Specialized Learning vs General Learning

[crosswire]

The "Technology" field is so huge.

Communication, Simulation, Robotics, Databases, and so many more.

Each one of these area is itself also huge.

By the time a student is ready to write code for a recognition app, he/she would have completed many years of study. That time would be shortened if they specialize from earlier.

Back in my time, there was no IT class at my high school, but since that I have seen teachers "Introduce" Microsoft Offices Products at the Primary level. Now it is at the secondary level.

Now there is a recognised curiculum in IT
http://www.cxc.org/subject/information-technology-csec/
https://cxc.org/SiteAssets/syllabuss...Technology.pdf

http://www.cxc.org/subject/information-technology-cape/
https://cxc.org/SiteAssets/syllabuss...Technology.pdf

Let's imagine Jamaican Programmers writting a traffics simulation app.
Should a high school student spend even quarter of his daily classes in a specific area of Computer Science/Information Technology, then he/she would acquire advanced training in that specific area. Let us say, each year they do the following:
C++/C# programing basics --> collection of apps in C# --> simulation basics --> exposure to code from other simulation apps weather, traffic, physics --> taylor a traffic app for Jamaica

Let's imagine writing a robotics app to control a robot with 4 hook arms to climb a tree like a monkey. Picture it traversing a horizontal tree limb by using its weight, or a counter force to push away/down from the limb while it places its hooks on the limb. Then it moves by putting one hook in front of the other and so on.
Another aspect of the app would be "visual recognition" of the tree limbs from the output of a camera on the robot. The major components of the app would be done by different specialist.
C++/C# programming basics --> collection of apps in C# --> recognition basics --> exposure to code from other recognition apps audio, visual
C++/C# programming basics --> collection of apps in C# --> robotics basics --> exposure to code from other robotics apps

One student could have more than one special, because the special field is "narrow" and specific.

It is possible for a student to pick up something specific on his/her summer break, during a specific summer class. Let us say game developing, or it could be web site developing. Each summer, they would acquire a specific skill set in game developing. However, they would lack the overall skill set to get a certificate in IT. (I can relate to this and a computer science major.) And there is no recognition for high level skill in a specific area.

I could go on and on with other specialized areas but that is a disadvantage of specialized learning, among-st others. There would be fewer students in a class, and more classes in a school.
However there is the advantage that a student could go further with code within a specialized field. Jamaican could pioneer apps that would take others longer

Discussion
Should there be more specialized schools, so that there is a mix of specialized and generalized schools?


Furthermore, organisations should have temporary projects building unique apps, which would employ these specialized student, and give them specialized training to carry out the project apps.

In my first year of exposure to computer science, I wrote almost the full code for C++ console app for playing chess. I was exposed to just the code to write the bishop piece. I took that and coded every piece. The letter 'b' would represent one's player Bishop and a capital 'B' would represent the other side's Bishop. 'r' would be Rook and so on. I even placed the correct weights to the value of pieces. I had great difficulty with the AI, and at that time I could not get a three move ahead search to work.
I did that before finishing a whole year. And I was at university doing other classes. Furthermore, I it took me about half a semester, about two months, before I understood the "code flow" in an app, and to write my first hello world with application. Had I been exposed to more AI code then I could have done a good C++ chess game. I was also using a book on my own to learn MFC and create windows app within my first year because I was having so much fun. There was no curriculum for that but I liked it.

[crosswire]

If you are wondering:

The benefits of a traffic simulation app tailored at Jamaican bad roads

After considering factors like
Delay spots on road - narrowness, potholes, water settling spots after rain - use an object with a variable value
Timing Settings with Stop Lights
Various Driving behaviours


Then a simmulation app could be used to optimise traffic, and see what changes could be made and how greatly it could improve traffic congestion

Example, how would fixing these potholes help.
how would resurfacing this stretch help.
how would these stop light timings help

The app could do its own search and provide answers to these? (by doing a search on all the possibilities of fixes of each irregularity of the road surface, and pick the best one in each budget category. Also the best stop light timing in 0.25 seconds adjustments.

