) survey responses from my social media FOI as well at: https://www.surveymonkey.com/sr.aspx?sm=mE_2bvebDvXLOHHKrCnKP79IaCNkFpWjDocQJAN5aEvCQ_3d)

2013 hot The right way to release a mobile app - Human Services' new student app

I'm pleased to say that with all the apps now being developed by Australian governments, the Department of Human Services' new 'Express Plus Students' App, has managed to address almost all the criticisms I've had previously regarding government mobile apps.

What were these criticisms? And which did the Department fail to address? Read on...

Have a clear purpose

The first criticism I have about government mobile Apps is that sometimes they seem to be created without much thought about whether they actually are needed at all.

It is important to resist any urges to create a mobile App simply because you want to make one (as a shiny toy, for experience or credibility), or a senior manager wants to look good to their peers or Minister.

There are aspects of government business which, frankly, the community just isn't interested about. Apps, particularly in government, need a reason - a good reason - to exist, as well as an audience interested and ready to download and use them.

Don't create an App when you need a mobile site

One of the most costly mistakes governments (or anyone else) can make is in developing a mobile App when a mobile site would have met your needs and be more cost-effective.

If you're mainly providing a wrapper around website content and functionality, or providing textual information with a few images and buttons, it is usually faster and cheaper to build a mobile site than a mobile App.

This is because, well, building websites is simply cheaper, and while a mobile App needs to be recoded for every operating system and screen size, a mobile site will work across all internet-capable mobile devices without the coding overheads.

It is far easier to update as mobile site to suit emerging devices - by creating device specific style sheets, which are automatically applied when someone using a particular device visits.


This saves the money that would otherwise be spent in developing versions of your App for different devices and keeping them all up-to-date.

Mobile sites (should if built well) allow you to update the content easily, quickly and cheaply without potentially requiring development time and a user download. Though note that with clever App design this can also be achieved through having a mobile App that presents content drawn from a website or even a text file online.

The worst case - and I have seen it in practice - is when content is hard coded into an app, then there's a need to update it urgently. Frankly it's not easy to push an App through the iStore in less than two weeks, and this is after development. Apps are bad news for urgent updates.

Where your content is mostly words, mobile download speeds aren't generally an issue. It is when you get to video content and sophisticated functionality, or where your users are likely to operate beyond cost-effective 3G or wi-fi range (such as boat owners, remote communities and foreign travellers) that you may wish to consider a mobile App approach actively.

Design to standards including accessibility

When designing apps it seems that many basic usability and accessibility features can get forgotten, with many apps designed to operate in non-standard and non-intuitive ways. There are standards for a reason and standards-based apps will stand a better chance of feeling easy for regular app users to adopt (just like most Windows and Mac programs follow standards).

This means using the design paradigms for iOS, and Google's design principles for Android.

It also means tapping into the accessibility features built into iOS, and Android.

Use inbuilt controls

Using the inbuilt features and controls in mobile operating systems is also important. For example rather than building a map feature, use the one provided on the device.

I have seen Apps where the developer has built all kinds of nifty features that already existed in the operating system. This is sloppy, expensive and rarely results in a better experience.

Built in a reporting system

While you can find out how many App downloads have occurred from most App stores, tracking actual use of mobile Apps requires a reporting system hooked into the code itself.

This is fairly easy to do today, with Google Analytics supporting App reporting, and a number of custom reporting packages available from other organisations that are simply embedded in your App's code.

Having this reporting information is about more than accountability to the Minister, it is about understanding where, when, how and why people are using your mobile App, and helps you build an understanding of your audience so you can keep improving the App - and build new ones - that are even better.

Too many government apps are released without a reporting system, and it's very hard to reverse-engineer one in after release. People who previously downloaded an App can get mighty sensitive about the information you are suddenly collecting plus you miss the initial burst of activity that helps you identify issues and strengths.

Have an official agency account at App stores

This is one of my biggest frustrations, as seeing an official government App listed in an App store as having been created by 'Silly Mobile App Company' instantly reduces the credibility, trust and the ability to actually find the App by searching on the agency's name.

Also when an agency is making several Apps, often each is with a different Mobile App developer due to tender processes or skills. They then get listed under the name of the developer in the App store, which then cannot list your Apps together in a single place ('see other apps from this organisation'), reducing your agency's ability to cross-promote.

