 We're excited to welcome Dr. Ajit Kumar for today's first keynote. A bit about Ajit Kumar. He has done his MSC and PhD in nuclear physics from University of Calicut. And yeah, he has been senior scientist with inter-university accelerator center in New Delhi. So over to you. Thank you. So, morning everybody. And I'm very happy to be here today talking to you all, Python, whatever I know. So first of all, let me just share the slides. Is it okay now? Yes, yes. Okay. So what we will talk about is the role of Python. We already had a discussion yesterday about how it should be used in our education system and all. So I'll talk about those things in a different context because that will be mostly a couple with some project which I have been doing. So I'll just talk about a little bit about what I have been doing and that's like this is a problem when you call old people. So they will tell old stories. And then we will just try to see whether our present situation is it really very great or is there a scope for improvement and about our open source contributions and all. And then we'll come to the IT training at our school level, the present status and what are the desirable options and all. And here we will spend some time and then we'll also discuss the topic which we had a panel discussion yesterday, computer programming for everybody. And okay, of course this was the link to some keynote which I gave six years back. So there will be some reputations actually. And let me start with something about myself. I've been working at the University Accelerator Center in New Delhi from 85 to till this year and now I'm free from there so I have more free time. And here our main job was actually developing and maintaining particle accelerators. And here the computers come into the picture in two ways. One is designing all the structures you use computers. And then in the control systems and the data aggregation systems we use computers. And mostly coupled with the different hardware systems. That aggregation means you have to couple it with the real world devices. So and also we had a lot of interaction with universities because it was an inter-university center. And we were providing particle accelerator facilities to the university and the academic community. So what we noticed you as the entire system was actually suffering from lack of good equipment. We had if you see many of you know the college laboratories. The equipment currently we have is a little bit outdated. And the new equipment it's all available in the market but they are expensive and proprietary solutions. There are companies like Pasco, Feeway and all selling all kinds of equipment but it's not really affordable. Here I have seen only the ICERs buying some of them and not very many colleges or universities. So using our expertise in developing control system and other things we thought why not get into the area of developing some cost effective lab equipment. And the concept used was very simple. Suppose even if you take a thermometer it is a certain equipment. So if you see it has three components. One is a tip of it is the sensor and then you have the display and user interface. User interface means there is a small button to select say Fahrenheit or Celsius display. And then in between you have signal processing. That means some electronics will be there. So as a cost cutting measure what we thought is let us don't do this part. Like let a computer take care of that display and user interface. And the signal processing along with some interfacing circuit. So we make only this middle portion. And for sensors whatever available in the market like consumer electronics a lot of sensors available. Try to use them to design science experiments that can be used at mainly we targeted BSE level. And later on it has spread to higher secondary and other levels. And it was decided all these things should be open source. So that there are different reasons for that. One reason is anybody could just take it manufacture it and use it. So we have I mean we didn't want to have any control on that. So it's an hindered you know development. And then these things can be modified and the other somebody else can come and improve that. And most importantly is today we stopped. That's what we thought at that time. So for the Institute IAC stops this development. So it could continue if someone else is interested. So that's how we started a project because named us Phoenix. And it stands for physics with homemade equipment and innovative experiments. So the first version you can see very old kind of thing. There's a box sitting here. And all this history you'll find in the IAC website. And this was using panel board. And the interesting part was it was all based on some kernel code. There was a device driver. So all these real time measurements like suppose yesterday you saw in that demo like if you have a pendulum or something when you measure accurate timing the kernel itself freezes it make the measurement. But later on then we shifted to other designs like serial port version with a microcontroller so that at that point the real time work was shifted to a microcontroller. Then later on versions came with when the USB came we had to catch up with that. And finally this is the current version. I mean it came in 2017. There it did a lot of work in this. And it's far better than the earlier one. So here just if you ask what is this box size. It is for students. It's an affordable tool for doing experiments. And the key is it's like anytime anywhere freedom from lab timing. This was something initially we thought of. But today it has acquired a special significance because almost all our labs have stopped working because students are attending online classes so theory is going on well. But there is no lab sessions. And here actually this is a very useful tool because it's a small handy thing. It can be sent to the students and they can do experiments from home. And some software has been developed so that the teacher can monitor it over the web. So that yesterday it was presented. And again