If you are getting this via RSS and want to stay subscribed, you’ll want to update your feed subscription to point to the URL https://davidmallard.id.au/feed/ – at the moment this redirects to FeedBurner, but soon I’ll be updating the server configuration so the feeds will be delivered directly from my website.

No promises that there will be new content coming or how frequently you might get updates, but with the fragmentation of social media communities and the experience of seeing how publishing on third-party sites can come undone (e.g., a decent chunk of the articles I wrote back in the day and linked to from here have disappeared from the Internet), if I do want to publish anything substantial then it’s likely to happen here.

The post Updating this site’s RSS feeds appeared first on David Mallard.

What and Why

What’s HTTPS?

Web traffic has traditionally been sent between the server and your web browser using HTTP, the Hypertext Transfer Protocol. With HTTP, the content being sent – both the web pages you’re viewing and any data you’re sending (e.g., form submissions) isn’t encrypted. If someone was to intercept those packets of data as they’re being sent across the internet, they could view the contents. They might even be able to modify or replace them.

Secure HTTP, which might be referred to by various relevant abbreviations including HTTPS and SSL/TLS, encrypts all of the data that is communicated between server and client. Everyday internet users have been aware of HTTPS for years, especially in terms of financial transactions. Browsers have shown a padlock icon to let the user know that their connection is secure.

Why switch to it?

As general awareness of privacy and security risks has grown in recent years, there has been a growing movement for all sites to enable HTTPS and to redirect anyone connecting via HTTP so they receive an encrypted connection instead. Here’s some general reading about the arguments for encrypting the web from the Electronic Frontiers Foundation and Mozilla Developer Network.

In short, switching to HTTPS is the responsible thing for website hosts to do. And what’s more, it’s now easy and cheap, thanks in large part to Let’s Encrypt, which allows you to get a free security certificate for your site(s).

And if you’re not convinced that there are enough carrots to justify the switch, there’s a big stick looming. While Google has already been rewarding sites that serve encrypted content in their search algorithms, they’re now beginning to flag sites served over HTTP as ‘Not Secure’ in Chrome.

And what’s HTTP/2?

It’s the recent major update to the HTTP protocol, which had seen limited changes and improvements since the early development of the World Wide Web during the 1990s. Without attempting to dig too deep into technical details, it changes the way web clients/browsers and servers communicate to remove some shortcomings and performance limitations, and to allow things like allowing servers to push content to the client/browser without needing it to be requested. If you want to get into the details, you could start with Rachel Andrew’s “Getting Ready for HTTP/2”.

Why switch to it?

Because it offers improved performance for clients/browsers that support connecting with it, and the server will fall back to serving via earlier HTTP versions to clients/browsers that don’t support it. It’s a win-win.

The Process

To avoid breaking anything that would have required me to wipe out my work and restore the original server configuration, I worked through the upgrade in stages. It allowed the sites to remain live through the whole process, and while there were a few wrinkles along the way it all went pretty smoothly and didn’t take long at all.

One important thing to be aware of is that although an encryption requirement wasn’t included in the HTTP/2 standard, some of the most widely-used web browsers will only use HTTP/2 over secure connections. This means the migration to HTTPS needed to come first, then HTTP/2 after the server was delivering encrypted communications.

My server hosts multiple separate WordPress sites, so I migrated one site at a time. Once I had a site fully set up to be delivered via HTTPS and HTTP/2, I repeated the process on the next one.

The general sequence for each site was:

1. Enable HTTPS, then configure and test the WordPress site with HTTPS enabled.
2. Enable redirection of HTTP traffic to HTTPS, then test and fix any issues with the WordPress site until all pages display correctly and the browser’s address bar indicates a secure connection.
3. Enable HTTP/2, then test that the site is being delivered by HTTP/2 where possible.

Enable HTTPS, configure and test in WordPress

The first thing you need is an SSL/TLS certificate for your site via Let’s Encrypt. The recommended tool for this is Certbot, and the commands to install it are in the Certbot instructions for nginx on Ubuntu 16.04.

Once Certbot is installed, the key command to start generating your certificate and/or updating your server configuration is sudo certbot --nginx, which initiates a pretty straightforward Q&A process.

As mentioned above, I migrated one site at a time, so when I ran Certbot I had it generate a certificate for just that site’s domain name (e.g., davidmallard.id.au) and the www-prefixed version (www.davidmallard.id.au). Importantly, at this stage I told Certbot not to redirect HTTP traffic. I then visited the site by typing the HTTPS address into my browser’s address bar to test that the basics were working. I then manually typed the site address (with https:// as the protocol prefix) to test whether the basics were working.

