subtype doubt

[Shivam Tripathi]

Q. Gender can be seen as a subtype of which of the following basic datatypes?

    Boolean

    Integer

    Character

Accepted Answers:

Character

My doubt: - Why gender can't be a subtype of a Boolean datatype?

[Hardik Minocha]

Because you simply cannot have gender as "True" or "False", in a commonsensical way. However, if you choose to treat the output of a Boolean status
indicator as Male or Female, it can work. 

I think this question was meant in a normal sense, than a programmer's approach.

[Ganesh P]

Hardik has given a valid explanation.

I'd like to add to this:

Boolean datatype is one of the oldest/simplest datatype in computer systems because they can be easily represented by the presence or absence of current flow. However, it doesn't seem correct to think of male gender as the absence of female gender or vice versa. Moreover, the representation of gender in the dataset given to us is with the characters (alphabets) "M" and "F". If you write them fully as "male" and "female", then they would be word (string) datatype.

Regards,

Ganesh P

[Dev Raj]

I think this post raises a great question about choosing the appropriate datatypes.

You might think it would easy to choose a certain data type for a certain variable but there are lot of things that can possibly influence this decisions.

Let me start off by saying , Under the right circumstances you can use any of the data types for gender. But there are certain problems with each case. 

( Do note : I may use extreme cases to illustrate certain points)

Let us consider each case one by one 

1. Boolean

Now of course since in this classification of gender, there are only two categories male and female so we can store it as boolean variable b

and say that a true in b  means male and a false value means female (or vice versa ) (you can think of this like b being the answer or value of the statement 

" You are a male ")

Now the advantage of this is that it is very storage efficient i.e it takes only one bit ( as usually truth values are stored as 1 and false as 0 in memory)

but the problem is that it is non-intuitive ( it raises the question why should true values refer to males or why should it refer to female ) 

and it is also non-expandable ( like you can have only two different values ) so you cant store a new value of gender like "other"

2. Integer 

similarly to boolean you can say that when you store a certain integer  ( 0 for example ) it is male and for another certain integer ( 1 for example ) it is female. 

This has problems similar to that of boolean also. It is non-intuitive which number should be male and which should be female and although it is expandable ( you can include more genders by using the other numbers  2,3,4 etc ) it usually takes up more space than a character data type, 

and its great expandability isn't that useful ( it will be unreasonable to think of having 100 different genders )

3. Character 

Now i feel this is the most interesting one because it is the only one whose stored value is intuitive ( like you can expect a reasonable person to figure out that a stored value 'M' would represent Males and 'F' would mean it is Female)

Further more since you can store more than 2 different  values in character , it is also expandable . But one disadvantage is that it will take more space than boolean. But we choose this because that intuition is important. Now you could say that for greatest intuition you could directly store the value as a string 

like "Male" or "Female" but that doesnt really increase the intuitiveness or simplicity of the code that much and it greatly decreases the storage efficiency.

Good code/ Program would be used for years to come and it would most need a few modifications here and there ( like maybe the addition of an 'Other' option as 'O' ) and often the person modifying the code wont be the same as the one who wrote it. So for him/her it will be hard to figure out the stuff that is non-intuitive / unreadable . Good code is ( relatively )  easy to read and comprehend and hence modifying it wouldn't be much of a hassle. Your code wont be useful for long even if it is very efficient but very hard to read.

and usually the priority of a programmer (in expressing computation )is as follows :

1. Correctness ( of out put and program)

2. Simplicity

3. Efficiency

While coding even though there are lots of possibilities, we stick to a certain convention while writing to code to avoid confusions and misunderstandings,

 

Another factor is the storage as i mentioned , the difference in used storage would not be apparent to us, and often times it does not make a difference in

our normal programs. But real world programs would often have to deal with hundreds of thousands of records / variables like these .So , the storage efficiency would make a great difference there . ( for example in C++ the programming language I'm most familiar with, the size of bool is 1 bit and the size of character data type is 8 bits or 1 byte and that of integer is 4 bytes or 32 bits  by default.   1 byte = 8 bits )

 

Note : I don't have much training or practice related to real world programming , these are my Ideals and thoughts that are guided by the programming principles and ideals mentioned in " Programming: Principles and Practices Using C++" By Bjarne Stroustrupp , the creator of C++ [ second edition]

Any advice/correction regarding my thoughts / ideas are gladly welcomed

Thank You,

-Dev Raj R

On Tuesday, October 6, 2020 at 10:12:40 PM UTC+5:30 shivamtri...@gmail.com wrote:

[Computational Thinking Support 2]

First of all, I would like to thank Hardik, Ganesh and Dev for their brilliant answers. Especially Dev, thank you for your great answer. You have carefully gone through each case and considered the pros and cons in a balanced manner. I really liked the idea of "expandability", assuming some such word exists. Your explanation is so comprehensive that I have nothing more to add. Thanks for giving us this wonderful perspective on this problem.

I would please request you all to chip in and answer as many questions as you can.

[Ganesh P]

Excellent explanation by Mr. Devraj.

His in-depth analysis of an apparently simple question is wonderful to read. Efficiency, expandability, intuitiveness and therefore maintainability!

I especially liked that intuitiveness has been addressed in great detail.

As they say, "great code is not something that just machines can understand, it's something that humans can understand!".

Thank you for your time and effort!

Regards,

Ganesh P.

Get Outlook for Android

---

From: Computational Thinking Support 2 <ct-su...@onlinedegree.iitm.ac.in>
Sent: Friday, October 9, 2020 10:53:18 PM
To: Discussion forum for Computational Thinking <cs1001-...@nptel.iitm.ac.in>
Cc: Dev Raj <dev.ra...@gmail.com>; shivamtri...@gmail.com <shivamtri...@gmail.com>
Subject: Re: subtype doubt

 

--
You received this message because you are subscribed to a topic in the Google Groups "Discussion forum for Computational Thinking" group.
To unsubscribe from this topic, visit https://groups.google.com/a/nptel.iitm.ac.in/d/topic/cs1001-discuss/1YmSCzUQIh0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cs1001-discus...@nptel.iitm.ac.in.
To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/cs1001-discuss/c48c3a65-1cc5-4d28-af24-a639af909991n%40nptel.iitm.ac.in.

[Harikumar P]

One doubt arise after reading this.If we provide three Gender M,F,and O.then how the Result type is always Boolean.