for teachers it could be used for classroom demonstrations or intentional once teachers should be able to develop new experiments. So that is where the Python also came into things. And for engineers or engineering students it's an open system that combines actually basic physics electronics and microcontroller programming and some computer interfacing. Also this if you go through this design of this system that sort of code everything is available. One can get some idea how these sort of things are developed. And for hobbies of course it is like you kill more time with less money. It's not very expensive toy. And an important factor was the collective effort. It was not the team at IUAC doing everything. We had only two or three people doing something. But when it was announced on the web and we could attract more number of people and the significant I mean there are many many but I only just mentioned three, four. And first was actually for both he only came up with the idea of Python. So I was doing everything in C and then he just showed that it is better to use Python. Then we had a devian developer who is a school teacher in France. He came into the picture and he started packaging the software for devian. He also did a lot of work in this localization like translating into different languages and all. And Praveen was very active in this. And in 2014 he has developed some experiments under a Google Summer of Code project. And later on in 2017 actually around 14 onwards he started working on it. And after two three years of continuous and full-time work he released a new version. And the older that 2006 version actually it was used in his school of jobs. And so this collective effort actually this idea of collective effort came from the free software movement. Because I was having some familiarity with free software starting from 1994 using Linux. And this is something which I feel that everyone in the computer field especially students should have a look at this. Because there are so many people like Bill Gates everyone knows. And the charge shall man smaller number of people know. But the importance of this movement is earlier also even before say in 1983. There were some software which is distributed freely but they were called freeware. And it was available on the internet. But what was more important was putting it on a right footing with a proper license. Growing the new public license. And it became a very well organized thing. And it was a very daring step on Stalin's side to announce that we will try to completely free operating system. But as you know around 1991 they completed many things except the kernel. And but some other very interesting development completed it. And that should be an inspiration to the students actually. And if you can look at the search internet you will find around say this 25th August 1991 there was that historic posting in a news group that was done by Linus Talwar. He was a computer science student. So this tells you even a computer science student you know how much one can do if properly you know that motivation is there and the capability is there. So he announced he is doing a free operating system say just as a hobby one be big and professional like GNU or the very modest announcement. And he released he could release the version 0.99 in 1992 itself. And in March 1994 the kernel version one and the existing GNU system we have a complete OS. That was the preferred to call it as GNU and Linux. And I was fortunate to you know come across this somewhere in February. That means just before the version one was released. So I was using 0.99 or something it used to come in some three copies in those days. But now actually the Linux has come a long way and it has become a very major force in this field. And now I just mentioned how I mean just how I came across Python and what was the reason. See when I wrote this first version the software the only language of mostly I was using C because earlier it was Fortran then just to see and because C is good for communicating the hardware. But when I started writing an oscilloscope and just to give some things quite rudimentary looking like this it was running into hundreds of lines of code with the existing tools available in those days under Linux. So that time only promote role this is another option. So I switched over to Python and the result was really fantastic. Because when you could make very complex system like this with a very little effort because all these things each experiment now is only 100 lines or 200 lines at the maximum with all these features. Because Python libraries like PyTorrent all gives you all the you know required support. So just one example experiment we have put here. And what you see here is this thing measures the frequency phase difference and all but actually there is no frequency meter or phase meter. All done in using this data analysis of sci-fi because you have two sine waves. And if you capture two sine waves into software you can fit it with a mathematical equation find out the frequency and you can also find out the phase difference. So a lot of things are just done in software because of the Python libraries mainly NumPy and sci-fi. And all these developments I am not going into the details. Everything you will find in this website xyz.in. So here what are the experiments about all those things if anyone interested they can see there. I mean even the hardware availability or design and software everything is there. And we were also conducting teacher training. It was mainly for basic teachers from colleges and universities. So 2005 onwards we might have trained nearly 1000 teachers in our one-week training program. And then also conducted more than 100 one-day workshops at different places to promote this device as well as Python and free software of course. And these details you will find in the IUNC website. So the topics what we taught during this program were basically some scientific computation and after teaching a little bit of Python and some experiments and all based on open hardware and