Updating IFWiki

General IF Community Resources
21

Just to be clear, I think replacing IFDB/CASA with a wiki isn’t even something to wish for. At a minimum, we want a thing where people can post reviews, and search/sort for games with good reviews. No wiki I’m aware of offers functionality like that (and certainly not IFWiki).

The whole idea of wiki software is to collaboratively edit the data, but you absolutely wouldn’t want to allow anyone to edit someone else’s reviews. The whole idea of a review database is contrary to the idea of what wikis are/do.

3 Likes
22

I can’t imagine any of the self-organizing IF communities wanting to agglomerate in order to create an IF knowledge base with the scope of the entire genre.

But there does need to be something like that, because there are social science and media courses encouraging students to try their hand at IF. Whatever that impetus is, it needs to consider resourcing some supporting material.

Apologies if this happens already, should I not be aware of it.

4 Likes
23

I don’t think there’s an easier way. Let me elaborate on Zarf’s suggestion to “set up cross-site indexes.”

If you want to automate copying links from CASA to IFDB, you fundamentally just have to know which pages on CASA correspond to which pages on IFDB. That’s a “cross-site index.”

But how would you know that? It’s plausible that an initial attempt at a cross-site index could be automated by doing an approximate (“fuzzy”) match on titles and authors, but there would be a number of questionable cases and/or missing matches that we’d have to manually fix up. (Not to mention the issue @8bitAG brought up, where sometimes there are multiple IFDB pages for one CASA page, and probably vice versa.)

If someone had that data, or was willing to compile it, it seems plausible that we could store it in IFDB. It could look kinda like the “Baf’s Guide” links in IFDB. Older games that had an ID in Baf’s Guide have a “Baf’s Guide ID” section on the IFDB game details page. https://ifdb.org/viewgame?id=ju778uv5xaswnlpl “Baf’s Guide ID: 255”

IFDB could have a “CASA ID(s)” option, which could have an ID (or more than one ID?) for cross-linking to CASA. Similarly, CASA could list the IFDB ID(s) (TUID[s]) for games it tracks. (Or both! But then both would need to be updated. It’s certain that if the data appeared both in CASA and in IFDB, that the databases would contradict each other some day, and someone would have to sort that out.)

And the task of keeping the ID mapping(s) accurate would never end… as long as new games appear on CASA, their IDs would have to be cross-posted to IFDB and/or vice versa.

Once we could know confidently which CASA pages exist for which IFDB games, a tool could then add IFDB “download” links to CASA solutions and maps. Note that this tool would also need to run forever-ish, updating IFDB as new solutions/maps appeared on CASA. And the tool would need to be able to detect whether a CASA link had already been posted to IFDB, to avoid posting and re-posting the same link multiple times.

As Zarf says, it’s a lot of work.

But, if you’re reading this, maybe you think you could do it more easily? Well, none of this strictly has to be done inside the walls of IFDB, or CASA. Anybody is allowed to post links to CASA on IFDB, so anybody could maintain a mapping of CASA IDs to IFDB IDs on their own computer, in a spreadsheet or whatever, and then post the links yourself.

And you don’t even necessarily have to post the links by hand. The IFDB API allows you to automate posting links.

You can get a full backup of IFDB on IF Archive. I don’t know how you’d get a list of all CASA pages; maybe you’d have to automate crawling CASA…?

Maybe it’s easier than I think? But don’t say I didn’t warn you…

2 Likes
24

To address one or two questions about. The CASA policy is to have a single page/entry for a “game” that includes references to all the various versions and remakes of that adventure. Other sites often have entries for different versions of the same game, e.g. the BASIC original, the TADS remake, a later Inform reworking. So it’s not a one-to-one mapping.

(There are instances when we have multiple entries for games… particularly in complex cases like the many games derived from Colossal Caves; however, if anything we’re moving towards less entries per game rather than more… I’ve been spending a lot of time this year researching the fascinating history of many BASIC text adventures that started out on early platforms, such as the TRS-80, and then ended up with uncredited, and often differently titled ports, on other systems)

I’m still not completely sure why a topic about a completely unrelated site, IFWiki, has morphed into one about “harvesting data” from CASA. I don’t own or speak for that site, but as a user/admin, I’d hope anyone would have a formal conversation with Jacob at CASA before they go about hotlinking CASA content.

Perhaps, a focus on the original topic would be in order? If you wish to discuss the development of CASA then you’re perfectly welcome to come and chat to the team over in the forums there.

7 Likes
25

I’m fairly sure that a MediaWiki website could combine the general information pages of the various existing sites together with a database of the games (editable by all) and reviews (editable by the reviewer or sysops) which appear on those existing sites. Wiki software wouldn’t be the only answer, by any means. But there is such a desire to maintain separate websites that I do think it’s just a pipe dream, and anyway centralisation does come with its own risks.

