 I work at the Australian Astronomical Observatory in Sydney. One of our flagship projects is Manifest, which is an instrument that we're building for the GMT. And we have a large team of engineers, project managers and scientists working on this instrument. And so we're all pretty excited about being able to contribute to this large telescope project. So I'm going to tell you a little bit about what Manifest does and how it works. First I need to teach you a little bit about how astronomical instruments work. So these very large telescopes, you don't just look through them anymore, right? This went out some time ago. You have to build some instrument that you put at the focal plane of the telescope in order to analyse the data that's coming from it. And there's two types of instrumentation that we can build behind a telescope. The first is an imaging instrument. So here's schematically what I've got here is a telescope. The primary mirror is the secondary mirror. And the blue lines here is light coming from very distant star, very distant galaxy. Bounces off the primary mirror, then bounces off the secondary mirror. And then it comes to focus at this point here, which we call the focal plane. And this is where you form an image, you know, a simple how any simple imaging camera works. And what we find at the focal plane here is a view of the sky. Depending on these optics, et cetera, adjust the scale appropriately. So what we then do is put a CCDRA and probably some form of lenses and then we can take some, you know, nice, usually impressive image. And this is something that the GMT could do. This is what you're familiar with from lots of telescopes like Hubble, for example. However, that's not really what we're about for manifests. What we're going to do is spectroscopy. So for any astronomical source, be it a galaxy, be it a star, be it a nebula, be it a planet, whatever, there's a lot more information that you want to obtain from it than simply looking at what it looks like from an image. And this means that we need to take a spectra of it. We need to split the light up and understand how much red light is in it, how much blue light is in it, how much yellow light, how much infrared light, how much UV light. So what a spectrograph looks like to do this, again we have our telescope here, light's coming in here, bouncing off the primary, bouncing off the secondary, comes to a focus. And then what we do, what we can do is put an optical fiber at the focus of the telescope which collects the light from our source of interest, the star, galaxy, whatever, and then I input the light into a spectrograph which looks something like this. I'm not going to try and explain this because this would take some time. But essentially what this does is disperses the light or spreads out the light into different colours of different wavelengths of light. And this tells us things about the chemical composition of the object, the distance, the heat, et cetera, and the environment that that object is working in. However, building our billion dollar telescope and getting one object at a time is not really that efficient. So what we want to do is collect spectra from many objects at the one time and again they may be stars or galaxies or whatever. So this is what we call a multi-object spectrograph. And in a multi-object spectrograph I say, rather than just collecting a spectrum from one single astronomical source, we collect multiple spectra, sometimes hundreds, sometimes thousands, from many different source at the same time. And this is essentially what Manifest does. To give you some reference of the current technology that we use to do this, here's a picture of the AAT that's Chris mentioned before. This is a four-metre telescope up in Siding Spring near Coonabarraband, northern New South Wales. Behind this we have a fibre positioning robot where, and the job of this robot is to place hundreds of little fibres in the field of view, in the focal plane of the telescope, where we know there's going to be an object, again a galaxy or a star. So we need to very precisely position many, many fibres to collect the light from many, many objects at the same time. And this is what 2DF does. Here's the primary mirror of the telescope. We're looking down. The light's bouncing up here. And then we have some robot here, which does the configuration. Here we have a field plate we call this. It's in the middle of the telescope. And we've got a little robot. You can see these fibres. It's positioning 400 fibres individually, and it does this sequentially, so it's one after a time. And it's quite, you know, it's quite slow. It's as fast as it could be for its time, built about 15 years ago. But it takes some time, as you can see, to position all of these fibres very accurately in the focal plane of the telescope. So rather than doing things sequentially, what we've been working on is a concept of the AO to do this in parallel. So we want to move all of the fibres, position them accurately in the focal plane, again, of the telescope. We want to move them all at once. So this is what Starbucks is. And Starbucks is these little individual robots, and the concept is kind of similar. So here we have the focus of our telescope. Again, light's coming here from the primary mirror, the second mirror bouncing here. And this is where, this would be like a view of this focal plane of our telescope. I'm looking up into the sky. I can see, you know, some galaxy here, some stars here, whatever. And I want to take a spectra again of hundreds of these at once. So what I do is I place a Starbucks robot on the bottom side of