software. So this was the scheme that we were running. And incidentally I also wrote a book that was meant for BSE mathematics students. It is downloadable you can either get it from xyz.in or in this site size school.in that will come to that later on. But it is downloadable. It is mainly focusing on applications in like science and mathematics. And Python of course we started using Python just for this outreach project. But later on we found that our particle accelerator control because let me just say something about this accelerator control. This is something called a beam line. You can see here the high voltage structures are there. But all of them are interfaced with the computer. Like here what you see this blue thing is a huge electromagnet and the current through that everything is set from a computer. So what we do is we have some special interfacing hardware that is connected to a computer. And this whole thing the network runs some server like that as well written in C. But the front end we could write it in Python to make the job very easy. Like here three lines of code and this is just reading one parameter from the system among these 2000 signals. So any parameter in this system can be accessed with one line of code. So this helped our other experts I mean the accelerator field who doesn't know programming. So they could write small scripts and access it. So Python was very useful that way. And now so far I have been talking about something which I was personally involved. And now the question is we are doing all these things. Our intention is situation should improve somewhere or other. So that way it is relevant to ask a question because sometimes people already say we are already a superpower in software and all. So it sometimes it is better to have a sort of assessment ourselves on using maybe simple parameters. I may be wrong in this so but it's just introducing that like just putting forward the idea we should do that. Suppose if we are really strong in software probably we should have had many of these packages I mean Indian products. Because when I look for a CAD package electronics mechanical or anything I don't find there all. Companies are from abroad and other applications like weather forecast or even language compilers or operating systems. We don't really see local products right. And this is one area where I was I mean having some experience is simulations that is mainly for its nuclear physics computations and modeling. Nuclear modeling and another one was basically designing of accelerators. In both cases actually we did not have much indigenous software. So so many people working in software and we are earning a lot of money in software then why these things are not happening. So that is something one should think about. I don't know I'm not that an expert in those areas but that question should be asked. And also incidentally in an interview with the Lennestro words so interviewer asked him about the contribution of kernel code from India. So his reply was very polite but he made some important points what he what he mentioned was you can go and see this you know watch this video. It is there but it could have been much more concerning the fact that India has a large number of educated English speaking people but still the contributions are I mean not that much. And he also makes some of the observation like for people like him coding came as a hobby first and then it became a job. And he says suppose software first it become a job or a profession then probably open source may not happen then open source will happen. Suppose if you take it as a hobby and then you move into that profession so that that is his you know kind of use on that maybe you can watch this video and. But all I'm trying to tell you is like probably as he mentioned we could have done better. And if you are not doing then what could be the reason one observation is we have education even starting from school level. Somehow or other we are see all CVC schools you will find a computer lab is there but what are we teaching. What I observed is we teach computer applications how to use it how to use it as a typewriter how to use it as a calculator. But we are not really teaching programming this programming is we are teaching only the small number were selecting computer science. All others are just that behind and why that is done maybe they are not capable. So here I have some personal observation but it cannot be taken as a sort of you know authenticated fact is only an anecdotal evidence. In 97 I along with some of my friends he conducted a camp in one school a vacation camp after these in May after the exams. And around 100 students were there they were all just finished their 10 standard written their 10 standard exam and all our middle and medium students. And what we thought is we will teach them to use this because those days those was prominent will next was not quite common. And basic concepts like files directories and all those things. And after that if time permits they were they were permitted to do some basic programming because it was part of another the interpreter was there. And that time what I noticed was those who are just doing the basic programming they were picking up things very fast. Because they were really fascinated by drawing a circle or something on the screen and just to making a multiplication table or something. So that means that it is not a such a great thing probably you could teach it to anyone. But in the beginning when the doors was there you had the basic interpreter but then later on things changed probably because we got into these windows for us. And at that point one thing very I would say personally a very harmful thing happened probably is we started pushing the office packages as educational software. Even today you will see that in all this what we are teaching at the school level it is like word processing and spreadsheet or just making presentations. So to plot a graph you will struggle with the spreadsheet Excel or Calc or whatever it is. And then another very futile exercise is making