What is the state of IF preservation?

General
21

Lots of folx have already mentioned things the IFTF is has done and continues to do.

There are two other larger collections I am in communication with as part of my own preservation efforts. The first is Consortium on Electronic Literature (CELL), which has been trying to import existing metadata on works as part of the Wikidata project (and as found on Wikipedia, among other services). The second is The NEXT, which is a smaller collection but has made great progress toward emulated systems and configurations for older electronic literature of all sorts.

Are there any major issues being faced that I could look into tackling as part of a thesis or independent study course? I come from a computer science background, so I would be looking for technical challenges that I could potentially help with.

Major challenges include, but are not limited to, the following:

Platforms

  • Death (i.e., the death of Flash for early online works).
  • API changes (i.e., what happens when the Facebook API changes and older games don’t work any more.)
  • Policy changes (i.e., Steam and Itch’s removal of “adult games”)

Preservation and Access

  • What does it mean to play older works? Should hardware controls be emulated or re-constructed? Should “lag” be programmed into emulated systems to mimic older access requirements and processes?
  • What does it mean to save distributed works? Some older MMOs have now died, and the stories that existed within those might not be recoverable. The same, too, with stories across social media platforms. Trying to preserve what it means to access, for example, for electronic literature across TikTok or Twitter posts when those platforms change is now a problem.
  • What does it mean to save configurations? Some older visual novels only work on Windows 98. Some only work on Windows XP. These versions are no longer maintained by Microsoft. What does it mean to have virtual systems of different, older operating systems to still play games only designed for them?

Metadata

Something I have been working on this past summer was trying to identify where the larger, funded collection aligns on fields. For example, The NEXT has a comprehensive set of fields that cover accessibility concerns (json-elms/schema/schema.md at main · videlais/json-elms · GitHub) that others could also implement or add as optional fields.

A major challenge, in my conversations with people from The NEXT and CELL, is in just trying to figure out what metadata to save or maintain and how to get different databases to communicate with each other.

Traversals

As part of efforts about the Electronic Literature Collection, traversals of older works has become a growing concern. Myself, among other people, have been trying to capture video and audio recordings of works to save them along with metadata about the work. This has become a renewed effort as platforms like Steam and Itch are forced to remove works that may or may not be saved anywhere else.

11 Likes
22

Honestly, even knowing that IF is still a thriving, if niche genre, that barriers to game making have decreased over time, IF has several widely used game creation systems, some of them catered to those without a coding background, that the rate of new computer games has been trending upward in general, etc., it still feels somewhat counterintuitive that records suggest games published since 1992 outnumber games published prior to 1992 by at least 2:1.

And yeah, as much as I think most dichotomies are false and tthat the optimum of most spectrums is usually somewhere between the extremes, Copyright and Preservation are pretty strongly in direct opposition and hard to reconcile… and it doesn’t help that authors often don’t own their own work and that the publishers responsible for the disgustingly long duration of copyright will gladly let any work they own rot if it isn’t actively making them money and will have no qualms about shutting down efforts to preserve works they otherwise don’t care about.

3 Likes
23

It makes sense to me—Crowther/Woods ADVENT came out in 1977, and 1977-1992 is 15 years, while 1992-2025 is 33 years!

8 Likes
24

I did upload my IF-related writing, ten years’ worth of it, to the IFArchive. I very much appreciate having a say in the matter, unlike on the Internet Archive. We’re lucky that most interactive fiction is in formats that can be easily emulated: if not Z-Machine, then 8-bit Basic dialects. Flash is more of a problem, though projects like ruffle.rs have been chipping at it. The social aspect remains. Can we easily identify areas where preservation is lacking and do more advocacy or something?

4 Likes
25

There are two different metrics there. The rate of games being published is different from the number of games.

It is not surprising that there may have been more games published since 1992 then there was before, as there has been a lot more time since 1992 than there was from the dawn of text adventures to 1992.

Garry’s graph, however, does show that the rate of games being published/created was much higher at points in the past, with 1984 as the focus point. It was pretty much as easy to create an adventure game (or IF) back in the 1984 as it is today (we had great tools aimed at non-programmers too!). The higher rate of games there is simply down to text adventures being a more popular medium back then.

3 Likes
26

I think the methodology is flawed enough that Garry’s graph doesn’t really show that: he points out that people submit very few modern games to CASA but completely ignores the likelihood that what this community submits to the IFDB is only a fraction of the text games being created around the world today.

And even if you’re only looking at the IFDB for modern games, Dan’s numbers show a peak of 828 games published in 2024 (after Garry’s graph ends), every bit as high as the rate CASA shows for that peak in 1984.

So suggesting that the rate of game creation today was much higher at points in the past is simply not true.

Unless you’re referring to only classic-style adventure games.

And of course, as several people have pointed out, recording the names and descriptions of games isn’t full preservation.

5 Likes
27

Has the IFDB or IFarchive archived all of the 2,000+ text adventures found at, e.g. World of Spectrum?

4 Likes
28

that’s a good point–there are several pockets of text adventure creation sites/archives/communities aside from tumblr that aren’t part of the ifarchive.

the answer in this case (re: ifarchive) seems to be permission from the creators again. reading the site, WoS got permissions from copyright holders to store/display games on their site. uploading it to ifarchive would need different permissions.

ifdb has another issue again that without the file it’s just metadata. Still might be worth doing, or it might be another CASA thing.

2 Likes
29

Registering games in IFDB is important to keep a record of its existence, regardless of whether it’s legal to upload it to the IF Archive. After all it’s hard to know what games we’ve failed to preserve without knowing about them.

