User talk:Nescientist/Project Trace

Active discussions

Template format

A couple issues that arose previously with {{User:TehPerson/learnlist/prevgen8}} that I realized no one ever thought to communicate:

  • The Generation columns are bolded twice over. Once with the header ! and once with '''.
  • The links to the past learnset subsections are probably not needed. They could even be confusing if a Pokemon learns a move in more than one way. If people can just get to the page, they can search it themselves.
  • Also, the sorting has a couple issues. First off, the numbers just don't sort right. Second, for Pokemon like the Alolan examples, it doesn't work at all. This is probably due to hiding some of the Generation columns, a problem I've had in the past. (I believe the page just thinks the sort button for the second column applies to the second column it has data for, even if it's hidden.) That's one reason it's better to just "erase" the column/cell entirely if you can.

There was also a suggestion that the Generation columns could just be turned into checkmarks ({{yes}}). The column "header" could link to the page instead.

Also, I'll note that #ifexpr is generally a parser we want to avoid using a lot. It'd probably be fine for the past pages, but on the main species page, it could potentially be a problem.

I'm also curious: when you're doing Gen III transfers, how are you going to deal with the fact that individual moves don't actually have a physical/special attribute? Just use the one appropriate to the type and call it good?

Tiddlywinks (talk) 00:34, 8 July 2020 (UTC)

I know the questions were not directed to me, but I think I can add a thing or two. For generations I to III, I believe the idea is to keep the pattern we already have at the learnset subpages for these generations, without putting the damage category. We also would only have to consider this when transferring from gen I to II or from gen II to I, as you can never transfer to gen III.
About checkmarks vs generation numbers, I have a preference for the generation numbers. They are at least just as easy to read, and they follow the same idea we have for listing game abbreviations in move tutor learnsets. And personally, I also think they look more interesting than plain checkmarks, but that's just me.
Another unrelated thing: we should probably have a small list of related talk page sections and user subpages linked in the project page. Suic (talk) 01:27, 8 July 2020 (UTC)
Oh, right, I was just trying to figure out where it might be an issue (thinking that maybe Gen I/II wasn't applicable since you could trade back and forth) and I completely forgot Gen III just wasn't connected.
That's a good point about the tutor templates on the past pages, I didn't catch that. Tiddlywinks (talk) 01:43, 8 July 2020 (UTC)
The project page is just pure basics at this point. I'll be updating it (basically to a more useful v1.0) soon, and I'll take care that includes referencing. Nescientist (talk) 18:12, 8 July 2020 (UTC)
The project page is now updated, and we should keep it updated.
Who's working on the other points? Can you please agree/decide on which templates are being worked on, and add them to the WIP section on the project page? Nescientist (talk) 09:33, 11 July 2020 (UTC)
Due to a lack of response, I have picked TehPerson's templates as the templates to be completed. Nescientist (talk) 11:56, 1 August 2020 (UTC)
If and when I decide to finally get it done, I'll have the stuff generated based on the parameters of my template. I know there are some glitches like the sorting and stuff, but that should not affect how the template's parameters are used. If someone wants to help fix that sorting, that would be much appreciated.
As for the point about linking to the individual sections, I prefer it that way because it helps to navigate more conveniently. I know there is often overlap, but generally follows a precedence that I have already written into the script, in order of accessibility: Level-Up > TM > Egg Moves > Tutor > Prior Evolution. If it can be learned via level-up, then it generally follows that it can learn via TM if one exists, etc. Tutor is near last simply because they're usually introduced in later games. There are a few exceptions like Feebas learning Light Screen by Egg Move but not TM in Generation III, but I consider those to be oversights that eventually got corrected later on- in later generations, they even make TM and Egg Moves mutually exclusive. I'm sure readers are well aware of the many overlaps, so I don't think including the subsection links would really be too confusing so long as the precedents are consistent. TehPerson (talk) 05:37, 13 August 2020 (UTC)
The section links are just an additional pain for anyone trying to update things in the future. Much simpler to just make it a binary "is it in this gen or not". I think it's a little simpler for a reader to search a page than for a "developer" to maintain the exact information to include the section link (i.e., doable but complicated). (And as a casual breeder, I *don't* know about those "overlaps".) Tiddlywinks (talk) 12:53, 13 August 2020 (UTC)
You don't have to get anything done, because I'm already generating based on Bulbapedia data (see below), so no worries.
I can try to modify the template so that sorting works. I'll also test with a different color for event moves (which logically has lower precedence than "Prior Evolution"; also see the section right below).
As for the individual sections, I generally like TehPerson's idea. However, there's what Tiddlywinks said, and also there are many caveats with section headers/anchors (multiple forms, different names for event move headers, etc.), so let's just go with the binary ("isn't needed, pages aren't that long anyway"). I will make sure to at least mention the method (the first one in that order) in comments while generating, we can either keep those or remove them afterwards. Nescientist (talk) 18:18, 13 August 2020 (UTC)