these presentations just making fancy things and you know it does not really contribute to the understanding. Then in word processing we focus more on typesetting how the font should look good and capital italics everything than the content. And if you go for Latin or something you would get a much better output especially when it involves mathematical equations. But we did not probably emphasize those things and we went this office of his packages. And probably that has done some harm and someone else also pointed out the same. This was if you go to the website of this particular website which talks about this Raspberry Pi history. And you know the original Raspberry Pi project it is very old I mean like it is a few people from Oxford, Cambridge and from industry. They were trying to address some issues and these were the issues and the main concern was the students who are coming to join the computer science program at Cambridge. In the 90s those who are coming for the interview they say they were experienced hobbyist programmers and the landscape in 2000 was very different. After 10 years a typical applicant there he has done some web designing or of course you have a spec sheet. So what they conclude is what something has changed the way kids were interfacing interacting with the computers. And one problem they identified is the colonization of ICT curriculum with the lessons on using Word and Excel or writing web pages. And I would agree with this thing because we also observed the same thing because I have seen like in 8th or 9th standard the question paper which my son wrote and the one question is what is the function of the home key. And what he is expected to write is how it behaves in the Word application. So it went to that level and what is the ultimate root actually. So let us think about that. Why not use IT as a supporting tool in education like we need computation, simulation, data visualization and where does Python pick it is. So can we use Python as a tool for that and how much effort is required what are the benefits. Let us have a quick look at that. First thing is it is easy to learn no doubt because a Hello World program in C++ is this much and this is in Python. So there is no mugging up required actually. So and here I would say I mean C is my favorite language but as a first language problem this is not advisable. C is good if you are programming hardware very close to hardware then it is good. So now how good it is let us just explore some simple examples. It is for you to judge they are useful or not. Like here say just one for loop and this formatting string is not compulsory. If I simply say print trial star file still the answer will be there but making a multiplication table is two lines of code. So that means we are not learning Python. Actually we are programming to learn something else. So we can explore these or make tables or print them all those things for that sake you learn this syntax of Python. So that becomes a secondary thing but indirectly you will start learning. So you start using it in a different manner. So another simple one like one equation and you want to see the time displacement plot. Why I put this is this equation is there in the 11th textbook and graph also is there. But how to get from this equation to graph? Either you have to do it manually in a graph paper or forget about it. But this few lines of code does the job. Now something slightly more involved because many people I mean you know that we are using a three phase supply. And all of us know it is a 230 volt supply. But if you take a voltmeter and measure between two phases it is 400 volts and the electricians know that. But even those who know that engineers if you ask them why it is 400, why 230 minus 230 is 400. Because when you measure between these two lines means you are measuring the difference. So difference between two phases why it is 400. But it is quite easy to plot the three sign ways and plot the difference. You immediately see that why it is so. Because when one phase is going up the other phase is in the other direction. So always the difference will be more. And here the advantage is now students can explore suppose a four phase phase AC how it will behave. It is a two phase how it will behave. Why people went for three phase instead of two phase. All questions can be asked and explored. So that pattern acts as a tool there. Now let me just take something from mathematics toy. So this is something which you can buy in the market. You can buy a toy store you can buy a paragraph. And you see here there is a big wheel with the tooth inside and a small wheel with again tooth wheel. So in this smaller wheel there are holes you put a pen or pencil here and start moving this. I mean without slipping because the tooth is provided for that. And then this pen will you know it will go through a trajectory you can see patterns like this. So it is an interesting you know kind of hobby plotting different packages. Suppose if you know a bit Python and you want to explore this you don't need this toy. So here you see some code because when you have big circle small circle and movement of this point is just given by its equation. And you can refer to the Wikipedia here. So all you need to know is this X and Y you calculate and we have plot. And here you can see this is radius of small wheel and this is the big wheel radius. And this parameter decides you know where is this hole in the from the center how far is this hole. So now I have three patterns here for different values 1.6, 1.5 and 2.5. So the sort of things probably when you are doing it you know in h standard or n standard thing it may be a bit you know fascinating that time. And in that process you will learn Python better. And this is another called its real simulation like how an electron or a charge particle moves in a electric and magnetic field this path. This is just picture is there in the 12 standard textbook but you can do the computation like you use the Lorentz equation. And from the force you calculate the trajectory. So such things I