Month: August 2020

{{This article was originally posted on January 2, 2017. Lost and found, it is reposted here.}}

Reading this thread on hackernews (https://news.ycombinator.com/item?id=13297250) about how to learn and then going through the syllabus for the course on Learning how to Learn (https://www.coursera.org/learn/learning-how-to-learn/) on coursera got me thinking about my own methods for learning.

Over the years I have become an avid learner, I guess, this is my way of taking risks in life. People take risks in business or work their do or in sports. I take risks in learning. Learn everything I can, most often, for the sake of learning.

Earlier on I realized and differentiated between formal learning to learning for self.

My way of learning is probably different to everyone else’s, as probably is for everyone else, in that I tend to learn new things in either “Binge” mode or in “short bursts” mode. In both I tend to use open ended questioning, perspectives and reflections as ways of enhancing the learnings. I really get excited by a subject when that subject occupies my mind while watching the most beautiful sunset, high up on the hills with a chill wind blowing. When I begin to think of new ways to look at the topics and start thinking about what possibly is, then I am ready to begin my learning gears.

Precisely due to these reasons, formal learning has been a challenge for me even from earlier on. In many occasions my paths have taken a wrong turn from teachers and others. Few incidences that quickly come to mind:

I really wanted to come first in class atleast once. Being of a calculative mind, I figured the best time was the first terminal exam in class 8. By class 7 I had started moving into the top 5 places in class. There were a few courses that, I realized, that I would never score high. So I needed a time when material in such courses were low and we discussing more the introductory material. This all suited the 1st Term, Class 8 thought. To add to this the person who normally came first was about to leave the school and his frame of mind was not geared towards the exam. Few others, who were up there, were just slowly getting into the mood of studies. So I bucked up and studied pretty hard. Come exam time I knew nearly every material. So I was flying through the questions. I even managed to help a few students around me including the person who used to come first before. It seems the teacher had noticed me turning around. Come the result time, I came first. He called me upto his desk and told me congratulations I was overjoyed, but few instances later he said “you cheated”. I told him “No”, but he kept saying that it is better to tell the truth. Since then I promised that I would never try to come out first.

I studied from class 3 to 10 in a school called Goethal’s Memorial School in Kurseong, Darjeeling, India. It was school started and run my Irish Christian Brothers and strictly adhered to all of us speaking in English at all times, except during second language classes. My mother tongue, Nepali, was a second language there. I considered my English to be pretty good by the time I reached my high school. In high school I had an English teacher named Usha Mathur. She literally taught English in Hindi. Come the first terminal exam, she asked questions regarding English grammar. I got a zero in the exam. Quickly she said to the whole class that I didn’t know English and I would never really catch on to it.

Another incident occurred in the Technical Drawing course. The teacher, Mr. Bharadwaj, came in the first class and told us that he would pick the best students and place them in the front of the class so that he could provide more guidance. Others would be placed behind based on their performance. Next day I and a group of hostel students were assigned to the last bench. Mr.Bharadwaj would not even come to check out work even if we made a racket. During the final exams I got marks among the top students in the class. When I met him next he failed to recognize me.

Despite my father saying go in to study management I went to study Engineering during my Bachelor’s. Few classes down the road I realized that I had made a wrong choice. I had a hard time going through most courses. Statistics was something I liked and understood little, but due to pressure of other courses I did really badly in that too through the semester. Towards the end of the semester, I thought maybe if statistics is the only course that I seem to enjoy a bit why not try and get good marks in it. Others I targeted only for passing. I did a binge session (explained later) on statistics and managed to get a fairly high score. I actually got an A grade in the exam itself, but due to bad grades in semester it was pulled down to B. Even the person, who would later on become my spouse, protested against me getting such high scores while they either failed or barely scraped through.

Many more stories in the formal learning sector, while didn’t seem to drive me at all.

Informal learning is where I excel and with Internet and online courses this has become better and better. I study any topic that I feel excited about — when people talk to me, when I see something in somewhere or just out of pure curiosity while browsing sites.

As mentioned above the two methods I rely on are — binge and short burst.

