 Hello, everybody. I've never spoken to so many people, so it's absolutely terrifying to be up here. I'd like to thank Vicky for that nice introduction. I've never been called very esteemed before. So, yeah, I'm here. As Vicky mentioned, I'm here to talk about the Python role in unlocking the secrets of the universe with the James Webb's telescope. So I'm sure you're all aware that yesterday was the first release of the science images taken with JWST. I'm very glad we could organize President Biden to present the very first one on Monday just in time for EuroPython. So, yeah, the timing has worked out very well. When I was first asked to do this, I thought, oh, this is great because there's so much Python involved in what we have done with Webb. And I thought, oh, I have so much to talk about, but when I actually sat down to do it, there was just too much. Okay, so I'm only going to give you an idea basically with my experience, colored by my involvement with the mid-infrared instrument. But bear in mind that 90% or more probably of the analysis tools, the simulators and so on, were developed in Python, okay, and used Python. So the way I'm going to structure this talk a bit is to give people an introduction to the telescope and the mission and what's been going on for the last few months, and talk a little bit about the Python tools we use to calibrate the data and to do simulations, which we use for testing and so on, and commission. And then finally, I showed the images from yesterday for those who haven't seen them along with a few others that weren't released as part of the announcement but have since been released. Okay, so the James Webb Space Telescope, it was built on the success of the Hubble Space Telescope. Okay, it was designed and built to answer questions Hubble just couldn't do because Hubble looks at a specific part of the electromagnetic spectrum from kind of near UV to infrared, to near-infrared. But Webb is an infrared instrument or a telescope, a pure infrared telescope, so it looks at a different part of the electromagnetic spectrum. It was designed to see the very first stars in the universe. Okay, so it needs a big mirror to collect enough light to see those. And the big mirror also provides fine detail in the images so you can resolve a very fine structure in your images. It needs to be cold because it's an infrared instrument. Most are an infrared telescope. Most people will understand infrared as being heat radiation. So if you want to observe heat in the universe, you need to keep your telescope very, very cold because if you don't, if this thing was on the ground, a great analogy I once heard was it'd be like standing in the middle of a football stadium with the floodlights on at night trying to see the stars. You just wouldn't see anything. You'd be overwhelmed by the heat from the telescope. Okay, it's a big thing. All right, it has a footprint of a tennis court and this is a picture taken from ten years ago when the full-scale model was here in Dublin at the Museum of Modern Art in Kilimanum. And this is that picture and in front here you see a lot of the scientists from NASA and European Space Agency and Ireland who were involved way back then in, you know, 2011, I think this was, but you can really see the size of the telescope, okay, and it's not trivial to put something like that in space. And it was overall, you know, in the announcements and so on yesterday, you heard a lot about the US involvement, but really there was a big European involvement in it as well and Canadian. So Europe supplied two of the science instruments in Canada, one of the science instruments. Indeed, the platform, the telescope and so on, was built by NASA who also provided one of the instruments, but the mission, the telescope itself, was put into space by Europe. It was launched on an Ariane 5 rocket. As I mentioned, it's an infrared telescope, so it looks at sort of heat from things, okay, that's a very kind of general description, but it's looking at the part of the electromagnetic spectrum, which we feel is heat, but it's actually looking at colder stuff in the universe and there's good reason for this. So Webb's four primary science goals or themes are shown here. There's the birth of stars and protoplanetary systems, the assembly of galaxies, the first light and re-ionization history of the universe, and planets and the origins of life. And they are pretty big questions to answer for a mission, okay, but these can only be answered using an infrared telescope the size of Webb with its instrument suite, okay. So it was from the ground up designed to address big questions in each of these four fields. Now I'm not going to go into detail about the science, because this is a Python conference after all, but the four science themes are just sort of how the telescope was designed. It's also a general observatory. You can observe anything in the sky and with that mirror and with that sensitivity and detail it and provide and the suite of instruments it can revolutionize pretty much all areas of astronomy and astrophysics, which is why the scientific community and the astronomy community is so excited right now and we got the first taste of this yesterday and again I showed the images at the end. So it was launched in December 2021 on Christmas day from Kourou in French Guyana. It was launched about lunchtime European time and it was the worst Christmas day I've ever had. I couldn't enjoy anything because I knew this was coming and I was so nervous and yeah I had lots of drink after it I can tell you when I went up safe. But it was launched and it was extremely successful launch. It's one thing you maybe don't hear so much about that the European, I don't know why it's flickering, but the Ariane space company did such a great job on the launch that they didn't need as much fuel to correct the path of the rocket after the launch which meant that that fuel or the path of the telescope rather which