Manual input of new roads that would be proposed could be done and the app would show how much it improves traffic. Likewise reordering the direction of one ways in some towns.


The benefits of a robot climbing a tree.

Once it is safe from falling and injuring others.
To PICK FRUITS

Some trees are so darn tall, over 30 feet, and dangerous to climb like an ackee tree. If a robot can be programmed to climb a tree then with some more tweaks, then it can be programmed to pick ackee. Maybe a different code would be needed to recognize different fruits.

Tasks like these could be projectized to specialized students by bigger organizations.

[khat17]

I skimmed - I didn't read. But I like the thoughts. Part of the problem is human nature - not just the Jamaican mentality.

If you want to build something well then all parties involved need to at least have a GENERAL concept. I did programming back in 1997/1998 and I know the fundamentals of where things should go - but I couldn't program myself out of a lunchbox currently. If I discuss with a programmer about how things are going and how they should flow then I may give him/her ideas, but I won't give any unrealistic deadlines to get things done.

Specialized and segmented work is good if you have a singular goal in mind. If you're doing robotics for example (not that I have) and you're making an arm to pick something up, the persons doing the mechanical work should have a BASE on how the programming works - and the same should go in reverse for the programmers. If everyone involved has no idea of how the other department works, then at the VERY LEAST you should have collaboration where things are discussed. It makes no sense to have the engineers design an arm that can rotate 360 degrees when the programmers only write something to turn it 90 degrees. Unless you have future plans.

In any case - it's a good idea as long as the persons managing the project/s have good heads on them. But the examples you gave as well are good ones - useful ones. If there's no end goal in sight, or any benefit to everyone, then the project makes little sense.

Will it get done? With this corrupt economy? Maybe - but how much of it will actually get the full funding it needs? We will see.

[owen]

IMHO most if not all schools apart from specialized departments with big budgets exist to teach general concept. they do not have the money to teach specialists. Allot of people ask me which school I went to to learn the things I do and I tell them you can't learn stuff like that at school. You have to go out of your way to get the extra knowledge and tools needed to really specialize in stuff nowadays. Its hard to find people to even fix basic electronics.

[sumo]

Specialization from high school is a bad idea. Learning the syntax of a language and slapping together some code that kinda sorta solves the problem does not a good programmer make. Good code comes from thoughtful consideration of the problem and associated constraints. It's more about the way you think about the problem and less about the code that you eventually write. So if you try to narrow students too early they won't get exposed to enough material to make them understand they should take one approach vs another.

I actually think the opposite is the best approach. Expose students to as many aspects of programming as is feasible. Let them see the different kinds of things that can be done and how smart people have solved problems before. Hopefully they get interested in at least one area and start digging further on their own because no amount of school/classes can make you a good programmer - you have to put out the effort on your own.

[khat17]

Specialization from high school is a bad idea. Learning the syntax of a language and slapping together some code that kinda sorta solves the problem does not a good programmer make. Good code comes from thoughtful consideration of the problem and associated constraints. It's more about the way you think about the problem and less about the code that you eventually write. So if you try to narrow students too early they won't get exposed to enough material to make them understand they should take one approach vs another.

I actually think the opposite is the best approach. Expose students to as many aspects of programming as is feasible. Let them see the different kinds of things that can be done and how smart people have solved problems before. Hopefully they get interested in at least one area and start digging further on their own because no amount of school/classes can make you a good programmer - you have to put out the effort on your own.

Agree/Disagree. Have the specialized departments available at the later stages so they can choose where they'd like to go. Give the option in the later stages but make it general in the earlier parts.

[sumo]

I think that's the goal with a masters in CS and beyond.

[crosswire]

All great responses, really.

Thanks Khat.

I am going to read them all again three more times because they are that good.

Bear in mind the problem proposed - involves: the long length of time needed to train.

How long would it take for a group of Jamaicans to learn how to build and code a robot, and form a robotics specialty team.