This all went fine for me, but if there are any issues you might need to:

1. Restart nginx (sudo systemctl restart nginx).
2. Look at the changes Certbot made to your server configuration file by viewing and/or editing the relevant file in /etc/nginx/sites-available.

WordPress itself then needs some extra configuration. For most sites, this meant going into the Dashboard, to Settings -> General, and updating the WordPress Address (URL) and Site Address (URL) fields to replace http with https. But one of my sites is a WordPress Multisite (Network) installation, and this required manual editing of the database instead – if you have to go through that process, back up your database first and then you can draw on the “Changing the URL directly in the database” instructions.

Enable HTTP redirection and test

Once I was as comfortable as possible that forcing HTTPS wouldn’t break the site, I ran the Certbot command again, telling it to use the existing certificate (which had been generated in the previous step) and this time telling to to enable redirection.

Once that was done, I visited the site with http:// prefix and checked that the site loaded and the address in the browser’s address bar had switched to https:// (ideally with the green padlock indicating a secure connection. Again, this worked fine for me but if redirection failed I’d look at restarting the server and checking whether Certbot didn’t get something right in modifying the server configuration file.

Once I established that the redirection was fundamentally working, I navigated around the site to test whether everything was being served securely, including things like:

1. If the homepage has a link to another page with a hard-coded address http://, follow it and check that you end up with a page served over HTTPS.
2. If a page is displaying an HTTPS address in the address bar but doesn’t have a clear indication that the site is secure (e.g., green padlock) then there might be some mixed-content issues. I had to do some extra work with some sites, and found that opening the site in Chrome, then opening DevTools (Command-Option-I) and viewing the Security tab (you might want to reload the page with the tab open) would let me find the issues. In some cases it was HTTP URLs that had been entered into the theme settings that I fixed via the Customizer. In one case a parent them I was using had hard-coded loading of Google Fonts via HTTP and I had no choice but to manually edit the theme file to replace those addresses.

WordPress also has a FORCE_SSL_ADMIN directive that can be defined in the wp-config.php file. Once the web server itself is redirecting all traffic I don’t know that it’s necessary, but once redirection was working I enabled it anyway.

After all of this configuration and some tinkering to deal with problematic theme settings/code, everything should be being served over HTTPS!

You can get a security analysis of the site by visiting https://www.ssllabs.com/ssltest/ (if all has been configured correctly, you should expect to get an A).

Certbot should have automatically set things up to renew the 90-day Let’s Encrypt certificate before it expires. You should run sudo certbot renew --dry-run and check that it doesn’t alert you to any problems. You can also check out the file /etc/cron.d/certbot to see the cron task that will run to check the certificates.

Enable HTTP/2 and test

Once all of the HTTPS configuration is in place, enabling HTTP/2 doesn’t take much more:

1. Edit the relevant server configuration file, eg sudo nano /etc/nginx/sites-available/davidmallard.id.au, find the line(s) that have listen 443 ssl; and change to listen 443 ssl http2;.
2. Restart the server with sudo systemctl restart nginx.
3. Test the site address at https://tools.keycdn.com/http2-test

What’s next?

That’s it for turning a somewhat-neglected server into one that’s running on modern web protocols that deliver benefits for security, performance and search engine optimisation.

My next steps are to overhaul my personal site. Jen Simmons has been promoting a great initiative with the hashtag #newwwyear and I’ve set my goals to get a lot more done in the next couple of weeks:

https://twitter.com/tobiasziegler/status/944096052918943744

Next up will be a mix of refreshing site content, wiping out the current theme and starting with a fresh redesign (so expect that things might begin to look a bit weird for a little while), and adding IndieWeb functionality to the site. I also still have some server configuration to do, especially in terms of enabling caching for the sites.

The post Switching a WordPress server to HTTPS and HTTP/2 appeared first on David Mallard.

They’re mostly low-traffic sites and have never needed anything special in terms of configuration, but it was past time that I did some important and straightforward improvements, including HTTPS (thanks to Let’s Encrypt) for all traffic and enabling HTTP/2.

Upgrading the Server

The server was still running on Ubuntu 14.04, which I kept updated but still had the default older versions of software packages, so I wanted to switch to the latest Long-Term Support (LTS) version, 16.04, which would also bring with it PHP 7.0 and other improvements.