(resetting indent)Done, sorting works fine for me, section links removed, and color shades implemented (also see section right below). Feedback welcome. Nescientist (talk) 19:37, 13 August 2020 (UTC)

I believe the headers should display all generations the Pokémon appears in (as other learnlist templates do), not only past ones.
If somebody could create the missing templates, I could then publish all relevant test cases. Nescientist (talk) 18:41, 19 August 2020 (UTC)

Event moves

Between Suic and myself, we've had some ideas what to do with special/event moves (see here). We basically believe that those should be differentiated, most specifically that

  1. Non-transferable moves shouldn't be listed at all (see WIP on these below)
  2. Moves only available by events should probably be listed as tranferable, but should indicate as such. This is because it's gone for good if you don't have access to the event-only move; my exact wording/condition/definition was "Can I get it with brand-new copies (or would I need to have done/received something in the past*)?"
  3. If a move is available by events in only some generations (such as we're in GenVIII, the moves is event-exclusive in GenVII but TM in GenVI), something should be done about it (waiting for ideas/discussion for these two; it might depend on how common that is)
  4. Being event-exclusive in a generation should not exclude that move from that generation's transfer-only list

Are there objections? Is there anything missing? Are there any good/better ideas on how exactly to deal with them? Nescientist (talk) 09:33, 11 July 2020 (UTC)

I'm expecting discussion here on how to deal with event-exclusive moves. Nescientist (talk) 11:56, 1 August 2020 (UTC)
If you keep the generation numbers in the column, you'll probably have to have a subsection for event-exclusive moves. If you go with {{yes}}/no, you could easily use an asterisk instead.
There's nothing special to "do" about 'interleaved' event-exclusivity. The generations where it's available "brand-new" go in the main table, and the generations where it's event only go in the subtable (or {{yes}}/asterisk if you're going with that). Tiddlywinks (talk) 02:08, 2 August 2020 (UTC)
I was hoping that there was something color-related with generation numbers in these cases, maybe light colors, and possible an explanation in the footer? Not sure how that would look, though.
Another thing that popped up: what to do with moves that are exclusive to pre-evolutions in another generation, for example Venusaur's Celebrate? I'm not sure how many cases these are, and if there's a case that's not event-exclusive, but I'm leaning towards including that as if it wasn't a prevo move (so Venusaur would display an event-exclusive Celebrate). Are there objections/better ideas? It's something that is hard to automate. Nescientist (talk) 09:46, 2 August 2020 (UTC)
I've added (partly) event-only moves for Bulbasaur, and made the template differentiate color-wise between event and anything else (also see section above). Nescientist (talk) 19:37, 13 August 2020 (UTC)
I believe prevo moves have to be checked manually whether they're event-only (search for "prevo" in the current code). Nescientist (talk) 21:28, 20 August 2020 (UTC)

Result data

I think the initial movelist at TehPerson's userpages (linked on the project page) should be improved in the following ways:

  1. It currently includes data from a public database; it should contain more comprehensive data
  2. It should incorporate whatever the result of the above section is and use the finalized template
  3. It shoudln't bold/italicize moves that deal direct damage and Hidden Power
  4. It incorrectly lists moves that aren't transferable, we'd need to research what exactly those are (see below)
  5. They're not complete, obviously

I'm working on these (for the first one, I'll try to pull Bulbapedia data), and plan to update the userpages accordingly. Automating what needs to be bolded will probably be easier than italicizing, which might have to be done manually. Or maybe I'll need a list of future move types per species or something!?