Binge is method I use to learn a lot about a topic in the shortest period possible. I start a binge with either trying to understand more about a topic or to try an see whether a hunch I have in regard to the topic is well worth following. In binge I start closing down questions to a few that I feel are relevant to me, reducing overall open ended questions. I home down to 1 or 2 perspectives and stick to them. I substitute reflections with serious studies. Good side of it — I learn a lot about something and extremely fast. Bad side — I tend to also forget very fast and need to do another binge the next time I need to learn up the topic. Some quick examples:

I wanted to learn this new fancy Data Science related stuff. For that I wanted to quickly brush up on my classical statistics knowledge as well as try and learn a few things that I had skipped before. Binge made a perfect sense for this. I limited my open questions to quick refresher about exploratory data analysis, hypothesis testing and starting of regression. I really didn’t need to look at other perspectives like Bayes stuff or Time series stuff as I would do that through the data analysis portion anyways. I downloaded 7-8 books — 4-5 pure statistics books, 1 Python book and 2 books that showed examples in Python. I read through the pure stat books. After the first 2 books I was reading only the last few chapters. Then quick examples through the Python book and some practicing. Few videos to understand how classical statistics is being done by modern programmers (statistics for Hackers videos). I did about 4-5 hours per day for 5-6 days and I was upto speed.

We had been having a discussion regarding why Nepali people are tolerant, sometimes even awed, by people who cheat or bend ways for personal gains. I kept arguing from the point of view that genetically we might be built that way. Probably our culture portrays that as being good. Cultures are written down in the stories that are told. So I decided to test my hypothesis by reading as much folk stories as possible to get an insight into most prevalent central theme. In this case I already was fixed on the questions and perspectives. So I did a binge reading of over 200 stories to try and figure out a central theme. I don’t remember the stories anymore, but I came up with options for central theme that met my immediate needs.

Short burst is a method I use to learn something more deeply and more over a longer term. In this I put in 5-10 minutes 3-4 times a day maybe for a period of about 2-3 weeks. Then rest and reflect for a couple of weeks and then go through the cycle again. This allows me to get into the topics, ask myself questions, reflect upon the answers, try and follow different perspectives and check up with further learnings. In this I tend to try and use as many different ways to get information as possible — books, videos, trying, talking, reflecting, just about everything. As I talk to as many people about this as possible, I tend to get a lot of different perspectives and this drives me more to learn about the topic. Also instead of trying to form 1-2 hypothesis like in Binge, here I tend to make many and test multiple in multiple different ways. The best part about this is the obsession of talking and learning from everyone, even from people who do not have a clue about the topic. That surge in adrenalin sometimes keeps me awake through the night and I tend to get hyper. Just the sheer feeling of knowing is worth it.

I have always wanted to learn how to play atleast 1 musical instrument. Beginning of 2016 I made a small promise that I would learn atleast something by the end of the year. Coming to November I kind of realized that I was really not going to learn anything. So to motivate myself I tried a hand at conch (sankha) blowing. I thought to myself it didn’t require rhythm and didn’t need to be part of other instruments. I guess that was harder than I had ever imagined. It took me the best part of 2-3 days for even the sound to come out. When sound came out, it came out in a monotone — for which everyone started shouting at me as that style was only blown when someone died. Further 5-10 days and then I started blowing a little better. I practiced 3 minutes in the morning and about the same time in the evening. During the day I reflected on the sound atleast 3 times. Then I listened to conch playing on Youtube atleast 2-3 times in the day for about a minute. Few days later I was a lot better and still going at it atleast once a day for about 3 minutes.

Gaining confidence from it I wanted to try another musical instrument. So I kept pondering over many instruments and who I would try and learn from. I listened to many lessons on youtube. Either I didn’t have the instrument or I couldn’t find the person to teach or I wasn’t motivated enough for it. So until December 20th I didn’t have a clue what to do. I thought this is a promise that I need to extend to 2017. Early morning of Dec 21st, my son was playing the maadal (a Nepali drum based instrument). I then had my motivation, instrument and teacher. I asked my son to teach me what I thought was a simple beat. He taught me a few times, and the young guy as he is, he got impatient and went away. I practiced the tune 4 times that day — in short burst of about 5 minutes with a gap of atleast 2 hours in between. Next morning I was a lot better. 2 days later I asked him to teach another tune. Again 5 minutes each 3-4 times a day, now both tunes. Then a tune from youtube — come Jan 1st 2017, I now know 4 tunes and can play along with songs on youtube.