When I created the original server I made sure to make the disk image take up less than half of the storage space available on my Linode account. The server didn’t need any more space and it comes in handy for duplicating the image as a backup before doing anything experimental. And for this upgrade, it also meant I could create a new Ubuntu 16.04 server image configured with the same IP address so I didn’t need to mess around with updating DNS records for the sites.

The migration steps were:

1. Export database contents for migration, which I did using mysqldump to export the whole DBMS.

2. Download the database backup (using Transmit for SFTP), all of the files from each website (in my case, the entire contents of /srv/www) and the nginx server configuration files (from /etc/nginx/sites-available).

3. In Linode Manager, shut down the server, then deploy a new 16.04 image and boot it. Here’s the main drawback to my approach, because from this point until I got everything configured on the new server the sites were unavailable, but an hour or two of downtime wasn’t a critical issue for me.

4. Set up the new server, which pretty much involved working through the steps in a few of Linode’s guides: Getting Started, Securing Your Server, Configuring UFW (including a rule to allow https as well as www), and then Installing a LEMP Stack (although I used MariaDB instead of MySQL).

5. Upload all of the files I had downloaded at step 1, move the website files and nginx configuration files into place, and Restore the DBMS from backup.

Extra Steps and Gotchas

That process just about had everything in place, but there were a few things that needed extra work along the way:

• The new database configuration defaulted to expecting all of the database user accounts imported from the backup (I create a separate one for each WordPress installation) to have user logins on the server itself. I connected as root (sudo mysql -u root -p) and then set native database authentication using commands similar to option 1 in this StackOverflow answer, but inserting the username for each of the websites’ database user accounts.

• Figuring out which PHP extensions the WordPress installation would need took some reading and experimentation. Note that you can use php -m to get a list of installed modules, which might be worth running on the original server as a guide. Everything seems to be working with what I’ve installed at the moment, although when I first got the sites running I found that Jetpack wouldn’t work properly because I hadn’t installed php7.0-xmlrpc.

Restarting the Websites

Once everything appeared to be in place, I enabled the server blocks (and also removed the default site – sudo rm /etc/nginx/sites-enabled/default) and restarted nginx.

This got the websites back to functioning as they had before, but on an upgraded server that would support changing the sites to have better security through HTTPS and performance by enabling HTTP/2. I’ll write up how I got all of the WordPress sites switched across to those protocols – and why it’s worth doing – in a separate post.

UPDATE 29 December 2017

I missed one thing that the server needed – setting up a mail server so that the WordPress sites can send email notifications. I used Linode’s guide to configure Postfix to send mail via Gmail/Google Apps using my G Suite account.

The post Clearing out the cobwebs: Upgrading my web server and WordPress sites appeared first on David Mallard.

I seem to have got it (mostly) set up to do work stuff, writing, data analysis, some web stuff, etc., so I figured I’d note down some of the steps I’ve gone through in case I need to repeat them or in case they’re useful to anyone. I’ll start with just a listing of the software I’m running and look to document some of the steps I’ve gone through in other posts.

I’ve attempted to go free and open source wherever possible, but will look at commercial software where it’s essential or genuinely worth the money.

App Installation Options

Some things are available through the Mac App Store, some were directly installed, and then I also set up Homebrew for installing some command-line things.

Mac App Store

Things I’ve installed through the App Store:

• TweetDeck, because I’m used to it and although I use Tweetbot on phone/tablet I don’t see the need to get it on desktop at this stage.
• Evernote, but then I uninstalled it and went for the direct download because I was having trouble getting the Evernote Helper to appear in the menu bar on login. No matter how many times I checked the option in the preferences, as soon as I closed the dialog it would reset to unchecked. Direct install has worked fine and in searching I found some comments that suggested other people had assorted issues with the App Store install.
• Skitch, which seems to have worked fine despite the Evernote issues.
• Slack, which I’m using for the WordPress core communications but hope we might explore as an option for work.
• OneDrive, which I’ve switched to now that I purchased an Office 365 subscription. I’ve cancelled my paid Dropbox account and will see whether I’m happy not spending the extra money, or can resubscribe if I feel like Dropbox is worth it.

