What is the state of IF preservation? I’m beginning a Master’s degree in the fall with a focus on technical issues in game preservation and am trying to find some research topics. 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.
4 Likes
we currently have the IF archive to keep sourcecode and game files, and IFDB to keep track of games. this goes back some number of years (the 90s I think). I’d say the state of IF preservation is pretty decent in that sense. In general, the IFTF has made it a mission to conserve a lot of things.
there are past discussions about IF that haven’t fully been preserved due to net/link rot and the community hopping from platform to platform, but I think a lot of stuff did get imported onto Discourse (this forum software) so even a lot of that’s preserved. The gaps, imo, are some lost games from the past and, I’d say, whatever’s going on in the tumblr twine scene, since that’s a world that’s seen as separate from this sphere (intfiction forums, ifdb, etc) but is also IF. the nature of the blogging platform makes it harder to keep track of things, and I don’t think most of those people have their games on ifdb. @manonamora iirc has a blog that collects them though.
15 Likes
I think that’s the big disconnect; it’s social, not technical. Communities that are not tuned into this one don’t always think about preservation as a community need. “I better upload a copy to the IF Archive in case my web site goes down ten years from now” is not a common sentiment!
Colin Post was getting started on research in this area. See thread: Let's Talk about Collecting Narrative Games in Libraries . His notion was to come up with a community license which would allow games to be archived by libraries. (Open-source and CC licenses allow this, but the vast majority of IF games don’t use those licenses.)
He got as far as getting an IMLS grant to support that research. However it got torn up in April as part of the US government’s general, um, tearing up of things.
14 Likes
On the whole, the state of preservation is only fair. The vast majority of IF pre-dates the IF Archive and there is no central repository for all IF games. Instead, you have to search a myriad of archives and repositories, most of which are just a grab bag of dumped files with no organisation whatsoever. Some archives are better than others and do at least allow you to search by title, author and so on. Many archives do at least acknowledge the reported existence of a game, but missing games are the norm.
New games are being discovered all the time. (By ‘new’, I mean games that were thought to have been lost, but have been discovered in a dusty attic or a garage sale or on eBay etc.) Quite often, restoration of these games is a challenge in itself due to degradation of the original magnetic media, typically tapes or floppy discs. Preservation of the original packaging, tape inlays and instructions is also a challenge.
For non-commercial games, there were a lot of public domain and shareware games distributed by user groups and public domain libraries. As these folded, their games became lost. Very few of these have been preserved online.
In the modern era, many games are hosted on private web sites or places like Steam and itch.io. These only last as long as the owner decides to keep them. The IF Archive is a dumping ground for mostly modern games, but this is poorly organised. The IFDB does not host games itself, but only provides links to them. It’s not uncommon to find that those links no longer work.
7 Likes
Two things that I would think help round out your knowledge base.
For me, the biggest “lost game” is “Blood & Laurels” by Emily Short. I am probably going to get the description of the issue wrong, but if I remember correctly the game only exists now on iPads that had it installed around 2014. It isn’t available on the App Store and the company (Versu) is not in business.
The other unrelated thing to know about game preservation is that a lot of it is being done with the eXoIF project. Gunther Schmidl has spent a lot of time getting games from past decades to easily work on modern systems through it.
7 Likes
I still have Blood & Laurels on an old iPad. I can’t believe it still works. I also have Versu on it with The House on the Cliff by Emily Short too. I never update the iPad. So I can still run 18 Cadence by Aaron Reed and the Lost Legends of Infocom. Too bad Apple messed everything up.
5 Likes
I wonder if there’s a way that app can be retrieved from your iPad 
I don’t know enough about apple software to say…
5 Likes
Long term preservatation should include game indexes with file hashes both as unique identifier and as corruption detection. Multiple versions should be archived where possible.
5 Likes
I will quibble that only a minority of IF games predate the IF Archive. The period from 1977 to 1992 was only fifteen years, and there were many fewer people making games back then. Probably the vast majority of IF games were made in the last 15 years! (2010 to the present, aka “the Twine era”.)