mean simple examples I have given only a few examples but that must give you an idea. So that means advantage is easy to learn then open source and we have modules for scientific computation that I visualize session. And of course there is a large user community and good documentation so that is the advantages. And now we will see what more we can do now I mean suppose somebody wants to contribute to this area what can be done. If you look in the web what I have found is there is a lot of material to learn Python programming plenty of tutorials books everything. And lot of simulations are also available but they are all slightly involved. So you can run them but you may not understand it fully. But like small program 5 lines 10 lines which I showed you like that paragraph or this charge particle trajectory that kind of code fragments I do not find very many. So probably if somebody is interested and they think this is a proper route to go. What we can do is we should make a collection of such programs. I started doing a little bit a small collection you will find in this XPISE.in that Python have you will find something. But the real big work is make a Python companion for all textbooks that you can do for physics, chemistry, maths everything. And what this companion book should have? Book can look into the problems or even the graphs shown in the book and equations. And probably we can just do I mean how to get those graphs from this equation or problem solving other things. And another area of interest is probably documentation in Indian languages. So people can contribute there. And this is about hardware like we do not really need to spend big money on laptop or PCs. Raspberry Pi should do our job if you see a school computer lab. And why I am saying that is even the old version of Raspberry Pi. This I gave in a one-side pi or somewhere I think this you can see the XPISE is running and some other simulation is running. Even Mayavi was able to run on one GB RAM Raspberry Pi. So it is a very cost effective and less polluting and less power consuming solution. So now next thing where Python can be useful is at least some of you may be electronics engineers or doing micro control programming. And all of you know about Arduino. It is a very good thing but only problem is their website itself clearly says it is for non-programmers. It is not meant for engineers actually. It is for amateurs, artists and all to get something done because the policy is to hide the microcontroller. So those who are very comfortable with Arduino and when you ask them tomorrow you go and do a big programmer other than this area. So they are having a little bit of difficulty because what you learn on Arduino does not get translated into that. Because the core microcontroller features are hidden. So for that we started a project sometime back I mean maybe 10 years back called Micro-Hope. It was microprocessor for hobby programming and education something like that. But later on there is another version called Kutipai was introduced. And that you will find here you can download the software from there. And here the approach is you have a Python program. And you have a small hardware board which is having an ATmega32 with a USB to CDL interface. And what it supports is you can access all the microcontroller registers from Python. And you can run Python scripts. Suppose you can connect in LED and you can write a Python code to make a blinking LED or something. And once you are comfortable with the microcontroller register all those things without coding the microcontroller actually. And then you can start writing in C or assembler. And like Arduino you can do a single click upload as well. So now you can search this website and the mainly the development is there is a GUI and all then also the bootloader. And I think there are bootloaders that you can put into Arduino boards also. So Arduino Nano or something. So you can convert an Arduino Nano using a USB ST programmer or something. Then it will behave like Kutipai board. So you can use all the software. So this is something which probably a useful development. And this I am putting here because it's mainly as I told you I have been working in one place for around 35 years. And this year that is over. So this website I have set up actually site school.in as a platform to you know put all my works and maybe coordinating the things. So what I am planning now is sharing some knowledge whatever way we may be conducting small courses or something. Python microcontroller and some science experiment and also the learning management system. And also how to set up similar things like this on the cloud how do you set up all these servers and all. And here I have put a link here if you come to site school.in you if you think and you should be informed about you know further development if I start doing something. You can put your contact information there so that you will know if something happens here. This was another area and now I am continuing to interact is this is in IUPAP the international organization. There is there are different commissions in different subjects for that there is something called commission C13 for fix for development. In that we have a small group for affordable scientific instruments. So here also mainly what we are doing is looking for proper open hardware things that are useful in the field of education and compiling it here. And also planning to support in different ways. So this is some work which is going on at a different level. And now let me just finally we come to this point just to make a couple of comments. So yesterday in the panel discussion there are some arguments like should we teach programming to everybody or we should teach only selected ones. I mean if everybody goes after computer programming it is not really helping the education that is one argument. And okay let everyone learn computer programming that is another argument. So but whatever is the fact I strongly recommend you people to go and read this document. So if you go to or you can simply search this but it