Direct downloads/installs:

• Chrome, which I’ve used across all devices including my old Android phone and am happy with.
• Evernote, after the App Store version didn’t go smoothly (see above).
• Office 365, because so much of my work remains Office-based. I’ll look to change that where I can, but attempting to use Pages etc. when our office systems are Windows-based with Office 2013 didn’t seem worth the effort. And the software licence allowed me to get Office for my family plus 1 TB of OneDrive storage so I could cancel my Dropbox subscription, making it almost cost-neutral.
• iTerm 2, which was recommended around the place as a good replacement for the default terminal. I’ve also set myself up to use zsh rather than bash, which I might try to document at some stage.
• Atom as my code editor. I haven’t installed any extra plugins at this stage but will look at whether there are things worth adding to it.
• Github for Mac to handle version control. I’m aiming to do some more coding work, both for web and for data analysis, and the Github client seems fine for managing repositories.
• R and RStudio to do stuff with numbers. I left SPSS behind when I moved out of academia and I’m intermittently learning to do things in R. I’m running the preview version of RStudio which seems perfectly stable for my purposes.
• QGIS for mapping. Doing GIS work and geographic data visualisations is on my list of things to do, so a free and open source GIS app should come in handy.
• Tabula for extracting tabular data from PDF files. There seem to be some Java issues with Yosemite so there’s a special bundled version, which seems to be working fine.
• GIMP and Inkscape for raster and vector image work, respectively. I’ll see how they go and if I have too much difficulty with them and/or feel I can afford a Creative Cloud subscription then I may go for the commercial option in this area.
• Skype for talking to people, which is nice because I am away from home.
• Spotify, which I signed up for a few months ago and have absolutely loved. I rarely bought music, whether physical or digital, but a paid streaming service means I can find a bunch of new or old stuff that I’m interested in listening to, have it pre-loaded over my ADSL connection, and then listen to it when I’m commuting to work or going on a seven-hour trip to spend some time at home.

Command-line and development tools:

Installed so far so I can begin figuring out what the hell I’m doing:

• node.js and Yeoman
• chruby and ruby-install
• Vagrant with VVV for WordPress development

Wish list

The main thing I would like but don’t have at this stage is a good all-round writing app. I’m looking at the new version of Ulysses, along with its iPad counterpart, for that.

Now that I’ve got most of the software set up and have some time to learn how to use it, I’m hoping to turn my hand to some new projects. And I’ll aim to write some more about what I’m up to on this site. Any tips, suggestions and comments would be gratefully received.

The post Got a Mac appeared first on David Mallard.

• At the end of 2012 I left my ivory tower career behind and resigned from Charles Sturt University.
• For the past year I’ve been working as a Policy and Legislation Adviser with Jan Barham. I also did a little work while on leave from CSU for Jeremy Buckingham.
• In August and September I took a short hiatus from my job and contested the federal election (big thanks to First Dog for the drawing I got to use to announce my candidacy).
• Home base remains Orange but work keeps me in Sydney from Monday to Friday most weeks.

I haven’t written much during that time, outside of work. As we get to the end of a year I have some ideas in mind for things I’d like to approach differently and ways I can improve my sense of wellbeing. I’m not a fan of resolutions but do like to have goals and some semblance of a plan for how to achieve them. And writing more, especially on this site, fits into that.

Although I don’t expect to formally publish much work for the foreseeable future, I do think communicating some of my ideas and experiences helps me by opening the door to more reflection and interaction. I still hang about on social media a bit, but I intend to start busing this as more of a home base for thoughts that won’t fit into a tweet.

Someone linked me to a post by Frank Chimero the other day about fragmentation and his plans to build a personal site – “the internet version of a quiet, cluttered cottage in the country.”

I have no more idea than him about what that will look like. It will probably be different for each of us. And I don’t expect I’ll really know until it’s done, which it never will be. But I’m starting by modifying the site’s setup to suit the things I’d like to do. Once that’s done I’ll be in a position to post more, and I hope to follow through on that.

Stay tuned. Meanwhile, enjoy Christmas time and have a happy New Year.

The post Previously… appeared first on David Mallard.

Villar, G., Arciuli, J., & Mallard, D. (2012). Use of ‘um’ in the deceptive speech of a convicted murderer. Applied Psycholinguistics, 33, 83-95. doi:10.1017/S0142716411000117

I’m also in the March issue of the King’s Tribune, writing about the Murray-Darling Basin Plan and the challenges of water policy.

The post A couple of new things in print appeared first on David Mallard.

The Tribune is available in print via newsagents in Sydney, Melbourne and Canberra, but you can also subscribe via the website and have three months delivered for $25.

The post Why I signed an open letter to silence Monckton appeared first on David Mallard.

On the upside, it’s no longer paywalled.

The post The NSW filibuster appeared first on David Mallard.