meant that the fuel that was supposed to have been used or could have been used wasn't which means they're available to extend the life of the observatory and instead of a mission lifetime of 10 years because that extra fuel is there and it was the only consumable it's now looking like 20 years okay so that's all from the quality of the launch and I was inserted into orbit. This is the final real look we got of Webb as it was moving away from the upper stage of the rocket and this video was taken by an instrument that was developed here in Dublin by a company called Reiltra Space Systems Engineering and it's just showing Webb being pushed off to where it's going okay and what happens I won't let it run because it goes on for a few minutes but you start to see the solar sail deploy when Webb started to get power and that was it that was the first indication that everything had gone so well because it was on such a perfect path that it could start to deploy you know as I said it was a big thing it had to be folded up to fit into the nose cone and this was a very nerve-racking time this deployment of the telescope took two weeks and was very slow and methodical and systematic and there was nearly 300 single point failures in this process so any one of those goes wrong and that telescope is completely useless okay so it was a phenomenal piece of engineering but they got it up there and deployed the telescope and it went absolutely perfectly okay and this was the day in January when when the last deployment went or it was successful and what's on the left here is kind of a visualization tool that feeds telemetry from the telescope directly or that takes telemetry directly from the telescope to show what's happening and this is the moment when the final latch went in on the primary mirror and we had a fully deployed telescope and there was celebrations in the mission control room in the space telescope science institute in Baltimore where Webb was controlled from so it was a great day so what's been happening since then this was back in January well the telescope has undergone a six-month commissioning phase where everything gets tested out to make sure the mirrors work okay make sure the instruments work okay and it's a lot it takes a long time and because you have to do this right the first was the mirrors okay this is a this is an amazing image on the left is an actual selfie taken by the telescope with one of the instruments so each you'll see that to fold the mirror up there's there's mirror segments okay that have to come out and fold together so each one of those mirror segments is a single telescope in itself okay and what you need to do is you need to bring each one into focus first and then bring them into focus together to get the best out of the telescope and what's on the left is the selfie and what's on the right is a star in the middle of each mirror and you can see that they're distorted you know the mirrors were just that completely out of alignment and putting everything back into alignment and getting everything phase in the center and focus is a very very methodical process and it and it takes about two or three months two months but they got there and this was the first real image we got from web where now you're seeing and this is the same selfie just with all the mirrors focused that everything is pointing in the right direction they're all lit up because they're all pointing into the into the telescope people and you get this image of a very boring star okay but it's just showing the sharpness you can achieve that spikes are just the fraction spikes because of the optics of the telescope but what jumps out of this image I don't know if it is clear on the on the screen with the lights but there's just that background filled with galaxies okay it's so sensitive that no matter where you point the telescope you will get just field of galaxies okay and actually for commission and for science it's great but for commission and it was just so annoying right they're photobombing everything once the first instrument was focused then they could move on to the rest so there's four science instruments on mary and a fine guidance sensor which which kind of you know keeps the telescope pointing in the right direction as it's observing and this was the first image released for all of the instruments including mary up here on the top right which is which is what the instrument I worked on and again I mean this this is a field in the large magillanic cloud which is nearby galaxy so it's another galaxy but the images are filled with stars and galaxies the thing is so sensitive it collects so much light and it has such fine resolution that it doesn't you don't know where you point it if you integrate for any little if you expose the instruments for any short period of time you're going to get this rich field of stars and galaxies it's amazing and so this is kind of a close-up now of what the previous best space-based infrared instrument could do compared to James Webb in the mid-infrared wavelengths where mary observes so on the left is a spitzer space telescope and on the right is the James Webb space telescope and you can see the improvement is unbelievable and with mary you start to see lots of interstellar medium features and so it's very well resolved and these were taken in a few minutes or you know a few tens of minutes so no no exposure time basically and now that's kind of how commissioning was going and then we you know could there are instruments were commissioned and so on and now I kind of move and tell you a bit about well how do we get these images because what's shown here on the right that's not how the images come down from the telescope okay you don't just get these beautiful things they they look a bit more messed up and this is a very simple example so on the left here um yeah I don't have a laser thing but on the left is a raw image okay and you can see there's a lot of crap in