Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[qalnor]

Fun thought:
 Allow database queries that generate reports on 'known data'. Allow players to construct reusable query definitions and bind these to some set of keys to allow players to get quick reports on things that they might find interesting.

Depending on how things are implemented in the game right now this might not be possible without a gigantic overhaul.. But I think that from what I've seen it might be plausible.

(I thought about this because it occurred to me that it would be good to have a 'seen objects list' report that could be generated and ideally interacted with to some extent, but then my data analytics background kicked in.)

[Kyzrati]

Hm, I'm not quite sure what you're referring to here. A reverse keyword search of some sort? Well, that guess was because I first assumed you're talking about the lore, but then it seems maybe you're not. (And you probably know that all discovered lore is collected in one place in the game menu.)

Going off the one example you gave, an interactive "seen objects list" is definitely something I considered early on, it's just really low priority since backtracking isn't very common in Cogmind anyway (it usually gets you killed unless you're really fast), and of course there's already the item labeling which is good for seeing what's in the local vicinity.

What other related concepts are you thinking of with this? (Because I don't quite understand other potential applications based on your description.)

[qalnor]

Well from a high level view -- anything your robot knows and remembers is data in its little databanks.

In its memory banks for an square, item, robot, wall, terminal, device, smelly hobbit, etc the raw data would be something like this:

- The known properties of it, when last seen.
- Coordinates for the last place I detected it?
- Turnstamp for last time I detected it.

I think that pretty much covers it. I might be missing something you'd really need but with robust queries those are the only things you need to associate in other data

for example:
QRYDFN for Where might I productively look for some butter?
SELECT coordinates
FROM items INNER JOIN squares
ON items.coordinates = squares.coordinates
WHERE items.lastseen = squares.lastseen
AND items.name = "BUTTER";



Note: 'last' is semi-ambiguous. What 'last' means depends on whether the robot is able to distinguish between two things that are the same (the game can, but should the robot be able to depends on many factors. Perhaps 'he' has a digital prong that sprays RFIDs on everything 'he' finds, but more likely it's only able to positively* distinguish things based on where they are unless they're transmitting an ID that your robot is scanning.) So for things that he can identify uniquely, 'last' refers to the last time he knew where a particular thing was, whereas if he can't uniquely identify a particular thing 'last' means (more or less) the last time he 'saw' a thing of type at a particular.

(I say 'positively' because if he knows things about the thing he might be able to make an even fuzzier match than 'location' (I mean, if an assault rifle is where you last saw an assault rifle, that doesn't ACTUALLY mean it's the same assault rifle) and for anything that has non-default properties that are visible to the robot this might make for a pretty decent match, although I'd say it's questionable whether the robot should really be considered to 'know' as much about objects as we're presented with in-game, especially about EVERYTHING that he's ever seen. Maybe after the player looks at the object it's not too unreasonable to think that he knows that the hover unit is at 29/30 integrity, but honestly if he can really, really do that it's surprising all the things he can't do. I'm not saying take that data away, but if you implemented what I'm suggesting here I could see an argument for having some corruption in the reports, especially for older data)

[Kyzrati]

That... sounds like a lot of work for something that I can't imagine many people would use. Interesting to think about, though!