Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add lageanes dialects #74

Merged
merged 1 commit into from
Feb 27, 2019
Merged

Add lageanes dialects #74

merged 1 commit into from
Feb 27, 2019

Conversation

mauscoelho
Copy link
Contributor

Description

This PR adds the dialects that come from the city of Lages and Amures region

@vercel
Copy link

vercel bot commented Feb 25, 2019

This pull request is automatically deployed with Now.
To access deployments, click Details below or on the icon next to each push.

Copy link
Owner

@mvfsillva mvfsillva left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hey, @mauscoelho thanks for contributing!!

I think PR It fits in the state of Santa Catarina because Lages is a municipality.

What do you think?

@mauscoelho
Copy link
Contributor Author

So, that's a really good topic 😂 There are some types of dialects in Santa Catarina and they are really different. One of them is the catarina and is common on the coast of Santa Catarina, but it is only a small part of the state.

As you can see on the image below, the lageanes represents the red part on the map:

screenshot 2019-02-26 at 10 34 43

But, is up to you to. If you prefer I merge it as a general dialect of the state of Santa Catarina

@hugoestevam
Copy link

Hi @mvfsillva, it's a great project, congrats!
In short, instead of dividing the state of Santa Catarina in two parts, people should bear in mind that although Santa Catarina is one of the smallest Brazilian states, it is the most culturally diverse. I wasn’t aware about the cultural shocks I would get by just crossing those different regions. It is crazy, but it is real.
I want to demonstrate by a qualitative dialectical map below.

mapa-dialetal-v5

I think the better approach is they cannot be considered state dialects, once there usually is more than one dialect per state. (Only a suggest)

@mvfsillva
Copy link
Owner

mvfsillva commented Feb 26, 2019

Guys this is amazing!

That is the idea of this project, the Brazilian cultural wealth is incredible, I did not imagine these divisions, let's improve!

Let's add a new node in this json, to facilitate the identify the state, what do you think @mauscoelho?

cc @anfsantana @RodolfoSilva

@mauscoelho
Copy link
Contributor Author

mauscoelho commented Feb 26, 2019

Yes, for sure!
Do you expect something like that?

{
    ...
    "state": "Santa Catarina"
}

@RodolfoSilva
Copy link
Contributor

Hey @mauscoelho,

Thanks for contributing!!

I think you can use this format:

{
  ...
  “commonIn”: [“Lages”]
}

I believe this is the better way to limit a dialect by zone.

@mauscoelho
Copy link
Contributor Author

@RodolfoSilva Lages is the city if we follow this pattern I should add 23 cities on the node 🤔 according to wikipedia.

I think the better way is to group by state :) what do you think @mvfsillva ?

@hugoestevam
Copy link

Hi guys

I think this PR should propose a major change in the file structure of dialects because we understand that the dialect can cross the state boundaries.

For example, "Gringonês" can be seen in Xanxerê (SC) as well as in Erechim (RS), as well as "Lageanês" can be seen in Lages (SC) and in Vacaria (RS).

Therefore, I recommend the following structure:
lageanes.json

{
	name: "Lageanês",
	description: "Dialeto utilizado na região Serra de Santa Catarina e Rio Grande do Sul",
	crossingStates: ["SC", "RS"]
	dialects: [
	  {
		"dialect": "Aleijada",
		"meanings": ["Mulher de bons atributos físicos"],
		"examples": ["Uh, Aleijada!"]
	  },
	  {
		"dialect": "É só lavá",
		"meanings": ["O mesmo que, ninguém merece"],
		"examples": ["Olha o que ele fez, é só lavá!"]
	  }
	  ...
	]
}

I know, this has a big impact. 😬

@mvfsillva
Copy link
Owner

@hugoestevam

Hi guys

I think this PR should propose a major change in the file structure of dialects because we understand that the dialect can cross the state boundaries.

For example, "Gringonês" can be seen in Xanxerê (SC) as well as in Erechim (RS), as well as "Lageanês" can be seen in Lages (SC) and in Vacaria (RS).

Therefore, I recommend the following structure:
lageanes.json

{
	name: "Lageanês",
	description: "Dialeto utilizado na região Serra de Santa Catarina e Rio Grande do Sul",
	crossingStates: ["SC", "RS"]
	dialects: [
	  {
		"dialect": "Aleijada",
		"meanings": ["Mulher de bons atributos físicos"],
		"examples": ["Uh, Aleijada!"]
	  },
	  {
		"dialect": "É só lavá",
		"meanings": ["O mesmo que, ninguém merece"],
		"examples": ["Olha o que ele fez, é só lavá!"]
	  }
	  ...
	]
}

I know, this has a big impact. 😬

Could you open an issue so we can discuss it?

@mvfsillva
Copy link
Owner

@RodolfoSilva Lages is the city if we follow this pattern I should add 23 cities on the node 🤔 according to wikipedia.

I think the better way is to group by state :) what do you think @mvfsillva ?

Initially, I think it is best to group by states

@mvfsillva
Copy link
Owner

@mauscoelho

I'm going to merge your pr and open an issue so we can think together about a better solution!

What do you think?

@vercel vercel bot requested a deployment to staging February 27, 2019 03:57 Abandoned
@vercel vercel bot requested a deployment to staging February 27, 2019 03:57 Abandoned
@mvfsillva mvfsillva merged commit 9559908 into mvfsillva:master Feb 27, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants