 So we put two icebreakers in here, what's the most surprising thing you've learned so far in this course, and most surprising thing you've learned so far in your career. And if anyone has any other icebreaker ideas, please. Well, you can also propose your own questions there. I'm trying to think of something specifically related to today's materials, which is a bit. We have a wide range of topics. Yeah, so what's our summary so far? Where have we come from and where did we go? We started with day one, and it seemed like it was pretty normal Python stuff, like Jupyter numpy pandas, I mean, doing Python code. Day two, did we start getting exciting? So day two started pretty normal with pandas and matplotlib. So again, using normal Python code. Towards the end, we started seeing more interactions with the rest of the world. So the data formats, emphasizing how it's not just about the code, but how you mess with this other, how you read and write data outside of Python can matter. And then of course the things like the lintrace and so on, which is user interface. But then yesterday, I mean, yesterday's the kind of stuff which is really important for using Python, but you wouldn't see in a typical course. So things like making the scripts, well, you might see that in a Python course, not a purely theoretical programming course, but Yeah, can introduction to programming with Python might not even use Jupyter, it might use just scripts. And the library ecosystem and dependency management. Well, that is, I mean, that that's really getting to the kind of stuff which we use, and we support people with on a daily basis. So my estimate has been that about half of our total support request are helping to install software on the cluster. A lot of which is in Python and could, in theory, be done independently. But it's so confusing and hard dependencies are a complicated topic and big unsolved problem. It's also the reason that all of these libraries that fact that they exist is the reason why Python is good. I mean, the language itself is good, but the main reason people actually use it is because there are already all of these libraries available. Hello from my side as well. Yeah, like I was just thinking myself at that, like, quite often when when using Python, it's more about reading what some other person has done, like a package and then reading how they want you to use it and then just using it. And it's like, is this, can you call this even like coding when you're like calling functions from other libraries and that sort of stuff. But of course, you are coding but but like coding is coding. Yeah, and and and using using like it would be wasteful to not use the capabilities provided by the other people. So it's like, and it's usually the thing you find is not exactly the thing you want, but you can coax your data to fit into that and kind of use it or modify it slightly to make it what you want. Yeah, and it's like buying like Lego blocks, like you buy Lego blocks and then like you have a world that you can create with them. And you can of course, like just follow the instructions and do the set that you have bought or whatever, or you can go wild and do whatever you want with them like combine the blocks in an unexpected way and create something new. And that's what like usually happens, but you don't start molding the Lego blocks yourself like that would be waste of time like getting like a 3D printer to like mold. Of course, if you want to, you can go like you can go to Python libraries and and like start writing C code and that sort of stuff. But then you're coding C code and you're not coding Python, right? So I think we need to do a volume check. Um, so, well, let's start with an order me Cmo. One, one, one, two, two, three, three. So I think Cmo is too loud. I can also drop it here. I'm actually not at 100%. Is it better now? Oh, you still seem pretty loud. Okay. Better. One. Oh, yeah, one, one, two, two. Okay, is this better? Okay, might be just my enthusiasm, rubbing off. So I'm liking these icebreaker answers here. I just thought of these last minute, but it's good to see what's surprising in the course. And also, I like these career lessons also. So last time there was some problems with a blurry screen. And one part might have been. Which, but also it might have been so people are sharing by zoom and then I render it on the screen. And then capture it. And then it gets rebroadcasted. So there might be a little bit of blur that comes here. So if you see a problem, let me know is it really bad or like a little bit of anti aliasing around the letters. And let us know right away and we'll try to fix that. Yeah. On your side, I recommend in the Twitch, there's like this cork icon at the bottom right of the video where you can click the video quality and you can make it to appear a source, source quality, because otherwise it's automatic. And then it might be graded based on the network performance that you have. So you might like get suddenly a blurry, blurry flash because it's buffering or something like that. So, so setting it to force it to be the highest quality might mitigate some of the problems. Okay, well it's 1001. Any comments from these surprising things or should we just go on. Should us instructors say the most surprising stuff. Maybe we should just go on, we can answer this in the panel discussion. Okay, in the wheel getting reinvented. But yeah, we'll talk about, well, at least we will talk about how to package your own stuff so that other people don't need to necessarily read themselves later. But first we have some parallel processing. Okay, so I'll head out. Yeah, I can take the share if you capture from my screen. Okay, see it. Okay, I will head off.