8 Likes
30

Speaking on the idea of ‘how many games are created each year that aren’t recorded in IFDB’, there was a time when someone wrote an automatic uploader for ifdb that added every game ever published on textadventures.co.uk. There were tons of them, all the time, and no one ever played them or rated them. So the automatic uploader was turned off.

I guess the truth is that a lot of the game missing from IFDB are just plain bad. I think it’s not a bad idea to try to record things and IFDB can definitely serve that purpose, and I think it would be cool to port over CASA entries to IFDB, but I’m not too concerned about gathering up large swathes of games that don’t get plays or comments.

8 Likes
31

To clarify my previous post, even knowing all the reasons one should rationally expect there to be more games published since 1992 compared to prior to 1992, the intuition is that graphics being primitive enough to make text-only games mainstream viable should trump everything else, so the data suggesting it doesn’t is still surprising… Also, I heard mention of 1965 somewhere in this thread I think, so I think my brain was thinking 1992 was closer to the temporal midpoint, ignoring that there was likely vanishingly few games published prior to the earliest personal computers.

Of course, the data is imperfect, collated across several sources that have different goals and might not be counting in the same way, and we don’t know what we don’t know… Also, as this is a primarily English forum, I’m guessing most of the sources discussed are also primarily English, and that there are potentially huge swaths of non-English IF unaccounted for.

2 Likes
32

Ugh… my ISP blocked that site. (?)

1 Like
33

Similarly, have they archived the 3200+ Inform 7 games from Playfic?

2 Likes
34

A lot of those might not even be full games, just little tests to see if they wrote the code correctly. I feel like it wouldn’t make sense to. Just my two cents.

6 Likes
35

You’re right… on a second look, I can see it’d be tough to tell the wheat from the chaff.

4 Likes
36

This might be getting into the question of preservation versus curation of history. At the extremes, preservation cares not about the quality or distinctiveness of each artifact and will archive everything, no matter how insignificant, of poor quality, or similar to another artifact a given piece is. Meanwhile, the Curator cares more about presenting the best of the past and will focus on the definitive versions of a given work, ignore all but the most distinctive alternative versions, and might even combine elements from different revisions to create something that is not quite any particular historical release, but arguably the best of all worlds.

Personally, I’d argue both are important and ideally, an archive would take the preservationist view to its logical extreme while also cataloging every artifact in a way that makes finding something specific and identifying what’s what convenient… Of course, in practice, actually cataloging everything takes too much effort, so cataloging is often limited to what’s curated and preservation often gets limited to dumping copies of stuff into an archive and hoping someone finds the time to organize the new additions.

5 Likes
37

Taking things to their logical extreme is illogical. :)

3 Likes
38

It would be nice if someone who has Blood and Laurels still installed could record a few game sessions and upload them to YouTube. It’s surprising how scare information about this game is, considering how many people seemed to have played it. I don’t think there is even one full session uploaded anywhere.

9 Likes
39

I wrote my undergraduate capstone on the analysis and preservation of electronic literature and I’m beginning my Master’s degree this fall as well (which is one of the reasons I’m so late to the party! I’ve been running around getting situated at my new university the past week :face_with_spiral_eyes: )!

Some of the main sources I used for my capstone in regards to the preservation of electronic literature were:

“Acid-Free Bits” by Nick Montfort and Noah Wardrip-Fruin on the Electronic Literature Organization website, “E-Lit after Flash: The Rise (and Fall) of a ‘Universal’ Language” by Anastasia Salter and John Murray, and “Challenges to Archiving and Documenting Born-Digital Literature: What Scholars, Archivists, and Librarians Need to Know” by Dene Grigar from the book Electronic Literature as Digital Humanities (this book was a lifesaver anytime I needed academic articles on electronic literature, it has a lot of very interesting discussion on the topic).

Grigar’s article in particular gets into the traversal work that Dan mentioned and the Electronic Literature Lab at Washington State University’s Vancouver campus.

Hopefully these articles will be useful to you and best of luck with your graduate work!

12 Likes
40

I’m not aware enough with the territory to know how big of a problem this is for IF in particular, but a general problem with software preservation is that the tech goes obsolete.

It’s one thing to preserve source code and walkthroughs, but those don’t necessarily give you the ability to preserve the original experience of running the game. Once games started to be written for interpreters, this became less of an issue, because there’s been tremendous effort in keeping the interpreters backward compatible.

But some games, especially older ones, were written in obsolete programming languages (or, at least, obsolete dialects of old programming languages) and relied on hardware architecture and features that are no longer readily available and for which there may not be accurate emulators.

For example, the IF archive has the source code for a few iterations of Will Crowther’s original Adventure game and a couple with the Don Woods enhancements. But these were written for the PDP-10 using a DEC dialect of Fortran that isn’t quite Fortran IV nor Fortran 77, so good luck finding a way to actually play them exactly was the originally worked. (The notes with those files warn against using those sources as the basis for a modern port.)

Shortly after Woods’s version was released, a follow-on version was prepared for a slightly more recent version of Fortran and targeting a different processor architecture that’s more similar to today’s mainstream architectures. That’s the version that was widely distributed on the DECUS tapes, and thus its the one most people played in the late 1970s. Although that port was intended to be a faithful to the original, there are some differences to the game, there was a bug or two that were introduced, possibly a bug or two that were fixed, and some meta features were removed.

I happened to play on an actual PDP-10, and I’d long sought to be able to re-experience it. But I have never found a playable version that truly matched the game I got to play back then.

I believe that, had more people been exposed to the actual original game, navigation conventions in parser-based IF might have evolved a little bit differently. In other words, because, even then, the language and hardware were a bit off the mainstream, the small changes introduced to make the game more widely available had a larger impact.

I think accurately preserving the experience of playing the most influential games would be a boon for researchers.

7 Likes