Proposed change to !CATEGORY

[Orion Pobursky]

What are the thoughts on abandoning the first word convention for !CATEGORY and have every part have an explicit !CATEGORY statement?

[Travis Cobbs]

What are the thoughts on abandoning the first word convention for !CATEGORY and have every part have an explicit !CATEGORY statement?

Is this because you expect parts to stop putting the category at the beginning their name, or just to make it more obvious what the category is?

[Willy Tschager]

What are the thoughts on abandoning the first word convention for !CATEGORY and have every part have an explicit !CATEGORY statement?

I thought we had that policy already in place?

w.

[Orion Pobursky]

Is this because you expect parts to stop putting the category at the beginning their name, or just to make it more obvious what the category is?

The latter.

[Manfred Schaefer]

In my view, this would make sense, as it would allow the parts to be grouped according to their use. It might even be possible to have more than one group.

Best regards,

Manfred

[N. W. Perry]

Are there examples of where the first-word convention is providing misleading results?

[Orion Pobursky]

Are there examples of where the first-word convention is providing misleading results?

There isn't any I'm aware of. It would just make things more explicit. It would also avoid new author confusion since the two word categories require a statement even if those words are first in the description.

Edit: I thought of one case: Minifig. Since there are two word categories that contain "Minifig" and Minifig by itself is a category, that can cause misfiling. It's why the Minifig category warning exists. We could solve this by depreciating "Minifig" in favor of, say, "Minifig Body".

[N. W. Perry]

There isn't any I'm aware of. It would just make things more explicit. It would also avoid new author confusion since the two word categories require a statement even if those words are first in the description.

Edit: I thought of one case: Minifig. Since there are two word categories that contain "Minifig" and Minifig by itself is a category, that can cause misfiling. It's why the Minifig category warning exists. We could solve this by depreciating "Minifig" in favor of, say, "Minifig Body".

Hmm, and probably the same with "Constraction", "Figure" and "Sticker".

I do like the first word convention, it seems to help impose some orderliness in the naming of parts. I don't mind requiring a !CATEGORY statement, either, but that statement could be taken automatically from the first word if it matches an existing category.

What happens now if someone submits a part with no statement, and the first word doesn't match any established category?

[Orion Pobursky]

What happens now if someone submits a part with no statement, and the first word doesn't match any established category?

It throws an invalid category error. 

And the PT can absolutely add the statement if it's missing and the first word is a valid category.

[Travis Cobbs]

The latter.

Based on what you have written here, I don't have a problem with making !CATEGORY mandatory.

[Orion Pobursky]

I think I'm gonna do a soft roll out where the PT will automatically add the meta to all unofficial parts.

[Chris Böhnke]

There isn't any I'm aware of. It would just make things more explicit. It would also avoid new author confusion since the two word categories require a statement even if those words are first in the description.

Edit: I thought of one case: Minifig. Since there are two word categories that contain "Minifig" and Minifig by itself is a category, that can cause misfiling. It's why the Minifig category warning exists. We could solve this by depreciating "Minifig" in favor of, say, "Minifig Body".

I would like that.
Might also make some submission warnings less confusing.

[Roland Melkert]

I think I'm gonna do a soft roll out where the PT will automatically add the meta to all unofficial parts.

I don't see a problem with adding the meta to all library parts. 

But adding them automatically might be a mistake. 

Like mentioned above it could cause misfiling.

Also I would prefer to keep the "fallback to the first word" rule in the spec (for 3rd party/custom content etc).

[Orion Pobursky]

I don't want to change the category spec. I only intend to change the more restrictive library spec. Implementation will be phased starting with all unofficial files