Are there objections? Is this missing something? Nescientist (talk) 09:33, 11 July 2020 (UTC)

Is there a defining feature of the moves that aren't affected by STAB? I already have lists of all the Pokemon learnset pages by type, it's just a matter of defining a parameter that can easily check to see if STAB is triggered. So for example I can say all Water moves learned by Goldeen get STAB, it's just a matter of needing to know what moves attract STAB I think, like do all moves with a listed power attract STAB? I can easily create stuff to exclude the direct damage moves too. --Spriteit (talk) 08:33, 29 July 2020 (UTC)
I don't think so, the best "algorithm" I can think of is any special or physical move that matches the Pokémon's type(s), except those in the category (or subcategories) and Hidden Power. (Return, for example, is mdash but STAB.)
As for Pokémon types, you can easily just get them from existing learnlist header/footer template params on the page at hand anyway, so I never considered that to be a problem. The hard part is getting evo's types for the italics, maybe the easiest way is to do that manually. Nescientist (talk) 16:31, 29 July 2020 (UTC)
I intend to publish data based on Bulbapedia data within the next couple of days. (The whole project should be done before the end of this month.) It would be great if the section right above was solved by then.
TehPerson, is it okay to override (and extend) your userpages, or would you prefer if I started something of my own that we'll be using? Nescientist (talk) 11:56, 1 August 2020 (UTC)
FYI, I have given the STAB task 85%, which means that I can automatically bold things that need bolding. Italics have to be done (semi-)manually once we have all the data. Nescientist (talk) 17:48, 12 August 2020 (UTC)
Actually, if there was a list of future STAB types (such as "Charmander: Flying; Charmeleon: Flying; ... Cubone: Fire,Ghost; ..."), I could use that to automate, so if someone has the desire to make that (and/or the desire to help me), that'd be great. Nescientist (talk) 19:37, 13 August 2020 (UTC)
Created the list myself (see here), so italics done as well. I've given this 99% coz I'm a coward and afraid of bugs/oversights. Nescientist (talk) 11:15, 16 August 2020 (UTC)

(resetting indent)Natural Gift doesn't get STAB either.

For those generations where templates are ready (VI and VII), I have published the exact generated data (links see project page) of around 5 interesting species per generation. (Note that the template include size would start being a problem at some point.). Let me know where the result data should look different (bugs, params including the comments, etc.), then I can try to change that (except for form names and prevo moves, also see above).

Also, I'm not sure if we should really create a bazillion of user pages with all data, or if I should look into adding them directly onto mainspace pages when we're ready. Any thoughts? Nescientist (talk) 21:28, 20 August 2020 (UTC)

Let me reiterate: As of 10 days ago, I'm finished with all the data, and how to present it. If something could be better, we need to discuss (then make it better). If it's alright, we need permission (and a plan) to get started. Nescientist (talk) 09:05, 30 August 2020 (UTC)
I would say, if you have your final data now, make a good handful or two of examples covering different usages of the templates as much as possible, so that we can give it one final check. Tiddlywinks (talk) 12:30, 30 August 2020 (UTC)
For those generations where templates are ready (VI and VII), I have published the exact generated data (links see project page) of around 5 interesting species per generation. That's what I would go with (and would request a final check for), except for form names (so "Alolan Rattata" instead of "Form 2" etc.), and prevo moves (which might be event-exclusive, but would need to be checked manually).
I could publish data for other generations, but I think it doesn't make sense before the templates are made. If you want specific other examples, I can also do that. Nescientist (talk) 14:06, 30 August 2020 (UTC)


I think we should really mainspace User:Tiddlywinks#Transfer.

As can be seen on its talk page, there's not much that's missing before mainspacing. I believe we should mainspace this now; it'll make a great link target from the transfer-only section's header, for example. Also, as mentioned below, it could be used to display what cannot be transfered. Who's willing to get this done? Nescientist (talk) 09:33, 11 July 2020 (UTC)

Who's willing to get this done? Nescientist (talk) 11:56, 1 August 2020 (UTC)

Non-transferable moves

Someone may have to do some research here, but I believe those include at least

  1. HM moves (possibly not all of them, and not in all generations?)
  2. For Generation I (Time Capsule two-way transfer), any move introduced in Generation II
  3. Possibly some field moves in some generations
  4. Event-only moves learned in Generation I or II for later generations (because GB cartridge Pokémon cannot be transfered to GenIII+ games)

