Talk:Terrain

From Ring of Brodgar
Wiki-edit-clear.png

Last cleared: 2018-11-18 12:34:02 (UTC)
( See the [History] option/tab for previous cleared talks or topics )
( Use the [+] option/tab to start a new talk topic )

Animals v Terrain (?)

Archived Terrain-v-Creatures Data tables
Archived Terrain-v-Trees Data tables


- Animals (creatures)
- Terrain (assuming being native terrain where they can spawn)

Situation:

  • Different locations(pages) where "Animal v Terrain" data is stored.

Target:

  • Merge the data, and give them a single input/storage location/path.

Logical inputs:

  • Terrain input (template parameter) at animal/creature page. (allows usage of terrain-group (not necessary a good thing))
  • Animal/creature input (template parameter) at terrain page. (terrain-group: if really needed, sure. Seems troublesome at first glance)
    • ... not sure yet, but I figure SMW sub-objects might also be an available option (terrain-group: not sure). Here the same question arises, which type will be the container, and/or which type will be the value data.

Currently not yet sure about what path to implement.

Points of view on this from others appreciated.

PS: Although worked out reasons are not mandatory, they do bare more useful information than just giving ones personal preference. --.MvGulik. 19:08, 13 December 2018 (EST)


When I first started making the terrain pages, I originally dumped all the flora/fauna data into the terrain template, just as a place to hold it.
thinking on the subject, it seems it could really go either way:
update the individual animal/tree pages which push data to the terrain page - lots of small edits across many animal/tree pages
update the terrain pages which push data to the animal/tree pages - fewer individual pages to edit, but with much larger data sets.
you could argue that if a creature/tree is to be removed (which has historically never happened), it'd be much easier to remove the data from one location - the creature - than revising several terrain pages.


As an aside to the terrain topic:
I recall reading some notes you made about the grassland/forest/heath/swamp categories and how they fit into the other 'terrain' data.
In Legacy Haven there were 4 or 5 terrain types, most prominent were grassland, forest, and heath. In the context of hafen, these 'types' dont exist (I remember Loftar making a post about this, and could probably find it if you want). Users have informally categorized the hafen terrains into these categories. It makes sense, many of the terrains share features (you can sprint on 'grassland', you must plow 'forest' tiles before paving), but mechanically these distinctions don't exist. If you feel like keeping these categories, perhaps we could rename them to 'biomes' to make them distinct from terrains.
Just a bit of context to their existence
--Ricky (talk) 19:32, 13 December 2018 (EST)


Ricky>"update the individual animal/tree pages which push data to the terrain page - lots of small edits across many animal/tree pages"
Ricky>"update the terrain pages which push data to the animal/tree pages - fewer individual pages to edit, but with much larger data sets."
True. ( 'edits' in these cases also translate into 'data entry points')

Ricky>"you could argue that if a creature/tree is to be removed (which has historically never happened), it'd be much easier to remove the data from one location - the creature - than revising several terrain pages."
True, although the reversed argument also hold some validity. ie: If a terrain is removed there are less edits if the creature/tree data is stored on the terrain pages.

will give them some more thought. ...

Ricky>"I remember Loftar making a post about this, and could probably find it if you want"
Yes please. I count not find a post from Loftar that seemed to fit.

'biome' versus 'terrain' ... To me there seems to be no clear, strait forward, difference between those two terms. Reading up on some related Wikipedia pages did not help (the terms clearly overlap each other). Limiting the terrain-term for actual ground-tiles (and perhaps elevation characteristics), and Biome for the biological environments in the game might work ... at first glance. ...