Plus, what happens if you make an App with a company, then have a falling out? It can be tricky, even impossible, to get the App out of the developer's account and move it to a new account on App stores.

It seems a no-brainer to me that agencies should register accounts on the main App stores before they start creating mobile Apps. This allows them to register their Apps under their own name, rather than that of developers and to use their reputation to build interest and trust.

Link to your Apps

Due to the wonders of modern technology it is possible to link from your media release and website to your App, as well as to link from your Apps to your other Apps.

Something that agencies still don't appear to do well is to link their mobile Apps together, with an in-App method of downloading other Apps from the same agency, or even government.

Also media releases still lack basic details such as screenshots of Apps or links to them in the App stores. I know it might come as a surprise to some people, but journalists understand how to use hyperlinks, as does the community - and both groups love pictures as much, if not more, than they love words.

Most media releases are read online, not on fax machines - so links can allow someone to get straight to the mobile App without messing around with a search in an App store.

With many releases now read on mobile devices, it makes sense to allow people to click to download the App straight away. It is inconsiderate to force someone to search when they can click.


And that final point is my only criticism of Human Services' 'Express Plus Students' App.

Go to their media release, which has been widely tweeted, and there is no link to the App in the iStore. Hopefully this is an oversight they will fix. It should not take long!

Note that I can't tell if Human Services' App has a reporting system built in either, but I'll give them the benefit of the doubt!


So how has Human Services' App been received by its audience?

This is a great 'good news' story already - with a number of five-star reviews. Check them out yourself at the App iStore (and note that there's more reviews to read if you can click through to iTunes).

Chrome beats Internet Explore in browser stakes

There was surprising news from StatCounter earlier this month when Chrome topped Internet Explorer as the most popular web browser for the week of 14-20 May.

Sourced from CNET: Chrome now world's top browser, but beware the math

While this is only one of the services reporting browser use, represents only one week and is a global figure - so may not represent the situation in specific countries (such as Australia) - it is indicative of the changes underway in the web browsing habits of people around the world.

All major international reports on web browser usage have reported that Internet Explorer has been on a downward slide for several years, with Chrome or Safari picking up most of the market share shift and Firefox and Opera being limited beneficiaries.

While this reflects the growth of mobile browsing (Apple iOS uses Safari, Android devices use Chrome), it also represents a significant change in desktop and laptop computer use.

While corporate and government organisations remain major uses of Internet Explorer due to its lead in corporate management features (though Firefox and Chrome have moved to match these), households are choosing their main web browser based on speed, usability and usefulness.

Reliable Australian web browsing figures are harder to find - it would be very useful if organisations such as Google or Facebook (the top sites visited by Australians) released their figures.

However I can say that, from Microsoft's figures, Internet Explorer 6 use in Australia has fallen to 1.2% of the browsing public. This is a GOOD THING as IE6 is an 11 year old vendor-unsupported, insecure and standards non-compliant web browser, unsupported by many major websites and which adds, in my experience, 20-30% on the costs of any web development project.

I should note that Microsoft is trying to end the use of Internet Explorer 6 and has even begun taken steps to automatically upgrade people to more modern versions (beginning with Australia and Brazil).

You can learn more about Microsoft's campaign to end IE6 at their website, The IE6 Countdown.

Sorry if you are one of the remaining organisations using IE6, however my FOI request on web browsing and social media use across government has revealed that largely agencies have made or are making the move to upgrade.

From the now 65 responses I've been able to analyse, only 7 (11%) indicated they still used IE6 on desktop computers. While this is quite a bit higher than the national rate (1.2%), it is much smaller than I had anticipated. Of course if this includes large agencies the percentage of APS staff using IE6 may be significantly higher.

I've provided a breakdown below of the browsers that government agencies indicated they used.

Notes and caveats

• this represents 65 agencies, large and small, of 166 approached - so is representative but not population data
• many agencies used more than one web browser, so the figures don't add up to 65. 
• I've excluded browsers that no agency indicated they used (and I asked about all major browsers back to the time of Internet Explorer 6's release). 
• I forgot to ask about the use of Blackberry's browser on mobile phones - essentially every agency using Blackberries use this browser.