An exhaustive list/table would be great. I'm not sure, maybe this would go to the transfer page as well, possibly together with species that cannot be transferred? Nescientist (talk) 09:33, 11 July 2020 (UTC)

Having such a list would be very useful. I think I started looking into this a while ago, I'll see if I can help to get this going. But if the list turns out to be too large, maybe we should include only the general rules and the special cases in the transfer page, instead of the whole list. Let's see.
Something else I just thought about is how we should deal with moves that cannot be used in Generation VIII. Should we even include them as transfer-only moves? If so, we should definitely have them marked as unusable. I'm also not sure if these moves even have power, PP and accuracy listed in Sword and Shield, but if they don't it would probably be better to list them as N/A rather than the values from generation VII (assuming we decide to include them at all). Suic (talk) 22:14, 11 July 2020 (UTC)
I believe they do have values assigned, yes.
I think I'd just list them anyway. Also, there are now updates/DLCs that may make them available/usable in the future; I think that was the case for Kinesis, for example. (I haven't looked into it in much detail, so I'd be interested in SnorlaxMonster's opinion, or that of anyone else who has.) Nescientist (talk) 19:01, 14 July 2020 (UTC)
I'm at a point where I would need those, if you've done it, Suic.
As for the unavailable moves (which only applies to Generation VIII), I really think we should list them anyway, so if that shouldn't be done, please stop me now. Maybe there's even a way to indicate those (asterisk or something), but it would come with the downside of having to continuously update the section should more moves be made usable, so I'm not sure. Nescientist (talk) 17:47, 12 August 2020 (UTC)
The converse of that downside means you don't "want" to update the data at all. If you don't want to update at all, then either you're only gonna have a snapshot of the transferability for this update or maybe you just decide to note transferability for the base game (which actually has some merit given the whole "out of the box" mentality).
But if those options aren't what you're aiming for, then very simply you have to update it somehow no matter what. (In short, it's not a downside, just a cost of "doing business".) That might mean changing asterisks or what-have-you, or it might mean adding whole rows (which, maybe if you do it smart, you included but just commented out). Tiddlywinks (talk) 22:52, 12 August 2020 (UTC)
Take Bulbasaur's Captivate, for example, a move that can be successfully transfered and known but not currently used. There are three options:
  1. Include it anyway, much like its Bind and Confide and all the others
  2. Include it but mark it as special (like an asterisk screaming "..but cannot be used")
  3. Don't include it (this is what TehPerson has done; I could certainly upgrade it to "include it but commented out")
Option 1 was my personal favorite (at least before people like Spriteit below shared their opinion), and Options 2 and 3 you'd have to update (and ideally you should probably update with a note "wasn't usable until SwSh Version 2.3"). Your base game idea is just a snapshot of the past, and while it's something you needn't update, I agree that other than that it's bascially just a silly idea.
I can do whatever you want me to do, and I'm not against updating, but as long as you can still get the move, see its stats and all, I'm personally not a big fan of Option 3. Nescientist (talk) 23:33, 12 August 2020 (UTC)
As for non-transferable moves (the list at the top of this section), field moves aren't mentioned at articles (only HMs), so item 3 ignored. Items 2 and 4 considered.
For item 1, a confirmation here would be nice, other than that you apparently cannot transfer forward any move that is HM in all games of Generation III or IV, most specifically
  • Cut, Fly, Surf, Strength, Flash, Rock Smash, Waterfall from Generation III
  • Cut, Fly, Surf, Strength, Rock Smash, Waterfall, Rock Climb from Generation IV
These should be the only exceptions. Nescientist (talk) 12:33, 16 August 2020 (UTC)

Regarding this topic, I've added a "x" field to the Gen VIII template to help indicate unusable transfer-only moves. I've only applied the change to Bulbasaur in the working sandbox, so please see if you like how it looks. TehPerson (talk) 00:48, 12 October 2020 (UTC)

Well, the idea comes a bit late tbh, but I certainly like it; it's simple and serves a purpose. Assuming this doesn't affect the template's approval, I'm gonna try to change the data accordingly. Nescientist (talk) 19:18, 12 October 2020 (UTC)
In case that makes things easier, it's just a simple find-replace operation of }}<!--unusable--> to |x=yes}}. I recall the original idea was to comment them out completely, but I also agree including them is relevant since they still exist in Gen VIII, to a limited capacity. The ×'s convention comes from Template:anmov's way of highlighting impossible moves, while also not confusingly coinciding with legends used in other move lists (such as † and ‡). TehPerson (talk) 19:49, 12 October 2020 (UTC)

I can actually get all this done in a day, but firstly...