I have always had a fascination for agriculture. My interest in it grew much once I got involved in hackproject.org and a project that I did with a student called “A bucket a day”. I wanted to start learning about it scientifically. My approach short bursts — I first divided my interest in it into categories — fertility, water, soil, weather parameters and other factors. Then I got a series of books and lectures on introductory agriculture, did a quick binge to get the balls rolling. Now I am doing short bursts into different categories along with testing out application development. Now I am testing controllers with raspberry pi and arduino. I have increased interaction with people at all levels to understand more and more — both traditional knowledge that people have and what has been done scientifically. I feel learning needs to be always backed up with discussions. So following along the path.

One thing that I have not talked about throughout this post is the the power of motivation. Unless I have motivation, I am unable to use binge or short bursts. So the first step is always the generation of motivation — for me it is the level of curiosity towards that topic.

{{This article was originally posted on December 24, 2016. Lost and found, it is posted again. }}

The Central Bank of Bangladesh was broken a while ago and over 81 million dollars heisted. Though there were traces to try and hack into over a billion dollars a minor error in the part of the hackers led to the heist being uncovered and duly blocked. 2 months on the case has completely gone away from our minds as we move on with our everyday lives knowing that such disasters might never strike again.

When people built the Titanic they claimed it was unbreakable, yet it was broken in its maiden voyage. The feeling or sense of security may not be in actuality the realization of security. As was the case with the Central Bank of Bangladesh.

I had been following this case as much as I could due to my varied interest in communications and communications security. This case was interesting for me from the communications point of view, mainly in how the communications infra-structure was utilized for the heist.

Due to the sensitivity of the case a lot of forensic analysis is being carried out around the world to try and figure out what happened, how it happened, what were the weak links exploited and how. Recently a forensic report article on the Reuters (http://www.reuters.com/article/us-usa-fed-bangladesh-idUSKCN0XI1UO) caught my eye. At first I just read through it as any other article on forensic analysis and accepted the findings by thinking it would happen to others too. Scrolling back up to the heading some words stood out with prominence — cheap switches and no firewall. Then I read the article again, but this time trying to put Nepal and Nepali organizations in the places where Central Bank of Bangladesh was mentioned.

Since I am an “IT” guy, during most of my visits to organizations I am shown either full or parts of their IT infra-structure. If I am doing work on system design or development then 100% of the time I am shown the entire IT infra-structure. In most cases I have seen cheap switches, that we normally tend to use at home, being used in even very large organizations or even in financial institutions. I don’t know much about the firewalls as I am not a working computer security professional, but I see the hardware.

I remember a time while talking to a security professional and him telling me that the Government of Nepal should not give licenses to people to use certain frequencies because the security professionals and institutions used them. I disagreed with him on the term that atleast one knows that these licensees could be using the frequency. What happens if a malicious use uses them without a license? These days a Software Defined Radio costs less than 20 dollars, is easily passable as a USB dongle and can be used to scan and read frequencies in the range from 100MHz to 3.7GHz. This means the device can scan the GSM bands, wifi bands besides the VHF and UHF bands that are most commonly used for all communications in Nepal. Wifi scanners on mobiles can break into Wifi networks within minutes if security features are not activated properly.

The advantage of IT far more outweigh the disadvantages and threats from it. So avoiding IT is not a solution, but being prudent and living upto the challenges is a better way ahead. Understanding that the weakest link controls the strongest of security is crucial here and should be given priority. The usual notion of it has worked and will continue to work should not be accepted. An openly available statistics on security breaches available at http://bhutabe.blogspot.com/2015/04/nepal-web-security-status.html on Nepal looks pretty dismal.

This brings me back to the Reuters article and the usage of cheap switches and lack of firewall. I feel we can agree that most organizations do use cheap switches. Also that most organizations do not run any firewalls. Also most organizations run softwares where quality is compromised to cost.
This I call the “Cheap Chinese” syndrome.

Since our markets have been flooded with low quality Chinese products (disclaimer : I am not saying all Chinese products are low quality) at extremely reduced prices we have flocked onto them  for all of our needs,

The biggest problem in Nepal is that most human resource in organization do not really have sufficient knowledge regarding the usage of IT. Therefore the hardware and software are compromised for price rather than quality. Most organizations lack policies or practices that help better use the boon of IT and communications technology. Most organizations do not have or do not commit enough budget to better enhance their IT infra-structure.