Although there has been a great deal of research looking at how accurate people are at lie detection and just how they go about it, there has been relatively little work done on lie detection in cross-cultural interactions. This is a gap in the literature that needs to be addressed, because (i) decisions about the credibility of a person from another culture can be crucial in high-stakes settings such as law enforcement, customs and immigration, and (ii) research and theory suggest that certain factors might lead to bias so that people from another culture might be regarded with extra suspicion. The issue of bias does not necessarily involve prejudice by the people making credibility judgments – the fundamental issue is that people of different cultures will have different social norms that determine their nonverbal behaviour, such as how much eye contact is appropriate. If a person is assessing the credibility of someone from a different culture then the violation of the observer’s norms might trigger suspicion.

The experiment presented in this study used video clips of people who manipulated their behaviour to be consistent or inconsistent with Australian norms. We then gave different information about cross-cultural differences to the participants who viewed those clips and measured the degree to which norm-inconsistency triggered suspicion for the participants in each information condition. What we found is that when participants weren’t told about cross-cultural issues at all, norm-inconsistent behaviour tended to result in greater suspicion that the person in the clip was lying. When participants were told that they were viewing people from a different culture then our results suggest it was possible to eliminate this effect of norm-inconsistency, but only when we gave specific information about what behaviours the participants should expect to be different based on cultural norms. When we flagged that there were cultural differences but didn’t explain what they were, the participants were, if anything, more likely to show greater suspicion of the norm-inconsistent video clips.

Here is the abstract, followed by the citation details:

The expectancy violation model proposes that people infer deception when the communicator violates social norms without obvious cause. However, social norms are culture specific. Therefore, discrepant norms between a communicator and an observer in a cross-cultural interaction might increase the likelihood of inferring deception, and thus resulting in bias. The present study investigated whether informing people about cultural differences in nonverbal behavior could counteract cross-cultural bias in deception judgments. Sixty-nine Australian students were randomly assigned to receive No information, General information or Specific information about culture-specific behavioral norms prior to making credibility judgments of 10 video clips (5 norm consistent and 5 norm-inconsistent). The results suggest that cross-cultural biases in deception judgments can occur but may also be prevented by providing appropriate information. These findings require further investigation but have potentially significant implications in law enforcement, customs, immigration, and broader societal interactions.

Castillo, P. A., & Mallard, D. (in press). Preventing cross-cultural bias in deception judgments: The role of expectancies about nonverbal behavior. Journal of Cross-Cultural Psychology.

The post Deception judgments and expectancies about nonverbal behaviour appeared first on David Mallard.

Compare and contrast two lines from Tony Abbott’s budget reply speech, delivered last Thursday. First:

I understand that government should live within its means, value the money it holds in trust from you the taxpayer, avoid waste and, above all else, observe the first maxim of good government: namely do no avoidable harm.

A solid statement of fiscal responsibility, and framed in a way that might resonate with many voters across the spectrum, not just conservatives. After all, nobody wants to see public funds wasted. But then there’s this:

Little in this budget is quite what it seems. The $1.5 billion in new mental health money is offset by a $580 million cut in Medicare psychologist consultations.

Valuing taxpayers’ money and avoiding waste? That’s good. Cutting spending from psychologist consultations? That’s bad. The problem is, those “cuts” – or, to use another term that’s equally applicable but much less pejorative, “savings” – actually look like a pretty good example of a government avoiding waste and improving value for money.

Here’s a brief rationale for these savings, from the Budget Overview:

The Government is also improving the cost‑effectiveness of the Better Access initiative to save $580 million over five years. Informed by an independent evaluation, these savings will be achieved by:

• more accurately reimbursing general practitioners for the time it takes to provide specific mental health services; and
• rebalancing the number of annual allied health sessions to better align treatment to the needs of people.

The savings have been redirected to new mental health initiatives to better target the Governments investment and to benefit more Australians with mental illness.

Let’s dig a little deeper. The savings come from two changes to the Better Access to Mental Health Care initiative which was introduced, to his credit, by Minister for Health Tony Abbott in 2006. The initiative allowed people who were referred by their GP to receive Medicare-subsidised mental health services, with funding usually available for up to 12 individual treatment sessions per year. The aim was to increase access to the services of psychologists and other mental health professionals for people who were not receiving treatment through the public health system.

The savings listed above propose two reforms to this program. The budget papers detail the proposals here; search down the page for “National Mental Health Reform – Better Access Initiative…”. The reforms come after the independent evaluation of Better Access referred to in the budget papers, which was published in February 2011.