  • I need to make sure my cross-generational data is complete and accurate.
  • Need a list of edge cases (e.g. which gens do not allow HM transfers)
  • Regarding moves that have been removed in Generation VIII -- should I have them totally excluded or simply commented out (e.g. with the <!-- --> tags around them), as we saw IoA re-introduced some moves, though most were just signature moves of the re-added Pokémon.

I already have all the learnsets in my data to do all the cross-referencing, but it may be missing event-exclusive moves, which can probably just be manually added in. I've already ran my script for the Gen VII and VIII sets, all it takes is for me to re-run them with the other generations, with some minor tweaks. I guess all of this has already been covered above, just wanted to give an update on my end. (Honestly been pre-occupied with other stuff and forgot about this for a month.) TehPerson (talk) 04:37, 28 July 2020 (UTC)

Personally, I'm not a big fan of manually adding in event moves, since that's obviously error-prone, and also requires people to actually do it. As I've said above, I'm automatically grabbing Bulbapedia data anyway.
I also think we're not in need of checking your spreadsheet, my userpage(s) and template(s) already got that covered. (But just FYI, if you desperately want a spreadsheet, Spriteit had done something similar, and that we checked.)
My best idea is still to coordinate. Please have a look at the above sections (maybe particularly the first one at this time). Those are the bottleneck, I believe the rest, as you correctly point out, really can be done "in a day", basically. Nescientist (talk) 18:16, 28 July 2020 (UTC)
I'd suggest commenting out the ones that are missing from Gen VIII, since we don't know what will come back and what won't for sure till the games are fully out. --Spriteit (talk) 08:33, 29 July 2020 (UTC)
Just for my understanding: you can actually get them, see all the assigned power values etc., they just don't work, right? Because that's why I argued in the section right above that I'd just list them anyway. (If there's a majority for hiding them regardless, that's also fine to me.) Nescientist (talk) 16:36, 29 July 2020 (UTC)
Yes, every move has assigned Power, Accuracy, and other effects. But they all have a "usable" tag that, if not usable, is set to 0 (False). Kinesis was previously had the "this move should be forgotten" description, but was changed with IoA. Also thanks for that sheet- it should save me a load of work, once I find the time to do it. TehPerson (talk) 17:00, 29 July 2020 (UTC)


You tell me everything is basically ready. So after reviewing everything, you're all good to move forward with adding transfer moves as you are able. Cheers. Tiddlywinks (talk) 12:56, 7 October 2020 (UTC)

I'm currently sorting out some remaining details, and when I'm done (within the next few days), I plan to do a select few pages myself, and have BulbaBot do the others. Nescientist (talk) 19:18, 12 October 2020 (UTC)
Okay, finished. I will now start to do some pages myself (might take some days, but trying to finish asap), and BulbaBot will do the others soonish. FYI, neither of us will do Generation VIII learnsets for Pokémon set to appear in The Crown Tundra (The Crown Tundra#Returning Pokémon) at this time, we'll wait with those until we know the moves they'll learn. Nescientist (talk) 23:42, 17 October 2020 (UTC)
Just as an update: As you might have seen, BulbaBot has finished adding the data onto pages, including the Crown Tundra sets that have meanwhile become available. Users already discovered and fixed a few minor bugs (if there's others, please report). Unfortunately, BulbaBot also lost all hidden comments along the way, but that is going to be rectified (but since it's only comments, that should not affect the looks of pages). What's still missing is the remaining species that I deemed easier to do manually (i.e. Lycanroc, Meowstic, and Deoxys); I hope to get them done by the end of the weekend hopefully. Nescientist (talk) 19:15, 30 October 2020 (UTC)
And as of now, all this should be complete. Nescientist (talk) 16:29, 1 November 2020 (UTC)
Return to the user page of "Nescientist/Project Trace".