Unpacking the gig economy impact on The Future, This Week

Disclaimer We'd like to advise that the following program may contain real news, occasional philosophy and ideas that may offend some listeners.

Sandra So Kai what should we talk about this week?

Kai Well, there's been some interesting news around electric vehicles. When we talk about electric vehicles, Tesla always comes to mind, and we widely regard them as the market leaders and certainly technology-wise, they're right up there even though some of the German companies tend to catch up. But the news is that Tesla's cars are by no means the top selling EVs in the world. And we're looking to China again. So apparently the Hong Guang Mini EV, sold only in China, has outsold Tesla's most popular model, which is the Model 3, by a long mile. So there was 36,000 of this Mini EV sold compared to only 21,500 Model 3s. And it's a very basic car. And the point here is price. It retails for about $4,500. It only makes about 170 kilometres on a charge. So in terms of specs much, much lower than the Tesla, but the Tesla Model 3 retails for $40,000 in China, so you can see where the appeal is. And I think it points to the future that in order for EVs to become a real success, we need much more cheap cars and models on the market.

Sandra And I think what you're seeing in China is actually echoed in other parts of the world as well. There's been recent news out of Germany, too, that EV sales are dominated by brands other than Tesla. So Tesla has fallen out of the top five best-selling vehicles there with companies like Volkswagen taking out the first place, and then Smart, but also sales from Renault, Hyundai and BMW, all of them selling better and more EVs than Tesla. And Tesla having had only less than 500 registrations in Germany in January, and most of them Model 3S's points to the fact that they're really falling out of favour, especially in the more affordable range. There was some other interesting news, however, coming out of the US, Amazon workers have been trying to unionise for the first part of this year. But on the back of that there was some interesting news around deep fakes and the fact that deep fake Amazon workers are now active on Twitter and are trying to sow some confusion around work practices that Amazon has. And we just want to make it clear that this is very likely not Amazon trying to impersonate workers on Twitter, but fake Amazon employees nonetheless have cropped up on Twitter. And they are leading to some real confusion around what the work practices at Amazon are. And I think it points to a larger picture of how deep fakes can be used, especially when they're used in large numbers. So these Amazon workers have profile pictures that are computer-generated and contribute to the conversation. But it becomes very quickly very difficult to discern them from real people. And I think the bigger story there is how deep fakes are starting to infuse real conversations when you can do this at scale.

Kai And that resonates with a Wired UK article, which we're gonna put in the shownotes, which basically raises deep fakes as an issue, saying deep fake apps are here and we can't let them run amok. So the democratisation of deep fakes to be put into the hands of everyone, and there's already an epidemic of deep fake revenge porn happening on certain websites. So a real issue, that again, calls on tech companies to weed out these apps and again, something for the content moderation practices of those companies. But I also wanted to raise another story. Sophia is back. Remember Sophia? Back in the day the robot that became a citizen of Saudi Arabia that has been making the tours as a speaker. Sophia has now become an artist and has put in artwork as an NFT, a non-fungible token, on the blockchain that was then subsequently bought for $688,888. That is significant because it was bought by a digital artwork collector known as 888, or crypto888crypto as his Twitter handle. And of course, Sophia is basically a puppet with a machine learning algorithm behind it that has now been used to create this artwork. And to me personally, this marks sort of peak bullshit on the blockchain. It creates a lot of publicity, but we all know neither is Sofia a real person, nor is she an artist, nor is she a she.

Sandra To be fair, we've seen the hype around the NFT movement and we're also slowly seeing it dissipate. However, it is really difficult to tell what is happening with non-fungible tokens because they are illiquid in nature trying to understand how the prices are moving and how much things are worth and in which direction the overall market is moving is really really difficult to tell. However, we're starting to see the first signs that the hype or the craze might be over.

Kai So Cointelegraph reports that the so called floor price in NFTs is undergoing what they call a silent crash. The hype around NFTs has risen sharply at the beginning of February with prices jumping up, but according to their analysis, they have since plummeted or undergone a crash fallen back to prices in January. And we know this because they are tracking the so called crypto punks, which we've mentioned previously. They are little icons faces that are being sold on the blockchain. And their prices had skyrocketed during March, but they have now fallen back to January prices. So a strong indication that the early and short hype might fade and there seems to be some saturation in NFTs at the moment.