Another more generalist team would look at the world, and find applications for robotics, and decide what problems and what type of robots could be used to solve them.

That same team would do the feasibility study (and include the head of the robotics specialty team, and a project manager). For example, they could attach a recognition app to a school camera system that would alert on violence in the school such as pushing, fighting, drug use, knives and guns present. OK it would be a little more that recognition. Some simple Artificial Intelligence* would be involved. The output would be a silent alert, and a pic sent directly to the principal's phone. (It costs to have security people looking at cameras, and if there are many cameras, even more people are needed. However, a cheap multi-core processor could assist with this task. That would be evaluated in the feasibility study. Also, such an app could be commonly applied, after small tweaks, to more than one schools, or to other places, such as airports, bus stations or other places where security is needed to be improved.

This team could be comprised of people in the Research & Development department of a bigger organizations, and a Project Manager from that organization.

Based on their research, they could projectize such an app development to a specialized robotics team.

Think about the experience that would be gained by the specialized team as they work on a recognition app for this one problem. Then another big organization, and so on, could summit other project ideas based on the need for a type of recognition app, audio or visual. Likewise, more experience could be gained in building different types of robots that could do various audio and visual recognition. Just like how there are established specialties such as database development and web server development and certification for these. Such specialized teams doing these fields can readily get projects from bigger organizations who may need a tailored database need, or web service need.

Not much Jamaicans reach University/Tertiary level, even less reach Masters Level. However, I am positive that many Jamaicans could build a robot or develop a game, if they could get an earlier start on it, because it will take time to learn. You can have someone who did not get to spend 3 years at University, but instead spent 2 years at an associate degree in web development, and became pretty proficient in that specific field. Likewise it could be database development field instead. You can have someone finishing a IT/Computer Science degree and have a general broader IT knowledge based, which no doubt has its advantages, but its disadvantage as well, in terms of time, maybe another year would be needed to fine tine the web development skills, a total of 4 years.

Why does it have to be so far up before some specialty skill is learnt. Why would game development/robotics/simulation be any higher tier than web server/database. Think of an associate degree, and subject prerequisites. IMO you would need at least 2 years of study in any one of those areas to be of any use. Plus another 1 or 2 years needed in common programming basics. Plus, in the case of gaming, some general basic knowledge of maths, physics, networking is also needed, but also at the level to be used in that specialized field, namely gaming physics, peer to peer and peer to server concepts. Another example is robotics, some mechanical, and physics knowledge is needed up to the level of that specialized field. Then even further learning into any AI to do a basic control of the robot. Technology intertwines with other technology, but it takes forever to learn it all.

I agree that integration of different technology requires knowledge of more than one technology. Example a robot could need to employ a database routine. This is self explanatory. Not possible if you do not know basics of each. The overall consultancy for projects that involved technology from more than one specialized team could be done by IT professionals with Tertiary BSc/Masters. They would be responsible for integrating different technology components. Imagine a database mining app that feeds relevant data into a simulation app. That project solution would involve 2 phases. The components could be developed by specialized teams having associate degree, but the integration and planning could be done by another generalized person, even if that person is not extremely good in each specialized field as long as he/she has a good head on there shoulder.

I am not saying that generalized learning should be done with. And I agree that a general learning foundation is key. But how the curve moves from general to specific along the learning path in time does not have to end at about 18 years. If more fields with shorter specific associate degree learning are available, e.g. game development, maybe more students would get involved. A lot exist online, example on YouTube, there are tons of tutorials on how to do Android development, Unity tool, and how to build games for phones, but no real accreditation or associate degree in this field. It is difficult for a student to go out learn on his own, and build an entire game. However if he could prove that he/she has the skills in that field via certs and degree then he could get hired as part of a team that would be working on a bigger game app for phones. There is also nothing stopping someone from doing both an associate degree and a bachelor degree. Either can be done first then the other while you work. And both will allow you to get a job or start your own thing. IMO, there could be more specialize Associate degrees. Plus if that field is a passion, then some people would jump start into that. They can know what specialty they would like before 18 years.