All this helps me frame a dire question in my mind — are we on the path to become the next big victim???

{{This article was originally posted on December 24, 2016. Having lost and found, it is posted again.}}

While being obssessed with climate change and looking into making applications that could be effective for people to manage climate change in multiple ways, I came across the story of these 2 guys — Jadav “Molai” Payeng from Northern India and Yacouba Sawadogo from Burkina Faso.

Both build forests out of barren lands, showing that it is possible if one tries. SALUTE to both these guys.

Inspired by their example I wanted to learn about how their methodology worked and what we could do in similar lines to create new forests. Also how could we use similar ideas in agriculture for food security.

Article of Jadav — http://www.mnn.com/earth-matters/wilderness-resources/stories/indian-man-single-handedly-plants-a-1360-acre-forest

Article of Yacouba — http://www.odditycentral.com/news/meet-yacouba-sawadogo-the-man-who-stopped-the-desert.html The commonality in both seem to be usage of native plant species, compost and bio material from nearby areas, optimum utilization of rain water and other water sources and working in partnership with nature.

Looking around found that the technique (though age old) is contributed to a Japanese person and is called the Miyawaki Technique.

Based on this technique an Indian company has started a commercial venture called Rapid Forest (www.afforestt.com) and has open sourced all their techniques, tools and methodologies.Shubhendu Sharma, the founder of Rapid Forest, has done a ted talk and ted blog : http:/http://blog.ted.com/shubhendusharma// and http://dirtyclassroom.com/diy-build-your-own-forest-alex-eisenreich/ This lead me to a binge on agriculture, irrigation, local knowledge, etc… that is documented in another blog article.

{{This article was originally posted on December 23,2016. After loosing and retrieving it here it is posted again.}}

I was (still am) fascinated by how micro-financing works, specially they social insurance and social credit system. I got hands on, in the field sessions with 2 micro-finance Organizations (MFI), Chimek Bikas Bank and Jeewan Bikas Sanstha, is regards to some work that I was working with them on. During these trips I was able to see how the credit was provided for individuals as being part of a group (social credit). This created a little bit of restrictions in that the person needed to be known and was bound geographically.

With the blockchain revolution I thought that the micro-finance’s social credit could be taken online. But I see one issue, the initial credit that needs to be provided for one to become known to a social unit. In the current scenario that is being done by physically being known among someone in the group. The analogy does not look possible to replicate in blockchains.

So I have been thinking maybe an initial credit rating of a person should be done that could then be used to continue forward in the blockchain. As the person does not have any collateral or even a social network to show, we might need to be imaginative in doing this credit rating.

The initial credit should be :
Profile + weighting factors

where Profile :
Personal Income level
Household Income level
Mouths to feed
Potential of income in near future
Land
Business
Vehicles
Any terminal illness (I feel this needs to be there, but am not comfortable as it marginalizes)

Weighting Factors:
How long has someone stayed in the current place?
How many times has someone moved locations so far in life?
Current family members vs members outside (within Nepal and outside)
Own land location
Own home location
Current field size VS soil type in that region
Current animals VS national requirements for those animals
Personal education vs district education
Personal income vs district income
Animals possessed
Variety of animal vs animal requirement of district and country

As all financial facilities go, there cannot be an access to major funds right a the beginning. Credit has to be built. There always needs a starting point. So making initial credit rating is starting point. Incremental building of credit rating by honoring the guarantees — this will allow to develop trust over a period of time.

Individuals would try to get into the blockchain so that they have a credit rating and social guarantee to be eligible for loans and other financial transactions. Banks and other institutions could look into the blockchain as a way to make informed decisions regarding particular person.

In this case not having anything on the blockchain would be penalizing for the potential user. So incentive to put everything there could be rewarding enough for the motivation to just do it.

Note : when I discussed this with my friend, Lewis Kaneshiro, he deeply feels that putting a person’s information in blockchain itself seems not like a good idea as it puts someone’s entire life into a permanent storage that people can query into way past the person. Looking at it from that perspective I left this thought as a thought only and didn’t do a binge on.