there right this is a simulation of a galaxy core but if you look around you see all of these stripes and bits and pieces and these white things are cosmic rays hitting the detector okay there's lots of you can see the sort of tree ring structure up here so there's lots of detector effects all of that is due to the effects introduced by the detector by the by whichever instrument you're using what you need to do to produce the images that are released that are used for science is get rid of all of that okay to produce something like what's on the right now there's still some bad pixel areas but that's okay because you can set up your observation to move the detector slightly and fill in those but you want to remove all of the effects introduced by the detectors and by the telescope the telescope will introduce them as well and leave just the pure image of the sky that's well calibrated so you know the flux of what the object you're looking at you know the wavelengths and the position on the sky and so on and that's where the instrument teams come in like myself are you know like the team I worked on where before launch we did years of testing of other of our detectors and so on to you know understand all of these effects that were introduced how we could correct them in orbit understand the behavior does do these effects change in time create calibration files which we could use in orbit to correct the data and you know refine those in orbit and this basically what I'm describing is a calibration pipeline okay what you understand various different effects and you create a pipeline where you send in raw data you correct one effect you correct the next effect and so on until you remove everything and it's what we call a calibration pipeline and this is kind of a schematic of the stage one pipeline and on the right is just a series of steps to introduce our understanding of the detector on the sorry on in the green is just the list of steps what the gray is shown is sort of the reference files so these are the calibration files that we've created based on our understanding of the telescope which are kind of can easily be changed but the calibration step the operation that that does is fixed we just change the values based on our understanding these calibration files are created by us and we're finding commission and the blue is just a description but I just giving you an idea of what I'm talking about this is how we get from raw data to to the to the public release data so we've been over in in Baltimore at the the control room where I showed you and I'm working in the space telescope science institute for the last few months working on commission in myri where we were looking at all of these effects we're finding everything creating new calibration files making sure everything worked okay and the instrument worked as expected um and this is the first main area where python comes in now I'm not going to show any of it because I think it's still proprietary but all of these call all of these checkouts we were doing all of this commission and tools we were testing and running and so on we're all developed and built in python okay they live on a get on a proprietary github repo um which the which our instrument team you know develops on and it was used to basically commission myri okay and other instruments were exactly the same so when we finish this process our instrument is science ready and images such as the one you saw yesterday can be taken okay we tick a box it's the the instrument is science ready and those observations can be taken other instruments have to do the same and the final check out was a few days ago where the very last science mode for James Webb was commissioned so it's fully ready for science now um but what you will see here is that there's actually a bunch of science modes okay so I've kind of talked about the various steps in calibration but really in practice you don't want every astronomer doing run these steps themselves you need some sort of software package that will take the data from the telescope automatically run a 3d steps and make that process data available to observers and scientists around the world so they can do the science okay our job is to make is to remove most of this from there stop them worrying about it okay it's true that a lot of scientists will rerun everything maybe tailor some of the steps maybe tailor some of the values and that's great if you if we need to do that we need to have a flexible software package that can do that which we do but there's basically 17 science modes for a range of different uh kind of purposes across the four instruments it's a big pain in the neck okay but a software package was created that basically takes the data figures out from you know metadata in the in the files what's been observed what mode is being used sends the data into the specific pipeline paths and processes everything and sends it straight to the archive where astronomers and scientists can download it and this is called the JWST calibration pipeline okay it's a pipeline software suite it does have some c plugins because speed is required for a very small part of the of the pipeline python and slowing things up quite a bit but it's but 90 like you know more than 90 percent is written in python it automatically processes data that's sent from the telescope and as I said you know it it decides which data it is sends it through the correct path and you know calibrates it combines it if necessary and produces the science ready products that astronomers will use to win Nobel prizes hopefully and why is it in python well there's a there's a substantial I don't know if there's any astronomers in the room but there's a substantial knowledge of python in the astronomical community we've had tools like astropy and of course numpy and sci-pi for years now and you know it's it's perfectly suited to to astronomical data analysis because in the end are just images okay and