--.MvGulik. 12:54, 14 December 2018 (EST)

Well I Thought I had read a post about it. Can't find it currently, however. If I ever do find it again, I'll be sure to link it here. Maybe i'm finally going crazy.
--Ricky (talk) 20:33, 14 December 2018 (EST)


I can't make up my mind on which system to pick as default for location data. Both have pro's and con's that kinda cancel each other out. As such I will go with both. For creature-location data I'll go for storing the creatures at the terrain pages. And for tree-location data I'll go for storing the terrain's on the trees pages. That way it allows for some practical experience on how both systems work. --.MvGulik. 02:56, 17 December 2018 (EST) --.MvGulik. 11:51, 29 December 2018 (EST)


Ricky>"... If you feel like keeping these (terrain group names/) categories, perhaps we could rename them to 'biomes' to make them distinct from terrains."
Aha. So only for the terrain-groups.
... Not sure yet how to implement that in a good way. With that I mean where the term biome would actually be useful to use in the potential data-locations (like: page-name, category-name, property-stuff, etc).
(Nts) if for example used in page-names. like for example for the "swamps" group. "Swamp biome" would be a correct title, while "Swamp biomes" would be ambiguous/suggestive in labelling the different individual swamp-terrains as biomes.
Will ponder about it some more ... but after the terrain data cleanup.
--.MvGulik. 11:51, 29 December 2018 (EST)


Working on the last part. The specific target-page editing.

General data adjustments in play:

  • Group-terrains are replaced with there related terrains.
  • Terrains on related creatures are merged so that those creatures use the same terrains. (ie: chickens[chick, cock, hen], rabbits(excluding Bunny), Wood Grouse, Mallard Duck)

If anyone spotted some stuff I might have missed ... do post.
--.MvGulik. 05:23, 1 January 2019 (EST)


...


Troublesome cases
  • Dryad - All terrains (currently skipped due to missing property - ie: "The Heartlands" text terrain parameter)
    • Temporary fixed by making "The Heartlands" a property. (and a related local terrain group)
      • Excluded terrains from local group:
        Cave, Mine, Deep Water (Dryad not a cave dweller I think, nor a good swimmer),
        Shallow Water, Mountain, Plowed Dirt (Although it might walk along them, I doubt its spawning on them. (easily fixed if such a case is encountered)).

...


Consolidated the terrain-tree links, and related templates --.MvGulik. 02:46, 10 January 2019 (EST)

Edit Animals

Where and how edit now? ~upd: realised that on personal biome pages.

Mole for example never spawn on that list of biomes it has now, and spawn where it not listed.. Firefly on forest and swamps..

I guess you use old data from the Terrain Page - it was just copypaste without actual info(and the more actual info was on personal pages).. and i was unaware when you update this.

P.S. I just had a heart attack when realise how many info now corrupted..

--Kitsuneg (talk) 18:31, 23 January 2019 (EST)


I was kinda afraid of the bad data potential. :(
As the whole merging process was dynamic, corrective changes could have been made at any time. They would have just been picked-up.
The (merged) animal data is now stored on the terrain pages.
--.MvGulik. 19:33, 23 January 2019 (EST)

Leech - order hiccup.

Odd. Leech is resisting being properly placed in the #ask query results (Bog, Fen, Swamp: On this general Terrain page). ie: as intended, and specified, on the related terrain pages.
Fact: Leech was added separately to the already present items.
No effect so far: Refresh(all-4), null-edit(all-4), dummy-edit(swamp page only), hard-purge(all-4).
Might just need some additional time to kick in ... (nts: keep an eye out for similar cases)
--.MvGulik. 00:46, 12 January 2019 (EST)

Some additional elements where added at the source page, which also where placed (in the related property field), at the end of the already present elements. (with the Leech element maintaining its position.
Ergo: Property elements/(data-values) are (at least in this case) saved and ordered based on there adding-order (kinda unexpected).
(nts: some other page for which some property-data assignments where ditched. Are still showing as active. (only has one real edit that cleared the page (and property assignments)
Exception to the rule ... SMW Special:PageProperty/fen::terrain_fauna page is listing data-values in alphabetical order.
... move subject to local-space (later) --.MvGulik. 13:08, 24 January 2019 (EST)
Clearing and refilling the (target property) data on a parameter line (while keeping empty parameter case (might or might not matter though)) did reset the displayed values order on the property page (which also changed the display order on #ask result). (not preferred as used order fixup, but at least its available)
Initiating "Data rebuild" SMW Administrative function to see if that might work. (fen and swamp cases) ...
Result: No effect on stored-order of elements.
--.MvGulik. 08:14, 25 January 2019 (EST)