Sandra So just to clarify, floor price is the lowest price at which NFT can be bought for a particular project. So in the case of crypto punks, for instance, the lowest price at which you could buy one, and being one of the earliest and most popular NFT is a 40% drop in the floor price for crypto punks does signal that prices might be going down. But also market tracking website NonFungible also reported in the Cointelegraph article says that this cooling of the market can also be seen in the total number of sales, the total value of sales and the total active wallets which are all down on a seven day and on a 30 day basis. And maybe one more story before we turn to our main stories for today. And this has just come in over the long weekend, over Easter weekend. And that is that Microsoft has data around working from home within their organisation. And there's some really interesting learnings in this one. And something that we need to come back to. That is, first and foremost that it seems that the line between work and life is slowly dissipating with them seeing an increase of over 50% in the messaging that happens after 6pm and up to midnight. And there's also been a very interesting finding that's worth paying attention to. And that is that about 60% of leaders at Microsoft describe themselves as thriving in this new online Microsoft Teams-led environment. However, when they started looking a bit further down, it seemed that employees were thriving a lot less than their bosses by about 23%. So a very unequal experience that people are having online depending on the position that they have in organisations and on their ability to make decisions, and so on. So whilst hybrid workplaces seem to work for a lot of people, it seems that the experience that people are having depends a lot on where they are placed in the organisation.

Kai And we mentioned previously in one of our Corona Business Insights episodes that it is, in particular, younger employees and those new to workplaces that struggle in the remote or hybrid work environment with getting to know everyone getting to know how the place works, making connections. And so they would be much less likely to thrive in this environment, than reported by the executives in that study.

Sandra So a topic we definitely want to come back to and analyse in a bit more detail. But for today, we have to turn to the gig economy. There have been so many stories in the news. And we've had them in our short stories since the beginning of this season around Uber drivers being classified as 'workers' in the UK, big acquisitions happening in whether it's the rideshare category, or the food delivery categories, lots of movement, lots of impact due to COVID-19 and very unequal impact on gig workers. So all those stories have actually prompted two different aspects of the gig economy share economy, platform economy, so we thought we actually do need to pause and have a deeper look at this.

Kai Let's do it.

Sandra Let's do this.

Intro From The University of Sydney Business School, this is Sydney Business Insights, an initiative that explores the future of business. And you're listening to The Future, This Week where Sandra Peter and Kai Riemer sit down every week to rethink and unlearn trends and technology and business. They discuss the news of the week, question the obvious explore the weird and the wonderful.

Sandra So we thought we'd start today with a piece in The Conversation from last week titled, "A new deal for Uber drivers in UK, but Australia’s ‘gig workers’ must wait". And full disclosure here, one of the authors of the article is our most brilliant colleague, Alex Veen, who's a lecturer here at The University of Sydney Business School, also a DECRA Fellow, together with colleagues from Edith Cowan University and the University of Western Australia, Tom Barratt and Caleb Goods, has reported on the landmark decision in the UK where Uber lost a five year legal battle, and had its drivers now classed as 'workers' rather than independent contractors. Which is a significant development for what has broadly been titled the 'gig economy.' And the article reports on what Uber will now have to do, the fact that in the UK they will have to pay drivers a wage, an hourly rate, instead of a fee per ride, but only whilst drivers are transporting customers, hence more decisions to follow in that space. But also, more importantly, they will have to pay things like holiday pay and retirement contributions and sick leave, giving drivers a lot more entitlements.

Kai So the article asks the question, what are the implications for Australia? And the problem is in Australia, we only have two categories, employee or independent contractor. So the middle category of 'worker' doesn't exist, which has made it hard in the past to actually definitely categorise Uber drivers as employees, which on certain measures they're certainly not, because they have the flexibility to drive for another company. They do own their own cars, so they have some characteristics of independent contractors. While of course also, as the UK decision shows, have certain characteristics of workers as well, so it remains to be seen what the Fair Work Commission in Australia will come up with.

