Serializing or separate tables?

[nayen]

Hi,

I am building a simple browser-based text RPG (Role Playing Game) and my first step is to design the database. I have checked some samples and articles about the subject and I wanted to ask your opinions as well.

Basically, I will have the following tables:

- users
- characters
- items
- skills
etc.

Each character will have some skills and items in the game. I can think of two options for designing the database as below. Can you please tell me which one would you prefer and why?

1. Create separate tables such as character_items, character_skills?

Example:

characters table:
character_id
character_name

items table:
item_id
item_name

character_items table:
character_id
item_id

2. Create a single column for items/skills in the character table and store the info by serializing it.

characters table:
character_id
character_name
character_items
character_skills

[r937]

option 1 is by far the better choice

option 2 violates first normal form and therefore has all the issues involved with that

for example, if you're searching for something specific, you have to deserialize every string, which means that the search will require a table scan, which means your app won't scale -- the query will get slower and slower the more rows there are

[nayen]

option 1 is by far the better choice

option 2 violates first normal form and therefore has all the issues involved with that

for example, if you're searching for something specific, you have to deserialize every string, which means that the search will require a table scan, which means your app won't scale -- the query will get slower and slower the more rows there are

Rudy, thanks for your opinion. After asking the question, I made some more research about normalization and reading your reply, first option makes more sense to me now. Although I am starting very simple, there is no reason why it wouldn't get big in the future if I can create something nice. That's why I am trying to do it right as much as possible from the beginning.

I also read about denormalization and that it might be considered for really big databases. I guess I shouldn't worry about it at this stage. Besides, there is the database sharding topic. I am not sure if a browser-based text RPG will require this but as a programmer you have to consider all possible scenarios right?

Theoretically speaking, let's imagine my game reached 1,000,000 players, each player has 2 characters (average) and each character has 20 items (average) in their inventory. Following the first option, character_items table will have 1,000,000 x 2 x 20 = 40,000,000 entries. To read/write data from the database, I guess I will need a lot of joins because there will be many related tables. In such a case, would you still think that the first option is the way to go? Thanks again.

By the way, don't all searches require a table scan? Or do you mean looking for a key (such as item_id) would be much faster than looking up the content of a cell which contains data such as "item1, item2, item3"

[r937]

To read/write data from the database, I guess I will need a lot of joins because there will be many related tables. In such a case, would you still think that the first option is the way to go? Thanks again.

yes

[nayen]

yes

Thank you again Rudy, I added one last paragraph to my last reply if you missed it.

[r937]

By the way, don't all searches require a table scan?

no, not if the searched column is indexed

[guelphdad]

Or do you mean looking for a key (such as item_id) would be much faster than looking up the content of a cell which contains data such as "item1, item2, item3"

The problem isn't really storing of the data in such a form so much as actually having to work with the data. that is why normalization is the way to go.

For argument sake let's say you have 10 users and each of them have a minimum of 5 but as many as 20 items as you say.
with one row per item per user you would have as many as 200 rows rather than 10. BUT it is far simpler with the normalized table when you need to find out which users have a sword, or how many users are carrying a sword and a rope. Even more important when you have a list of say 5 items that a user must have to fight the "boss" and find out if each of the 10 fighters in the party have those five items.

It becomes more and more complex to do such searches when you have serialized data. And as you say, you might have a million players, so even more so.

[nayen]

The problem isn't really storing of the data in such a form so much as actually having to work with the data. that is why normalization is the way to go.

For argument sake let's say you have 10 users and each of them have a minimum of 5 but as many as 20 items as you say.
with one row per item per user you would have as many as 200 rows rather than 10. BUT it is far simpler with the normalized table when you need to find out which users have a sword, or how many users are carrying a sword and a rope. Even more important when you have a list of say 5 items that a user must have to fight the "boss" and find out if each of the 10 fighters in the party have those five items.

It becomes more and more complex to do such searches when you have serialized data. And as you say, you might have a million players, so even more so.

Your example gives me a better understanding, thank you. I am pretty sure I will have such or even more complex scenarios as I improve the game.