The first proposal would see the usual maximum number of sessions reduced from 12 to 10, saving a projected $174.6 million over five years. Is this going to cut the services available to people who need them? In practical terms, that seems unlikely. As noted by two of the researchers involved in that evaluation, Jane Pirkis and Meredith Harris, in a budget response published online by the Medical Journal of Australia:

The evidence from our evaluation suggests that this will not affect large numbers of consumers; we found that 75% of consumers received 1–6 sessions, 20% received 7–12, and 5% received 13–18.

Their evaluation found that the scheme is accessed by people with severity levels ranging from mild to severe, who access treatment for varying periods. But the majority don’t go beyond six sessions, at which point a GP review is required before the person can be referred for additional treatment sessions. In other words, the cut would probably not affect most people. Would it affect those with severe issues, who are likely to be most in need of more treatment? Pirkis and Harris warn that “ongoing evaluation will be needed to ensure that consumers with complex conditions are not disadvantaged,” but it should be noted that the budget expands funding for other schemes targeted at people with more severe mental health issues. The reform seems aimed at tailoring the Better Access scheme to efficiently focus on the mild to moderate range, while other programs may be more appropriate and, hopefully, accessible in the more severe cases.

The second proposal would reduce the Medicare funding to GPs for their role in developing Mental Health Care Plans as part of the initial assessment and referral process. This reform would save a projected $405.9 million over five years, so it accounts for the bulk of the savings. What’s more, the key point to note is that this proposal does not alter the level of service offered to people with mental health issues; it is a readjustment of the funding given to GPs, which the government says better reflects the work involved. As Pirkis and Harris note:

The rationale for this change comes not from our evaluation but from analysis of the Bettering the Evaluation and Care of Health (BEACH) data, which indicated that over 80% of Mental Health Care Plans were being completed in under 40 minutes. In our evaluation, GPs and allied health professionals had mixed views about the value of Mental Health Care Plans … It seems reasonable to provide a time-dependent rebate which is commensurate with the time spent in preparing the Mental Health Care Plan, although it will be necessary to monitor the effect this change has on the likelihood that GPs will prepare Mental Health Care Plans.

The direction of cause and effect is arguable. GPs might claim their efforts to generate Mental Health Care Plans were ignored by the mental health professionals, while mental health professionals claim the GPs’ plans were not sufficiently well-developed to be constructive in many cases. But the practical reality is that the process does not appear to be operating as originally intended, and the funding should be adjusted to reflect that. As Sebastian Rosenberg and Ian Hickie note in their eMJA article – the one that argues that the budget reforms to Better Access are not adequate, in counterpoint to Pirkis and Harris:

GPs have allowed their role in Better Access to dwindle to that of glorified referrers. No wonder the government is now backing better value services.

These are the “cuts” Abbott lamented in his budget reply. Despite what he implied, the majority of the expenditure reductions are not related to psychologist consultations at all, but to GP services. What’s more, on the current evidence the reduction in sessions for psychological services should not have a practical impact in most cases.

The Australian Psychological Society has objected to the first proposal. The Australian Medical Association has objected to the second. Both objections are to be expected – after all, the APS and AMA represent the professionals who work in psychology and medicine, and they will advocate against reforms that would reduce the funding for their respective members. But on the whole there seem to be few negative reactions to the budget’s mental health proposals, particularly from parties who don’t have a vested interest in securing funding.

In a budget that didn’t offer much that was innovative, these proposals actually seem to demonstrate a solid, responsible approach to fiscal policy. There are many reasons to criticise the Labor government’s handling of mental health during its first term. But in this case they have proposed reforms that, based on evaluation and evidence, look likely to reduce spending without compromising the practical integrity of the scheme. In other words, they’ve proposed an improvement to the efficiency of government spending in this area. Along with the addition of much-needed funds to other parts of the sector, it’s a solid outcome.

Meanwhile, it’s been widely acknowledged that Tony Abbott’s budget reply had very little to say about the budget at all. But in this area he actually did offer an opinion about a budget detail. It just happened to be an opinion that seems at odds with the principles he claimed were central to good government.

DISCLOSURE: Since I mentioned the APS, I should note that I was a member of the Australian Psychological Society but allowed my membership to lapse several years ago. I am not a registered psychologist and do not provide psychological services.

The post Abbott’s budget reply supports avoiding waste, opposes cutting inefficient spending appeared first on David Mallard.