Sandra And so Alex and his colleagues rightly call for a evidence-based nuanced debate about the changing nature of work and the experiences of people undertaking gig work like driving for Uber, for instance, to ensure that there are good outcomes for workers, consumers, but also for the platforms themselves. And we'll include links in the shownotes to Alex Veen's research, especially around the nature of algorithmic management systems that these platforms employ and their effects on workers and the quality of gig work. And the news reported on in this article has actually been taken up in many, many other articles across the media, and many articles then make generalisations about the changing nature of work on gig platforms. They make generalisations about experiences. But also generalisations about how we should think about policy or regulation in this space. And we thought it'd be a good time now to unpack this, as if we look at gig work or the sharing economy or matchmaking marketplaces, there's actually a wide variety of platforms, and a wide variety of companies doing very different things in this space, with very different effects on the quality of work, of workers, on the security of their jobs, on their long term prospects.

Kai As well as the future and sustainability of these services and platforms themselves. So the two of us have long-discussed the confusion that seems to exist in the media around what counts as sharing or gig economy marketplaces. So what we really need to do is look at what's actually in this market. So what are the different kinds of models, and then what the effects are on, you know, society and workers and the platforms themselves. So let's start by looking at what is actually in this space. And there's a big confusion around, you know, sharing economy, gig economy and marketplaces. And one of the confusions lies in the fact that platforms like Uber can be classed as marketplaces because you have both drivers and riders that come together where the platform then matches those and then offers a service, much like other marketplaces, but it's important to note that the category of marketplaces is much bigger. So things like eBay or Gumtree, or, indeed Tinder for dating or similar platforms, they can all be classed as marketplaces, but there are of very different nature of a more transactional nature where goods and services are offered or people are being matched. And we wouldn't see those as part of what we might call the gig economy.

Sandra And then there's all the sharing assets companies, which is actually where all the gig work started from initially. So initially, we had something called the sharing economy where people with excess capacity in their assets, whether that was their car, or the tools they had in their shed would share that with a community of people. So say you have a power drill that you're only using three times a year, I might be your neighbour, I'd be able to rent that power drill from you instead of buying a new one. And slowly that moved into sharing other things, for instance, sharing cars. In Europe, famously, we have BlaBlaCar, which is a carpooling service rather than a ride-sharing service. So if I say I'm driving from Amsterdam to Paris, I might tell people, 'hey, I'm riding on Saturday, do you guys want to drive with me and share the cost of that drive and share my car' with other riders. However, the sharing economy label has then been co-opted to mean things like Uber, where I'm not sharing in a ride that the Uber driver is doing anyway, but rather, using it as a taxi service to take me from A to B, except the car belongs to the driver rather than a taxi company.

Kai And so we want to set aside the actual sharing companies, and Airbnb would be in that category where you know, I'm sharing an accommodation or even renting out a room is not exactly gig work. We have something like GoGet in Sydney where people can take up a membership and then have access to cars that are available all across the city on an hourly or daily basis. So these are legitimately sharing economy companies. But we wanted to focus in on what we call gig work. So we want to distinguish really two groups here. On the one hand, we have person mobility, such as Lyft, and Uber, and all the various delivery services like Uber Eats and Deliveroo, all of which are transport-related.

Sandra And on the other hand, we have gig work that revolves around connecting people to people who have specific skills or talents. And they might be very simple things. For instance, the ability to hang a picture or stand in line for someone like you would have on Airtasker, to more complex caregiving services. Like for instance, Mad Paws, which was started by Sydney alumni Jan Pacas, where pets are hosted at the sitter's home, or there's pet day-care, or there's dog walking or dog grooming to a lot more complex tasks and creative work where you will hire designers or coders to platforms such as Upwork, or Fiverr. That might be for a very short-term project where you ask someone to design an illustration for you, or indeed recurring engagements or long-term work where you hire people or entire teams to develop software or to develop complex projects that require a variety of highly-skilled talent.

Kai And so we thought in looking at what these platforms do, and indeed, the implications and the impact for workers and the wider society, it's really important to pull apart those two groups.

Sandra And the reason we want to highlight this diversity in gig work models is the fact that it is extremely unhelpful to lump them together when you're thinking about the experiences that people have on these platforms, or the effects of something like COVID-19 on the work performed on these platforms, or even indeed about how you go about regulating them or ensuring that people have a positive experience on them. So for instance, when people talk about something like algorithmic management or algorithmic governance, while they might be talking about using computer technologies to automate processes around decision making, and control of the workforce, algorithmic management will manifest itself very differently depending on whether you're on a mobility or delivery platform or whether you're on a creative work platform.