is python.org, dope, essays, cp4 computer programming for everybody. So here is a very good document it was prepared in 1919 and when Guido himself submitted this as a proposal. And from this proposal there are very interesting points in this proposal I just picked up a couple of them. Like what will happen if the users program their own computers. So it is a very far-fetched idea and he says we compare this mass ability to read and write the mass literacy and call it as a software literacy. I mean it becomes a mass programming literacy kind of thing that is what they imagine. So and the requirements because this is written in 1999 and most of the requirements they have met actually developing a new computer curriculum. That is if you look at it many places it has already taken place and also the tool the python has become a good tool for. So I would recommend to maybe you should have a look at this and my view is that it should be taught to all and I take this argument. Like you teach a skill doesn't mean that that should become a profession. Like we all want to learn driving right even before the age of 18 you start learning driving but you are not going to join Uber or Ola as a program as a driver, taxi driver. But you want to still learn that skill. Similarly we should treat programming even if you are not becoming a professional programmer we are not going to take it as a job. It is better to learn that skill because it is not that tough and it may help you doing the other things slightly better. Okay that is my opinion. Now let me summarize what we talked so far. That means considering situation like our contribution in the open source seems to be less. And that means we need more number of good programmers. And to have more number of good programmers you should have much larger number of programmers because among the programmers when you do the sampling there will be good programmers. So that means you have to widen the sampling base. That means there should be more people exposed to the programming. And how to do that an Indian situation the best place is actually the school. The conventional school if we start using software as an enabling tool to teach the core subjects it may happen. And Python language seems to be suitable for that purpose at the moment. And another factor is the current system of teaching the office packages in the name of education software it should go probably. Because if there is no need students can do it themselves. And the Python language says like you probably you can help preparing material or read some projects. So that or you make something and go to your own you know wherever you have done your schooling and conduct them and just try to push this idea if you consider it is worth. And as such I am looking for collaborators as I told you somebody who wants me to you know be in touch with you just put your contact information on that website. So I use that as a platform. So more or less that is all I wanted to share with you today. And maybe you people can say something. I mean question means I am not going to give an authoritative answer. I'll just express my views and your question will be probably you can express your views as well. It may not be called a question answer that way. Please go. Awesome. Awesome. That was really insightful definitely. So let's wait for a couple of seconds. And there are a few questions or points to discuss. I will show those to you shortly. Can you expand a little on how x y's makes the actual interfacing between the sensor input and the Python happen. So there it is done like this. You just take a microcontroller and write some software so that you can interface it to the PC through this USB port now. You will say USB to digital converter. And interfacing the sensors which is done in two ways. One is the microcontroller. See what is going to that. But there are other pins where you can connect the ADCs are there and local digital converters are there. And it has the timer counters and all those internal resources where you can make skin generator all those things. And it also has the I2C interface and SPI interface. So either we use an ADC suppose say one simple example is we have a DC channel. So we make a digitizer. Connect a simple condenser microphone and you digitize sound. So you can do a nice experiment to measure the velocity of sound and it costs you only 10 rupees. And next step and the new version you have the capability to connect the I2C sensors. If you guys see the slides video yesterday just in percentage. There is a wide range of sensors available and they are only 100 or couple of 100 rupees. So they are getting interfaced to the I2C interface. So here the role the SPI is playing is I'll tell you right from the beginning in 2005 we put one line on our website is that is to develop science experiments without getting into the details of computer programming. There is one line or electronics. Because if you want to write I2C code a programmer can do that who knows C and all those things. But here what we do is expires provides the code which will communicate to the I2C device and all that low level code. And you can write code on the Python side just calling this code on the other end to plot the graphs and other things. So it's a kind of job division. So that is how it is done and the entire circuit schematic and all those things are actually available on the website. So if you really want to but I would recommend if anyone wants to make such a thing like a box whatever you call it which is connected to us before and you want to do a sort of communication and connect sensors. You should start with Qtify because it is a simple point to start with and AVR these processors are plenty and you can use any Arduino board also to do that. And there you will find if you go to Qtify website or the Macrobob website there is oscilloscope code which is 50 lines 60 lines still working as an oscilloscope without bells and whistles. So this is all I can say but complete details you can get from the site all the schematic everything. Do you feel making the