I’m sorry for my part in derailing this topic early on. To answer the original question:

I’d say yes and no…

Yes – It would be good if it IFWiki were to get an official or at least semi-planned/organised spring clean that we can all join in on. I see that the page for IF 2021 was added today (by @cchennnn I think) so it isn’t in too sad a state.

No – The game-related pages (see the “Works” category) should mostly just be deleted. It would be good if the owner were to issue a diktat that these pages have their content moved to IFDB/CASA as appropriate, and only be saved from deletion if there is anything left that doesn’t belong elsewhere.

I should confess that I’m not a big user of IFWiki or IFDB (though I’ve made minor contributions to both) and hadn’t looked at CASA in any detail until now (I like it).

2 Likes
26

The CASA site has a distinct community, and unless they want the site to close then I don’t see anything ever happening to it. The IFWiki community largely overlaps with the IFDB community, I think. But it’s still independently owned and run. It would be possible to merge it if they wanted to though.

I have thought before that a simple wiki using the Github wiki, and then published on a ifdb.org subdomain could be an option. We could populate it from the IFWiki, excluding game pages.

4 Likes
27

I completely agree with that. Though @Warrigal is the third-highest ranked Dungeon Master there, and said he dreams of the day when the four resources he listed are combined into one!

MediaWiki is such powerful wiki software (and easy to install and use) that it would be a shame to use anything else to create a wiki.

You would run into problems. The copyright situation isn’t entirely clear. And it would be (for want of a better word) a bit rude! It would be better to work all it out with the creator of IFWiki. And it would cause confusion – it would be like trying to find an Inform extension :stuck_out_tongue:

2 Likes
28

I think something that would help here would be de-formalising IFWiki pages and templates. Currently, most game and author pages use a rigid structure that makes them redundant with IFDB – and I believe that largely contributes to so many of them being many, many years out of date, because why bother updating them if IFDB exists? If they read more like Wikipedia pages they’d be more useful.

(But wouldn’t that make them redundant with Wikipedia? No, because the notability/inclusion criteria would be IF- and IF-culture specific.)

6 Likes
29

TLDR: keep IF Wiki; possibly make its focus clearer on the front page and do some (all?) of the things which bg listed above; don’t delete anything.

I’d say that IF Wiki has quite a lot of aspects that are worthwhile and not easily transferable to IFDB or other sites/formats.

1. Some things which it already has and/or could concentrate and expand on:

(partly reiterating what zarf said above)

  • currently running and upcoming competitions and jams

    • would be quite nice especially with the additional comps that have started to come up recently: Adventuron jams, IF Art Show Revival, Ink Jam, ParserComp, PunyJam, TADS Jam… and that’s not even counting French, Spanish, and German comps and the various events on itch.io and Indie Game Jams which could be interesting

  • master list of competitions and info/history about them

  • master list of IF-related websites

    • it’s not just a pile of links (although even if it were, that would still be very useful!), it also has a small bit of info/history about each

  • master list of communities/groups/meetups

  • articles about craft and theory, and collections of links to such articles elsewhere

    • past RAIF topics
    • it would be cool to have a similar, carefully chosen and categorized list of interesting threads from this forum
      (although of course, as usual, it hinges on volunteers who would enjoy collating that)
    • see also, for example, this useful article about Clothing, which includes a verb checklist (“DISROBE, DON/DOFF, REMOVE, …”) and ideas about the role that clothes can play in a game

  • collections of making-of/post-mortem articles, see here and here

  • collection of statistics

  • collection of puzzles and game mechanics

  • overview of and details about interpreters

  • overview of and details about authoring systems

    • (Yes, we have the sticky thread here, but the forum format is not necessarily ideal for this. Quite apart from the issues with the sticky becoming un-stickied as per Discourse’s default setting, which cropped up recently.)

  • overview of tools/utilities

  • tutorials

  • a people category, which offers opportunities to write something about yourself or notable IF-related persons which goes beyond a pure listing of authored games.
    For example, there is information about designers of authoring systems, comp organizers, prolific reviewers, diligent beta-testers.

2. About the game info pages: Keep them.

a) I’m not sure that the existence of the game info pages (whether outdated or not) is so distracting or off-putting that it keeps people from contributing to other areas of the Wiki.
If people want to contribute in the areas mentioned above under 1, they can do so, can’t they?
(And hey, if some also do want to contribute to game pages, why not let them?)
(Agreed though that the focus should probably be clearer.)

b) Deleting, pruning or force-migrating significant amounts of existing user-contributed content from the IF Wiki to elsewhere does not seem to be a good way to foster engagement & identification, IMHO. :wink:

6 Likes
30

I’m not arguing for anything in particular here, just thinking about how to make things work.