You finish high school o levels at about age 15-16 years (back in my time) and you still have a long way to go before you can work on a team for a traffic simulation app! But if there was a 3 year associate degrees in gaming/simulation then if you already started some programming before 15, then you could be proficient by 18. And there is no limitation on the amount of O level and A level subjects that you can do. And these are well accredited certs. You can do more to get a general knowledge, in Phys Chem Maths and so on. This could help you later on if you wanted to do a simulation app with actual molecules from the periodic table.

[owen]

I can clearly see that you have thought alot about this but you have to realise that we are constrained by an economy that wants to sell you brand new 2017 cars. Maybe in the future you will be able to change the structure of our education system but as it currently is they just want to put out generalists at the lower levels and specialists are expensive to train (back to the economy again). It might be easier for you to become an award winning dancehall artist.

I have to note though that " there are tons of tutorials on how to do Android development, Unity tool, and how to build games for phones" is a meme that you should investigate more. It is one thing to be lead down a path but its a whole other thing to actually reach the destination to solve a problem. Paper thin and oceans wide. Allot of these so called tutorial videos are bait bootcamps. You should also investigate app culture; who benefits from apps and how useful they actually are to the wider population. Back in 2015 I wrote THE INTERNET IS A TUTORIAL MACHINE

[crosswire]

I can clearly see that you have thought alot about this but you have to realise that we are constrained by an economy that wants to sell you brand new 2017 cars. Maybe in the future you will be able to change the structure of our education system but as it currently is they just want to put out generalists at the lower levels and specialists are expensive to train (back to the economy again). It might be easier for you to become an award winning dancehall artist.

I have to note though that " there are tons of tutorials on how to do Android development, Unity tool, and how to build games for phones" is a meme that you should investigate more. It is one thing to be lead down a path but its a whole other thing to actually reach the destination to solve a problem. Paper thin and oceans wide. Allot of these so called tutorial videos are bait bootcamps. You should also investigate app culture; who benefits from apps and how useful they actually are to the wider population. Back in 2015 I wrote THE INTERNET IS A TUTORIAL MACHINE

I am aware of our current state. I am also quite aware of who will listen and who will not. Maybe in the future, but it must be said/sown today.


Interesting article. I read about the first half until GO but I think I see you perspective.

Well, I have learnt over 50% of my knowledge from the internet. Maybe a lot more.
Sources include:
wikipedia
youtube
.edu sites
pdfs

really, there are too many forms but the trick is not to be tied down to one source

For example, I have MCAD (old), MCSD (newer), SQL2012 certs -> I would go tru the curriculum for the courses and watch YouTube video from multiple user sources so that I could get an understanding. Then read multiple other articles.

and I currently pursuing more courses, and Japanese for my own personal development. For Japanese, I have watch videos from so many users and I have subscribed to a few. I have even joined the web site for one and paid the subscription. It was very good. I learnt a lot, and would recommend it. I also watch a lot of Japanese anime and movies.

I have even bought 2nd hand books from amazon for Maya 4, Unreal 3, Android, and quite a few topics as well. Really. Many of the book came with DVDs with tutorial and lessons. The Maya 4 tutorial videos were awesome (from Sam teach Yourself series). Not all the book were great but that one had great vids. I have not gone tru all the books yet. I also have a project management book I got from Logos ship, and another from amazon.

Maybe I was baited but I am satisfied with what I have learnt and the little that I had paid.

So I would really encourage all to learn from as much source as possible. Furthermore, when you really want to learn you will take in all sources you find. I remember taking in all that was on the blackboard that was left behind by other classes doing Computer Science in my first year, classes that were higher than mine of course, because I visited the lab alot.

I even learnt from TJ

Maybe I phrased learning from tutorials on YouTube loosely. You should also take a look at the number of likes and dislikes on a tutorial. Most importantly, explore different sources.