Kai So algorithmic management on the mobility and delivery platforms pretty much govern everything that happens on the platform. The matchmaking, so finding a driver for your journey, but then also the actual driving is meticulously governed by an algorithm. The price is set by the algorithm. And of course, the scoring and the surveillance of the drivers quality, to the point where drivers often are automatically delisted with an automatically generated message. On the other hand, on creative platforms the matchmaking for example, is more much more difficult. So a lot more logic goes into classifying the diverse range of projects and then finding the right matching skills for that. But the price setting is often supported, but up to the parties to negotiate. And of course, the work itself is not governed by the algorithm because it's creative work that largely happens outside of the platform. So on the one hand, full service coordination by an algorithm of the entire service. On the other hand, a very complex matching between skills and projects. So a very different focus.

Sandra And whilst the algorithmic management and algorithmic decision-making does impact and fundamentally shaped the relationships on both delivery platforms, and task work or creative work, there are stark differences in how this is experienced by workers. And also stark differences in how the business models that these companies have, enables more or less transparency, for instance, or enables more or less precarious conditions, or exploitative or fair work conditions for people who do work through these platforms.

Kai So let's take a look at the impact of some of these gig work models. And we want to concentrate on the transport models, delivery and mobility like Uber and Deliveroo. And while we hear a lot about the quantitative effects around pay and working hours, we think it's important to also mention the qualitative aspects of these models, how work is experienced.

Sandra And the first difficulty here when you start talking about the qualitative side is the problem of there being no widely accepted definition of job quality. So different people looking at the quality of work on gig work platforms are taking different aspects into account. And the difficulty here lies both in how research is done, so for instance the fact that job quality and the literature around it is addressed in business, which is something we will do here in the Business School, but it might be also steeped in economics, or in the psychology literature or indeed, sociology. And it's also dependent on the complexity of the task and of the variety of the tasks that people are doing. So people researching the quality of work on a food delivery platform might be looking at very different things than people researching job quality for design tasks on Fiverr.

Kai And this is really why these definitions are quite important. And we're going to put an article in the shownotes that researches Upwork. The issues there are really with people fearing a negative rating and being delisted from the platform, mentioning things in their conversation with clients that might suggest that they might be bypassing the platform. So it's really about the conversations on the platform and the fear of the algorithm, of being demoted or not being given any further jobs. Whereas on the transport platforms the issues are really very different, because the work itself is algorithmically managed.

Sandra And well add here a research article by Alex Veen looking at the job quality on food delivery work, which revolve mainly around economic aspects of autonomy and enjoyment of the job.

Kai So drivers report on the one hand being constantly surveyed by the algorithm, and also very little in the way of making their own decisions. They can't set their own prices, they can't really decide who to take into their car, nor can they really decide which way to drive. The algorithm, step-by-step governs the entire work process.

Sandra And we should note here that while gig workers, especially in the food delivery and ride-sharing space objectively have little autonomy over their work due to the way they work is managed or controlled. Research, including Alex's work has highlighted that workers subjectively consider the levels of autonomy that they have as an upside of their work, and that they do highlight a whole range of enjoyable work elements, having the freedom to set their own hours or being able to engage in conversation or socialise with other people.

Kai But it's important to note that for anyone who wants to regulate these platforms, it's really not only about whether they provide good pay or job security or how to class workers as employees or contractors, but also to look at the quality of the work experience. And apart from autonomy and surveillance, there's a couple more things we need to say about this type of gig work. One is the way in which performance appraisal is being done, which is largely done through ratings systems. Uber drivers, but also users receive ratings and those ratings are very important to stay on the platform and to be a good member of that platform and being given work by the algorithm. And also progression in their job, which is often lacking in gig work, there's often no upskilling, there's often no career ladder to aspire to, it's just that kind of work that you do.

Sandra Whilst this might not be a big issue for ride-sharing or delivery, it is a very significant issue for platforms that enable more creative work. But these qualitative effects alongside quantitative effects, like pay or hours worked, are only the first order effects of gig work and are unfortunately, also the only part that most media articles seem to focus on. And all the debates seem to be had with respect to these first order effects.

Kai And that's certainly the case for the media coverage around the UK decision. But we want to highlight that when regulators take on these platforms, and try to figure out what the implications are, that there are significant second order effects, not on the workers themselves, but on the environment in which these platforms operate. And for transport-related platforms, that would be our cities.