this glass field plate. The light just is transmitted through it. I have a fibre stuck in the middle of the robot. And then I move all of the individual robots at the same time. And I can have a high density of objects. And I can do this quickly. And therefore I can do this efficiently. I can collect many hundreds of thousands of objects. So how does it work? Well, these Starbucks basically consist of two concentric, coaxial, piezoelectric tubes. And piezoelectric material is something you apply a voltage to it, and it changes its shape. So it can contract or expand or it can bend. So if I have two of such tubes, and then I apply some electrodes along the side of them, and then I apply the appropriate waveform, what I can get the tubes to do is to do this motion here, which is kind of like walking. The analogy is that this looks like a one-legged man in a Zimmer frame hobbling around the field plate. So what I then do to hold them onto the plate is I apply a vacuum in between the two tubes, and they're sort of sucked onto the underneath the glass. And then I apply these voltages, and the outer tube goes up and down, the inner tube bends, and this kind of just wobbles around the plate. And we can do this very accurately and very precisely. Here's an example of a bug under production here. We can see the two tubes are connected with all its services to a connector here. And here's a close-up of the foot underneath the bug. So what we put is an optical fiber in the centre of the bug. And then the inner tube here, the outer tube, and I placed these what are called metrology fibers here so I know exactly where I am. The precision I need to place this is something like 10 microns, which is a fraction of a width of a human head. We're looking through the demonstrator field plate. We can see these crosses are where they are and where they want to go. This is looking behind them, moving around. So they would be starting at some home position and then they're just walking across the plate like that. And we demonstrate this now for a unit of 10 bugs, but since that time we've progressed what we're doing. Because Manifest is some time away, because the GMT is still some time away, the end of the decade, we want to work out all this technology and we want to understand how it works and we want to prototype it, but we also want to do science with it. So what we're doing is we're building an instrument called Typan. We'll have 300 of those star bug robots on a field plate on the UK Schmidt telescope. Here it is here. This is a 1.2 metre telescope. It's quite an old telescope now. It's not that big, but it has a very wide field of view. So we can see a lot of the sky in one shot. So what we're doing here, right now, this picture was taken of last week actually, we're building this prototype system in the labs in Sydney and at the moment we're going to ship them up to the telescope next year and then start doing science. The aim here is to measure all of the stars and all the galaxies in the southern sky. So here we're now in the assembly and test phase. You can see a close-up of an underneath the field plate up here and you can see there's little metrology lights here again so we know exactly where we are and we're in the moment just testing the whole system. We did some simulations to work out. We've got 300 bugs. We want to move them sequentially so one of the big problems is like making sure they don't run into each other and damage each other. So we've run all this simulation software. You can see them moving around and this is what we anticipate using in the final instrument, kind of like flies on a pie. So then of course the ultimate aim is to build manifest for GMT. We're going to be a bigger version of Type-M. We're going to scale it up by a factor of three, up to a thousand star bugs. So here maybe is the field plate. Now no longer is it this big. It's 1.2 metres in diameter which is quite a challenging thing to make a glass field plate. We did a feasibility study last year, a couple of years ago where we sort of worked out what it would roughly look like. We have a bunch of optics inside all the star bugs to do various things and we're going to start the conceptual design study next. So what are we going to do with manifest? Well I can't really go into all the details here but various reasons what manifest allows us to do is to improve the efficiency of the telescope. So because we can move lots of these star bugs, there's lots of these fibres, all at the same time we can move them quickly, we can move them in parallel. Then we can take spectra of lots and lots of objects at the same time. We have hundreds of thousands of objects whether they be galaxy stars again. We can access the whole field of view because we can make the bugs walk around right to the edges of the field and all these things improve the science capability or provide a unique science capability to this telescope. Here's some of the science cases again. I'm not going to explain what this is but essentially what it's going to do is reconstruct the small scale structure of the intergalactic medium and we're going to look at the scales. We're going to look at chemical abundances in local group galaxies, look at what sort of chemicals are involved in the processes of those galaxies and we're going to do a survey of a large number of the chemical composition of stars in the Milky Way in order to look at the history of how these stars or how the galaxy evolved and formed. So I think that's it for me. It's a really exciting project and we're really looking forward to getting on sky with the next project. Thank you.