If we were to go the route of de-emphasizing game pages on IFWiki, one question becomes, what do you do when a game is mentioned on IFWiki? (For example, on a competition page.) Just make the game’s title plain text, with no link? Put in an internal link, but don’t put a wiki page at the other end? Or link to the game’s IFDB page instead? Then the person editing the wiki would need to look up the url for the game at IFDB each time a game is mentioned, instead of just typing the title in brackets. I wonder if there’s a way to shortcut that. I guess you could have a template that would automatically search for that game on IFDB, but that might be more annoying than helpful.

If some game titles are internal links and some are external links, is that going to confuse people?

I guess we could move in the direction of linking to IFDB once when a group of games is listed. If there’s a list of games authored by a particular person, you could link to a search for that author’s games on IFDB, and don’t link individual game titles. (Or maybe, in that case, you wouldn’t need to list the games at all.) Or if there’s a list of games in a competition, you could link to the competition page on IFDB.

2 Likes
31

When you talk about game and author pages being more like Wikipedia pages, are you suggesting that they would have different kinds of information than they currently do? Or would they have the same types of information, just in a different form? I’m not quite following.

32

I’ve slept on it and realised this too… and the same goes for the author pages. The answer maybe is to have minimalist pages for both. They could be linked together with automatic reciprocal links to help keep them updated. They could have automatic links to IFDB/CASA (search links and/or based on manually-added IDs).

I agree with what I’m sure he’s saying! The current game templates make the game pages look hopeless and too much work to improve. I think by “like Wikipedia” he means more free-form text.

33

IMO, the deepest problem this thread is revealing is how hard it is, culturally speaking, to delete anything, especially from IFWiki, but also from any public IF database.

For example, consider the IFWiki page about interpreters. Do you want to know the best interpreter available for Palm? Pocket PC? Symbian? You wouldn’t dare delete any of that, would you?

Half the interpreters on that list haven’t been updated in years, and are not remotely good choices for playing IF. But who has the temerity to delete an interpreter from this table? Not I.

Or consider @StJohnLimbo’s big list of things IFWiki is (or should be) great at. People often forget, but IFDB has pages for competitions, meetups (“clubs”) and a “download adviser” that’s intended to tell you the best interpreter for your preferred operating system.

The IFDB “competitions” page is pretty good; it was certainly updated faster than IFWiki! (Remember, this thread started when we noticed that nobody’d even remembered to create an IFComp 2021 page until this week.)

By comparison, the IFDB “clubs” page is pretty bad; it contains 13 clubs in all, and still manages to reference a bunch of defunct clubs. But are you sure they’re defunct? You’re not gonna delete a whole meetup, are you?? (IMO, this problem would be just as bad on IFWiki.)

As for IFDB’s download adviser, it’s a mess. It recommends Zoom for macOS, which doesn’t work on modern versions of macOS, and WinGlulxe for Windows, and it lists dozens of operating systems, most of which are no longer available. I’ve promised to work on it soon, but … what is the right interpreter to recommend on Windows 11? (…on ARM?) Could we just delete the download adviser, replacing the whole thing with a link to Lectrote??

As for authoring systems, IFWiki’s “authoring system” list includes a summary at the top:

The most commonly used (and talked about) systems appear to be ADRIFT, Alan, Hugo, Inform 6, Inform 7, Quest, TADS 2, and TADS 3. The newest authoring systems are ThinBASIC Adventure Builder and RAGS (released in 2006), and DreamPath (released in 2007).

This list is archaic. I bet IFDB could do better just by reporting which dev systems had released new games in the last year (or new good games, which we actually know about).

But if we did that, would anyone delete the IFWiki summary of authoring systems? (Or even update it?) I wouldn’t, that’s for sure…

4 Likes
34

I agree with the gist of what you’re saying here. For example, someone took the trouble to cut and paste the 2009 version Ron Newcombe’s Inform 7 for Programmers onto IFWiki. Despite thinking it’s not a good use of a wiki to cut and paste other people’s lengthy documents that are available elsewhere, and despite the 2009 version being out of date, I didn’t feel bold enough to delete all the out-of-date pages. I just added an introductory sentence linking to the more recent version elsewhere.

If something is not available elsewhere or should be archived for some other reason, it would be possible to keep a record on the wiki but separate from the main, up-to-date material. So, with the Inform 7 document above, it really just needs links to the 2009 version elsewhere, maybe with an upload of the PDF. With your examples, all the interpreters are probably historically interesting so should not be deleted entirely, but it would just need to be made clear which are out of date, which are useful for which purposes etc.