Sandra And for instance, three years ago on The Future, This Week, we reported on research out of UC Davis that was trying to answer questions around whether or not ride-sharing actually results in less cars on the road or less kilometres travelled by cars, and in turn, would there be a decrease in emissions as many of the ride-share companies such as Lyft and Uber have always argued. And we reported back then on transportation research that showed that if you look at aggregates of the likes of Uber and Lyft, that on the aggregate, instead of removing cars and trips from cities, it was actually adding more trips to cities and to suburban streets, and in many cases the second order effects on things like public transport was to actually cannibalise those systems.

Kai And we've certainly seen efforts by these ride-share platforms to integrate with public transport more. So you can now see public transport options or connect between public transport and ride-share platforms. And there's been a recent trial announcement in Sydney, for example, where the government will promote using the Opal card, which is the local public transport payment card, to use on bike-share platforms, or indeed on Uber to give a small discount and promote commuting via public transport to do the last mile in the city, using these ride-share platforms. So while this is happening, it's certainly the case that research shows that there are more cars on the roads often idling outside of peak hour, not being able to park anywhere, and so driving around looking for riders.

Sandra And not just that, but research has shown that between 50 and 60% of ride-hailing trips wouldn't have been made without the services making them available. So people would not have gotten into cars had those services not been available. So it's important to redo those pieces of research as time goes by to actually measure the effect of ride-sharing on the number of trips that are being done in the city, their effects on infrastructure, and their effects on public transport.

Kai And of course, on traffic on the roads, and with that, also, of course, on emissions, which has implications for the climate as well. And it would, of course, be helpful for this kind of research if these platforms would open up their chest of data and make available data on rides for researchers to do these kinds of analysis, which are possible for the public transport system, for example, based on Opal data. But to get a complete picture would be really good to also get access to this data owned by these private platforms.

Sandra And again, that data is really important to make sure that whatever we do to address first order effects of gig work does not have significant negative second order effects. And again, some research that we reported on this time a couple of years ago, and we'll put all these links in the shownotes was coming out of UCLA Institute for Transport Studies that was showing that ride-hailing apps were maybe benefiting poor and minority communities the most. So they were looking in particular at LA, where there was a pernicious problem in transportation for neighbourhoods where car ownership was very low, and where public transport was not necessarily at its best, where ride-hailing apps actually solved a real transportation problem for poor and minority communities. And so it turns out that ride-hailing services increased mobility, especially for people in low income neighbourhoods.

Kai And the reason was of course also that these areas were underserviced by the taxi system because they were not lucrative enough. And taxis, being more expensive were not accessible financially for these communities. So what's really needed is the complete picture where the interaction between those systems are being looked at as a whole, to make sure that regulation of one or the other system doesn't have negative spill over effects in other areas of the system.

Sandra Or indeed unintended spill over effects. Last year, again on The Future, This Week, we reported on a new phenomenon that was driven by delivery platforms, in which new brands of restaurants were created for these platforms alone, they were called virtual kitchens, or ghost kitchens. These were either done by restaurants that would open a different brand that would be delivery only. So let's say a pizza shop that now suddenly started making burgers, which would only be available on UberEATS under a completely different brand, or indeed, ghost kitchens that would appear solely to service a need identified by the likes of UberEATS or Deliveroo.

Kai And so ghost kitchens can be located off Main Street, they don't have to be in an attractive location, they can be in an industrial park, they don't have a shopfront. They're just basically a kitchen that caters exclusively for the delivery platforms.

Sandra And this was a very significant trend. So in 2019, alone, UberEATS had helped about 4000 virtual restaurants, spring up in the US, Deliveroo was testing RooBoxes in sort of derelict car parks in East London or warehouses in Paris, so trying and get delivery-only kitchens. And we were seeing the same trend in places like China. And with predictions by the likes of Morgan Stanley that this would grow to an industry of 220 billion by 2020. And this is a very significant growth with some unintended consequences on local restaurants.