spectra which are two and one you know they're just numpy arrays basically so and the benefit of having everything together like this as well in a single calibration pipeline is that it it's easier to maintain different teams can work a lot of the instruments have the same calibration steps so the teams can you know interface over the best way to do things that will suit all of the instruments so you've an additional expertise in there and you can also you know it's developed in the open for years the jwsc calibration pipeline is being openly and publicly available so the community could get used to it start using it and you know play around with it a bit it's very flexible it's because it's written in python and i go through it in a minute and the the steps are python classes and so on so it's trivial to move stuff around to better improve your calibration for things like Hubble people don't realize that Hubble had a lot of science modes as well but pretty much everyone had their own independent calibration pipeline which were these monolithic chunks of code that were written in different languages and it wasn't straightforward if you wanted to change something you just couldn't do it okay because it just you sent in something at the top and your product came out at the bottom whereas here you're going through it step by step you can save at any point um you know rerun from or you know pause the pipeline rerun whatever you need to do okay so before i get into how it works the just just some information for yourselves the observation astronomical data files are in the so-called fit standard okay and it's basically what you have is you have a header with some metadata that gives you things like what instruments were used um you know the dates just just lots and lots of data about the observation uh metadata about the observation and then you have a data extension which is your actual data like the image shown here and you can have additional things like if you have an error image you can have that in your fits extension okay so any any software to process these things has to be able to read these in and output them and run them through a pipeline we do that uh for JWST for the calibration pipeline this is done should i say um using software data models so these files are read in to a python object where members of the python object correspond to different aspects of the input fits file so if i read in the file as before just using this very trivial um so it's the JWST package so if i import the data models uh from the package i can just open my fits file and my image my data is just a member of that object so i can access a true model that data or if i want to check out any of the metadata i can just go through the meta um member and find whatever i'm looking for and it's very flexible and very very useful um and it's you know it's uh because it's python you can you can import uh you're sorry the defined by yaml schema so you can just start with a base and add as you need to go and create data models for every kind of file that JWST pipeline uses such as the raw data such as the output data such as the calibration data and actually it's extremely useful for the for the for the instrument teams because we can use the data models to produce the exact format required by the pipeline to calibrate the data um all of this is done through a package called stpipe so basically as i said we'll go through series of steps series of steps at one level is called a pipeline each step is a python class that's based on the step um the step class and the idea behind this is that it handles everything and leaves the instrument teams and the community um free to work on the scientific stuff so this handles everything from data input to um person configuration settings um input and output file management accessing the calibration files okay that's a big thing those calibration files live on a server and this package will automatically recognize from the metadata in the data model what is needed fetch that from a calibration file server takes all of that worry away from astronomers and the instrument teams so we don't need to worry about it we can just focus on what we want the corrections to be and the data to to do okay it handles logging and um it has a consistent interface for for users across all of the pipelines across all of the instrument models okay so this is this is what the JWST calibration pipeline is based on and i'm just going to give a quick example i've never given a python talk before so i don't know if you're supposed to show code or not so maybe this is a big faux pas in the python world but i don't know if you can see it but basically this this is a very simple calibration step and it it uses the step class as an inheritance step class okay it's called a fringe step all it does is divide a science image by your reference array um it's set up with it so it just takes an input opens it with the data models gets the calibration file using this one line uh will fetch the calibration file from the server or if it's on your local computer it will just get it from there it logs everything if there is no calibration file it will give you a warning and move out you know stop the step and and set it in the pipeline that it was skipped it won't crash the pipeline the pipeline will keep going but you will get this warning that this step has been skipped the fringe the calibration file is then opened as a data model a fringe model in this case and the correction is done in a separate function that's imported and as i said it's just a division of of two arrays and returns the output model so this is a very simple version but it gives an idea that if you know whatever the develop like whatever the instrument teams or the scientists want to do is done in a function that's imported all of the other things so just the input and the output the fetching of the calibration files and you know reading with configuration personal