primary language for students in 11, 10, 12 in Python better than in C? Yeah, okay. See this is the again it has been done now some experts from IAT and all in the committee finally said we should teach Python but even before that maybe for the past 10, 15 years I have been telling this for the following reason because we have been teaching C and C++ or mainly C++ in the 11th and 12th as a separate subject and it has no connection with mathematics or physics or chemistry or anything. So it was doing a parallel run but when you go to Python instead of C what will happen is you can convert that into a supporting tool for the learning the core subjects. And once you know programming for any language switching on to another language is quite easy. So that way switching from Python to C, C++ is not that difficult but starting with C++ I would say even though I started with Fortran then to C, C++ like that but after seeing Python I find that this is a good starting point. It may not be good for all the things but it seems like better than C as a first language and especially when it's you see it's I already showed you some examples considering its capability to support the core subjects like math, science or chemistry and so that is my view on that. All right. The next one is do you see the application of programming in non-science subjects like fine arts, linguists, linguistics etc. Very hard to comment you know fine arts and all yesterday we had the panel discussion right so you remember madam was telling that the way in which her field of work got really changed when she learned programming also. I can only just believe that because since art is not something which I should comment on but I think it should probably change slightly you know improve the things and it may help. That's what I feel. It won't do any harm definitely. Maybe check out the recording of the discussion of yesterday for those who have missed it. It was a very hot discussion actually because both the points were presented. Next one is for people who have come through the education system like we discussed right now do you see a way to still evolve out of the mindset of programming for jobs? Well see that is once you are through with it and if suppose you take computer science as your subject and all then obviously they will be looking for programming jobs. And so whoever has gone through that I mean there is nothing much we can because they already know programming they will stay there. All I am trying to tell you is suppose the other side was a physics or mathematics person who has come out of the course without any training in computers they could still start so they can make up for what is lost. So at least when they will be teaching that subject that becomes useful for the next generation so or it will help them in a different ways actually. See in this context I can tell you I mean someone asked me to teach some Python and finally asked about the nature of the people who want to they are basically they want to analyze them that they are mostly playing with totally spreadsheet and all. So I just had a look at it and I find that even in a spreadsheet suppose if I select some column put some conditions everything I can do following the GUI. But if I know the code and it just two three lines and writing and it becomes much easier so you are when you are using a GUI you can do everything but the amount of effort you are putting there is probably much more. So learning a little bit of scripting will help and you don't have to call it as a full stress programming or anything. I hope it answers I am not sure. The next one is what is Python support for DSP which is digital signal processing. Do we have Python modules. I have not come it may be there but I have not come across actually DSP have not used DSP. No, I cannot answer that it may be there. I don't see any other questions but there have been a lot of discussion in the chat. People are agreeing about the points they were putting up their thoughts and opinions. That's more important because their views are as important as our views. And ultimately they are going to decide the young ones are going to decide which way things will go. Because for them it's morning and for me probably tomorrow is evening so it makes a big difference. Definitely. I guess that's all the questions we had. Okay. So I hope they will continue to interact. There is one maybe we can have that question and then. Yeah, sure. Time is there. I do think having the ability to switch courses subjects more easily while pursuing BE which is Bachelor of Engineering. I guess it will help a lot. Do you agree and how easy or hard it would be? It's actually it's a question to an expert in education probably but still I can express my views. See we are the products like our system has been very rigid. Like you will learn once you select when you enter the college you make a choice and you exit with the same choice. You cannot make any change in between. But that is not the case in many places. It was not like that. And now we are slightly our system is also changing giving some flexibility. And we should also see that the whole scenario is changing. Now if you see the edX and Coursera yesterday there was an argument it is only helping people who already have some background. But I really don't think because I sometimes take some courses on biology or something where I have no way. But it's quite useful actually. And it's slightly better than reading a book or watching video or at least some assessment or those things that. So things are changing actually. So we should have the flexibility I would say. But at the same time some amount of certain stuff should be covered. I mean that means there should be a fixed part but there should be some part of it should be kind of students choice. So it should not be that rigid. But our system is changing now. So that is a good thing now. Our universities are also considering those changes. So I would say it should change for more flexibility. Great. That's all we had. Thank you so much for a great talk. Thank you.