Kai So for most restaurants, the actual delivery business isn't very lucrative, they often take delivery when they have excess capacity at certain hours as extra business. But because of the large cut that these platforms take, the actual Deliveroo of UberEATS business isn't lucrative for most traditional restaurant or coffee shop owners. And so there have been certain developments where restaurant owners have closed up shop, moved elsewhere, and taken their business exclusively to these platforms. And so what we're saying is that as regulators look at, on the one hand, the working conditions for these delivery platforms, safety of riders, which is worthwhile, it also needs to be taken into account that these delivery platforms, as they grow and roll out at scale, have these second order effects on what the makeup of our cities look like when restaurants disappear and become exclusively ghost kitchens.

Sandra And the effects don't stop there. There's actually third order or spill over effects as well. So whilst the first order effects might have to do with the quantity and the quality of work on gig platforms, second order effects might have to do with the interplay with local communities or public transport or indeed emissions. Third order or spill over effects have to do with gig work models and in particular practices like algorithmic management, finding their way into other organisations, or parts of other organisations, as a way to manage work.

Kai So we might refer to this as algorithmic casualisation, where traditional workforces in say cleaning businesses or retail, which were on contracts and receive their work schedules drawn up a week in advance by a human, are now managed by an algorithm that will allocate work automatically, sometimes only a day in advance, to create efficiencies for the company but also come with similar detrimental effects of being exposed to this type of algorithmic management, as we see on these transport platforms.

Sandra And we'll put an article in the shownotes by the Brookings Institute that makes that point. So before we wrap up today, it's worth thinking about where to next, and the future of good companies and in particular companies like Uber, as we started in the article by Alex Veen and his colleagues, and as he highlights there remain quite significant questions about the long term prospects of these organisations and the sustainability of their business model. The article highlights that Uber lost 6.7 billion US dollars last year. Deliveroo lost less, that was 309 million, and as a whole food delivery platforms have done better than mobility platforms during COVID-19.

Kai And we'll put in the shownotes a fairly scathing long read, almost like a rant, by American Affairs, which however makes an interesting point about the competitiveness of the Uber business model, by looking at it as an entire system that competes with an existing taxi system. And the point that the article makes is that when compared system to system, the cost of the Uber system is actually not lower but rather higher, than the taxi system. Because of the way in which each Uber driver has to purchase and finance their car individually, they cannot benefit from the same economies of scale that larger taxi operators would do. And also that the corporate cost of Uber are significantly higher given how much money they spend in breaking into new markets, or indeed for lobbying lawmakers as recently as the $200 million dollar expenditure to lobby the legislative change in California. So the article basically says that Uber as a system has made a big bet on gaining dominance and then automating the ride experience doing away with the drivers. And of course, it questions critically whether this automation will happen anytime soon, as we haven't made the kind of progress around autonomous vehicles that Uber would need to make that bet worthwhile for its shareholders.

Sandra And here we can come back to the UK decision, which will see the driver share increase and the share that Uber gets to retain decrease, which will make it difficult for platforms to improve their position, given that the financial position was improved not by efficiency gains, but by the ability of these platform services to drive down the take home pay that drivers would get to minimum wage levels.

Kai And so overall, Uber and similar platforms are still losing money, on average, on every ride that is being taken. They're effectively still subsidising rides, so it remains to be seen whether this part of their businesses can become profitable. It looks a little bit better on the delivery side UberEATS has been reported to be profitable by itself. So there's a growth potential there with ghost kitchens and all the kinds of innovations in food delivery for these companies to grow in that part of their business model.

Sandra So indeed, the delivery part of the business for Uber overtook its mobility segment to be the largest driver of net revenue for the company.

Kai So while these transportation companies have certainly brought innovation for users and customers, a ride in an Uber is on average much more pleasant than in a taxi...

Sandra And much cheaper.

Kai And much cheaper, access to good food via delivery has become much more prevalent, which was good during COVID-19 lock downs but also adds to people's lifestyle considerably. We will need to keep an eye on the impacts, be they first order effects on workers, second order effects on cities and communities, or indeed the wider spill over effects on how work is conducted in organisations more broadly.

Sandra But that's all we have time for today.

Kai Thanks for listening.

Sandra Thanks for listening.

Outro This was The Future, This Week, an initiative of The University of Sydney Business School. Sandra Peter is the Director of Sydney Business Insights and Kai Riemer is Professor of Information Technology and Organisation. Connect with us on LinkedIn, Twitter and Flipboard, and subscribe, like or leave us a rating wherever you get your podcasts. If you have any weird and wonderful topics for us to discuss, send them to sbi@sydney.edu.au.