configuration parameters for this step but but all of that is handled by this st pipe package and and the step class which makes it very very straightforward and each one of the sort of steps the calibration pipeline is step it's just a series of steps so the pipeline class is just a list of step classes that are run through and again very straightforward if we want to reorder these steps we can just move them around if we want to remove one we can just skip it or remove it i'm just giving an example here again you're taking your your input reading it as a data model and this this is where now you're you're starting to check for the different modes so you read the metadata from the data model you see that it's a mirror observation and you go through the mirror part of the pipeline okay and there's logic like that built in all over these pipelines it's extremely useful extremely powerful extremely modular and because it's compared to some of the stuff you guys do i'm sure it's extremely simple okay it's just a series of steps um so in the end what you want is you want to take all remove all of those detector things which i showed which is done in the first stage of the pipeline these second stages just do flux calibration wavelength calibration and astrometric calibration and so on okay which are just you know registering the images on the sky and making sure you have the correct flux and and then the third is stitching everything together so you know in an observation you will take a bunch of exposures um calibrate them up to level two stage two and then stitch them all together to produce a final mosaic such as the images you saw yesterday at level three and all of this is handled by the calibration pipeline wow ten minutes i'm gonna skip the simulators uh that's okay i'm sure everybody wants to see the images and not just me rabbitting on um okay uh so there are i mean i'm not very quickly go through the simulators but that that was basically a for the pipeline but you know to test the pipeline to prepare for commissioning to to test our commissioning tools we needed kind of real data not real data but simulated data at the sky because all of our ground test data is on calibration sources that's not representative of what you see in space okay it's good for you know checking your uh you know understanding your instrument and so on but if our commissioning observations are going to be of something in space then we need to create simulated data of something in space and that again is all python all python packages in miri team we we developed this miri sim miri simulator where we could know what we're going to observe during commissioning simulated with miri test our tools and in the end it was it was used pretty much by every part of the miri team um and what we had in the end sorry i'll just what we had in the end was something like this where we had knowledge of what we were going to observe in the sky which is on the left knowledge our simulations on the right which incorporates our understanding of the instrumented sensitivities and so on and this is what it actually looked like in commissioning so on the left is a simulation and on the right is the actual observation during commission and there's a slight offset in angle but that's just because we were out by a degree um on our way we were out we simulated a day uh two days later then was actually observed but you're looking at this the field on the right and it's almost perfect okay so our understanding of the instrument was extremely good because we could take that simulate what we thought this guy should look like and it was basically the same okay so really uh really allowed us to develop our it was perfect for developing our commissioning tools post pipeline so after the pipeline is finished and the science products come out um again there's a suite of um python tools to analyze the data and the images and the spectra and whatever uh these are all like a lot of these would be quite common throughout the astronomical community um okay the images now everybody wakes back up um so i'm on the knife we had uh president biden reveal the very first image uh science image with taken with awst and it was i don't know anyone watches a few okay that's good because it was it was really weird but um but it gave a whole nother level of publicity to the uh to the community not to the instrument and someone obviously not in this room because there's about five people who put their hands up but but no it was really something else and we didn't know that president biden was going to release the first image when he found out on sunday but this is the first image just uh it's the um jwst first deep field okay it's the furthest into the universe apart from the cosmic microwave background which is the first light that's absolute first light in the universe but this is the first sort of stars and galaxies that the deepest images stars and galaxies we've ever had this was done with 12 in 12 hours with james Webb okay 12 hours that's nothing the Hubble deep field ultra deep field took weeks of constant observing okay okay spread out over years but it's a weeks uh exposure and this is 12 hours this was done just in just over one shift one console put it here that way um the detail in there is phenomenal if you download the full resolution image you could scroll around it all day you're seeing a massive galaxy cluster in the center and you know Einstein's general theory of activity told us that if there's a mass in space it deforms space time so if you have a very massive object in the universe it acts like a lens because the life from behind it sees the deformity and gets bent around it and that's what you're seeing here so these are lens galaxies and this isn't even lens galaxies this is the same galaxy just seeing a different part of the lens here okay so phenomenal stuff the galaxy cluster is about four or five billion light years from here you're seeing stuff in