1 Like
35

I mean different information, and not necessarily the same structure of information for every page in a given category (that’s what databases are for.)

Look at IFWiki’s page on Galatea. There’s almost nothing there that isn’t on its IFDB page, and nothing that tells you how significant it is in IF history.

Look at Wikipedia’s page on Galatea. That’s much more informative and enlightening, and uses the strengths of being a wiki page rather than a database entry.

If you asked a member of the IF community to talk for five minutes about Galatea and wrote down what they said, you’d get something much more like the Wikipedia page than the current IFWiki one. With IFWiki as I’d like it to be, you wouldn’t have to collar an IF community member to get that information. And you’d have that resource for any IF work, not just ones that are “notable” by Wikipedia’s somewhat arbitrary and highly variably-enforced standards.

7 Likes
36

I think IFWiki should have the confidence to operate on a sort of “meta” level where it actively tries not to replicate anything that is or should be elsewhere. I think that would entail some deletion of content (although not deletion of all game pages… I agree now and feel like such a Philistine).

So for games, it should aim to have links to IFDB, CASA and elsewhere, but not links to reviews (for example) if those belong better on IFDB etc. If IFDB or CASA has an API maybe it could automatically take a summary from there (and attribute it clearly).

For documents like that Ron Newcombe guide, it should link to but not reproduce them.

Even uploads for historical interest/archival maybe should be uploaded to and linked from IFArchive rather than uploaded to IFWiki.

With the out-of-date information that is presented as if it is in date, maybe we need to start again at the top of the page and keep the old table (or whatever) at the bottom of the table for reference, maybe temporarily.

This forum as a source of information too, including it’s own wiki posts, so maybe IFWiki should link to relevant topics here, like the big Inform 7 one, Inform 7 documentation and resources, or even smaller ones like List of glulx/.gblorb interpreters. Alternatively, maybe these forum-wiki pages in particular are cases where the forum should bow to the wiki as being the more appropriate place for the information.

37

That’s a very good question which I deliberately left out, because my post was already very long, and because I admittedly don’t have a good solution myself.

Maybe something like a new minimal template would help, as was mentioned above, which would just contain the most important information (as described by the top-priority items in question 9 in the survey thread you linked to: IFDB link, Title, Author, Release date, and Platform/Authoring system). Ideally auto-generated/-populated.

Basically, what you said in the thread that is, in turn, linked from that other thread:

If we are serious about de-emphasizing the game entries, then such a template could also contain a stub info with a sort of “reverse intent”: something which does not say

“This is incomplete, please help expand it.”

but rather

“This is a minimal game listing, please consider contributing further game information to the IFDB, because we aim to bundle the info in one place.”

I dunno ¯\_(ツ)_/¯.

2 Likes
38

Yes, hm, fair point (about the interpreters and also in general).

Part of that stems from a dual role that IF Wiki has: it serves partly as an encyclopedia and historical record, and partly as a source of up-to-date, currently-helpful advice and knowledge. Or at least, one would hope so.

For this particular case, maybe one could one add a separate section or table at the top, where only current interpreters on the most widely used platforms are listed.

Yes, to be honest, when I wrote the list above, I had actually momentarily forgotten about the clubs feature. :slight_smile:

Hmm, on the one hand I see the appeal and sort-of-agree about filling a different role/niche.

On the other hand, the structure makes it easy for volunteers to create new pages, because they know exactly what to do and what to place where.
(But admittedly, it is too much at once, which is discouraging.)

The structure also makes some aspects less intimidating, I think.
For example, in the people category, I would find it difficult to write about a person’s contributions in free-form sentences, whereas it’s easier (and also uncontroversial etc.) to add a new bullet point under game/tech/testing credits and so on.

Maybe there’s a way to have the best of both worlds, I don’t know.

1 Like
39

This is a good point about IFWiki serving a dual role.

What if we put the currently-helpful information on the “interpreters” page (or at least, only include platforms that have come out in the last N years), but include a link to a separate page where we keep the older information, for the historical record?

Also, one challenge of pages like the interpreters page is that it seems like fairly specialized knowledge. Even if someone has been involved in IF for years, they won’t necessarily know all the right information to put in that table. For instance, if you never use a given OS, you might not have reason to know what interpreters are good for that OS.

3 Likes
40

This is a good idea, and what I tend to do when a page would be too unwieldy with the older information. The interpreters page isn’t that big so you could do that, or just have the older stuff further down the page under a separate heading.

You’re right about the interpreters page in that it’s so technical and daunting to edit. It’s like being asked for directions and saying, “Well I wouldn’t start from here.”

I have to admit I’ve been running a wiki in a totally different area of life for about 15 years which is part of my interest here. I’d love to upgrade the software and add some extensions…

1 Like