this image that's it was emitted 13 billion light years ago over 13 billion light years ago so the universe was still only a few hundred million years old again this is just in 12 hours with james Webb you could scroll around and look at all sorts of funny images or funny objects like this thing here again this is just deformed by the by the gravitational lens you're seeing objects in here like these red galaxies which aren't visible in Hubble okay because they're so far away the light from those has been redshifted redshifted means that the light has just lost energy as it moves across the expanding universe and Hubble physically can't look at these galaxies because it can't see in these wavelengths this is why james Webb was built to see those galaxies the image that was released was taken with near cam but on the left is the mirror image so this is the with the instrument we worked on and this sees even further back now we actually don't understand we don't know a lot of the objects are in the Miriam is because we need to identify them which is probably why it wasn't released yesterday in the in the announcement but Miriam has the potential to look even further back again okay because it's longer wavelength stuff that's further away gets redshifted to the longer wavelengths and we also saw Stefan's quintile yesterday which is a small galaxy cluster interacting galaxies nearby relatively nearby the one on the the one here is actually a foreground galaxy okay this isn't associated with these three with these ones here these are interacting so they've been in a kind of a cosmic you know dance for millions billions of years and this is the image taken with near cam so you can see the white is basically the stars but if you look with miri you start to see the interstellar medium and you can see all this dust being spun around the galaxies and in that dust there's lots of stars being formed because they're you know there's turbulent motion in there and there's compression of the dust and gas and stars will form in there and this is a near cam miri composite but the Miriam image itself is insane okay it looks like fireworks and again this this is so this is kind of a zoom in on the top tree here but this galaxy on the left again is a foreground galaxy it's pretty standard looking for for a spiral but on the right here you can see really how the interstellar medium is being disturbed by the interactions of those galaxies and you can see that there's these shock fronts and clumps and everything and here it's like there's you know it almost looks like the gas swirling or the dust is swirling into the center and this is a super massive black hole in here this bright object which is only visible in Miri or which becomes very apparent in Miri okay and again photo bombed in the background by all of these galaxies that are even further away okay which were you know consistent pain in the neck the next image was the southern ring planetary nebula this is a dying star this is a star like the sun which is in its last stage of revolution and it's throwing off the outer layers of the star exposing the hot core in the center the hot core is on the left is a near cam image and you can see the hot core star which is white dwarf basically or a proto white dwarf and the layers of the star being thrown out into the interstellar medium and the right is Miri but the detail in the near cam image is just phenomenal okay you can see all of this layers and you know there's dust and so on in here or there's just gas and dust and just layers and layers of mass loss from the from the star as it dies basically it was always known that this was a binary star but actually in Miri you can see the binary okay the resolution of JWST is so good you can see the binary in there next to it it's a dusty star so it's not visible in near cam but Miri resolves it it's phenomenal stuff this was released yesterday as well it's the spectrum of an exoplanet atmosphere okay so this is a hot Jupiter basically a big gas giant moved in front of its whole star while the web was observing and due because of the changes in the stellar spectrum we could infer the content of the stellar atmosphere and what was found was that it's filled with water vapor okay so steam basically this is right next to its star so it's very hot so it's filled with steam now this was kind of known what wasn't known or what what's different is that these water features shown here these bumps are smaller than expected the reason is that there's clouds in the atmosphere okay so this is actually inferring the presence of clouds in the atmosphere of this hot jupiter because that the the clouds will inhibit the formation of water vapor okay so just from this one observation uh and this is you know it will observe things like you know candidates that have potential for life and and so on but this is a very kind of standard hot gas giant and it's already you know producing these amazing results this wasn't released yesterday this is in our commissioning paper that was released today on the archive this is jupiter jupiter shines out in the infrared okay and i'm just showing you this it you know it doesn't look anywhere near as impressive as the Hubble image does it right it looks like someone shot a torch behind it but telescope or the telescope James Webb is a big unit okay it's a big thing but to observe something like jupiter it needs to be able to track it on the sky so it doesn't move in target analysis so it can actually follow jupiter as it moves along and this this observation when you're kind of demonstrated that and you can see the moons which are also bright the red spot so you're actually seeing a mission from jupiter itself now not just the reflected light from the sun um and you can see its rings okay you can see the faint rings here uh you can see lots of detector stuff in the background that's just because jupiter is so bright okay it's not um there's nothing you can do about it when it's so bright it's saturating the detector and someone this one was my favorite this is the kareena nebula the gas and dust in the kareena nebula um so there's massive stars up here that are irradiating this gas and dust and you can see the the gas and dust is evaporating off into the into the interstellar medium there's star formation going along in the dense parts of this um which you know can't be seen by Hubble because Hubble can't see into the gas and dust and near cam can't really either but mary can so if you look with mary you're seeing right into where these young stars are forming you can see the you know the plan is to to look at the formation of young stars and protoplanetary disks so look at the very earliest stages of planet formation and there's lots of other things here as well the ever present photobombing galaxies in the background uh outflows from massive stars okay you can see here's a there's a jet that's carving a cavity in the in the interstellar medium here just phenomenal stuff and all of all of these observations were done in one week okay one week and not even filled the week um this kind of thing is routine for the James Webb Space Telescope which is why the next few years are going to be extremely exciting um these these targets were also chosen because they're not on the list of targets that are going to be observed in the next year so the interest in targets have yet to be observed okay i'm going to leave it with this image because it's my favorite it took 20 years to build the telescope put it in space and now um yeah now it's very exciting time and all because of python thank you perfect thank you patrick that was amazing it's really hard to interrupt you when you're showing this is wow okay doesn't fit in my brain oh so we have time for a few questions if you want to ask a question you have to stand in that microphone that is there um anyone want to ask a question have to be quick um i don't think i i want to say a reminder if you are watching this conference remote you can ask questions right so if anyone is remote there is always a operator asking do you want to ask a question and you can get into a call and then we'll put you here in the screen live with everyone um so remember that for all the talks so first one uh so what are the interesting targets uh what there's a list of them uh so i mean okay i everybody's favorite thing that it's going to do is is the exoplanet atmospheres okay because potentially you could observe life and and we'll see so um web is not a planet finder uh you need to know what you're you need to know the planets before it will look at them okay and there's we have how many now three four thousand more probably exoplanets that are known some in you know some are rocky some are not of atmospheres all of this so they pick the best candidates of them they will be observed in the next year so really in the next year there's you know somebody only asked me yesterday in an interview you know will you find life well there's never been a better chance um potentially very quickly but yeah i don't want to get everybody's hopes up on that but yeah things like that i mean there's so many interesting targets so many cool next one hi first of all thank you so much for the for the presentation uh web is such an amazing uh a piece of piece of hardware and software um my question you mentioned something about the data pipelines being open and to the public doesn't also involve the data how do we get data okay so so everything that was released yesterday is publicly available okay that was these are so-called early release observations um the raw data from them and the calibrated data the fits files that will um will be available either today or tomorrow and they're free for everybody there's a second program called early release science which will also become immediately available to everybody um when they're taken okay so some have already been taken so that data will become available today or i can't remember today it's there tomorrow as well um but as they as they are taken in the first cycle they will become available the other programs the g so the so-called guarantee time and guest observer programs there's a proprietary period of one year for them so scientists have the right because they won that time to have the right to have it to themselves for a year to produce the science you know to produce papers and make the discoveries and someone after that it becomes publicly available right so basically in a year we can do just pip install jwst and we can well right well right now you can do that all right you can just type in anyone on your computer open type pip install jwst and they should install the software package um this the data that you want right so go get that data download that from an archive it's on mass it's called mass archive you can go and fetch it today or tomorrow and you can start playing around right cool so no that's two questions one more we have remote questions so only one more sorry for the other ones thanks for the great talk i was just wondering what is the size of the scale of the raw data uh to produce one image the scale it depends on so it all depends on how long you integrate so how long the exposure time is um for let me see now so basically the way the telescope works is it samples the detector uh at some readout time okay so it takes a sample takes a sample takes a sample and then you use that to calibrate your data and it depends on how many of those samples you take and if it's short it can it's only a few hundred megabytes and now you will have more than one you'll have more than one exposure right so you're talking a few gigabytes a couple of gigabytes but if it's much bigger then of course it just scales with the the amount of samples of the detector basically so it can yeah it can be tens of gigabytes probably for the biggest all right thank you okay thank you very much Patrick please give him a shout