However, your point is taken: early commercial IF is a significant chunk of history, and the IF Archive doesn’t try to cover that for legal reasons.
That’s not just IF, of course. It’s a problem for the entire history of games from the 1970s/80s/90s. If we had a complete collection of IF games, then a general game historian would look at it and say “Dammit, now we have N+1 repositories to search!” (Cue xkcd joke.)
I think the way to approach that era is to do a lot of work and then file it on the Internet Archive, which has a well-established “copyright? schmopyright” stance. I did this with the Infocom game file collection, for example: it can be found at https://archive.org/details/infocom-games-source in case my web site goes away.
20 Likes
I did this a few years ago using Apple’s “Apple Configurator” tool. I don’t know whether that’s still supported, but it worked in 2022.
This doesn’t let you play the game off the iPad. (I think if you try to install the retrieved game on someone else’s hardware, it just shows a “You don’t own this app” error.) It’s useful for examining assets in the app file package.
3 Likes
One era that is pretty well-preserved is the early online hobbyist era from 1995 onwards. The newsgroup communities (from which this one is eventually descended) used the IF Archive, rather than personal websites or a platform like Itch, as their primary distribution mechanism for new games. So almost everything from that era is archived as a matter of course.
Another aspect which you may or may not be aware of: many works of IF use either a standard data format or a virtual machine. The most obvious example is Infocom’s Z-machine, which was subsequently also used as a target by the Inform development system, but many other companies and hobbyist systems used a similar approach. This has made it easier to keep a lot of old games available since as long as there’s an up-to-date interpreter, no work is generally needed to get individual games working.
6 Likes
I don’t think that’s true. The IF Archive opened in 1992. IFDB is only aware of 2,790 IF games with a publication date from 1965 - 1992, (of which 78% were published in the 1980s) and over 11,000 games from 1993 to today. (There are 543 games with a blank publication date.)
Query of IFDB games by publication year
MariaDB [ifdb]> select count(id), year(published) as year from games group by year(published) order by year;
±----------±-----+
| count(id) | year |
±----------±-----+
| 543 | NULL |
| 1 | 1965 |
| 2 | 1967 |
| 1 | 1970 |
| 1 | 1972 |
| 1 | 1974 |
| 1 | 1976 |
| 2 | 1977 |
| 10 | 1978 |
| 29 | 1979 |
| 47 | 1980 |
| 61 | 1981 |
| 112 | 1982 |
| 221 | 1983 |
| 385 | 1984 |
| 343 | 1985 |
| 273 | 1986 |
| 306 | 1987 |
| 182 | 1988 |
| 257 | 1989 |
| 186 | 1990 |
| 189 | 1991 |
| 180 | 1992 |
| 131 | 1993 |
| 94 | 1994 |
| 96 | 1995 |
| 112 | 1996 |
| 107 | 1997 |
| 155 | 1998 |
| 157 | 1999 |
| 207 | 2000 |
| 283 | 2001 |
| 268 | 2002 |
| 266 | 2003 |
| 215 | 2004 |
| 195 | 2005 |
| 189 | 2006 |
| 180 | 2007 |
| 206 | 2008 |
| 158 | 2009 |
| 233 | 2010 |
| 261 | 2011 |
| 429 | 2012 |
| 619 | 2013 |
| 502 | 2014 |
| 572 | 2015 |
| 558 | 2016 |
| 508 | 2017 |
| 466 | 2018 |
| 495 | 2019 |
| 568 | 2020 |
| 425 | 2021 |
| 578 | 2022 |
| 796 | 2023 |
| 828 | 2024 |
| 181 | 2025 |
±----------±-----+
I suppose it’s possible that there are over 8,000 games published before 1992 that don’t have an entry in IFDB, but I don’t think there’s any reason to think there are that many. Prior to 1992, the year with the most games known to IFDB was 1984, with 385 games tracked. In 2024 there were 824 games.
Even if we assume that there were 400 works of IF published every year in the 1980s, and IFDB just missed them, that would mean IFDB is missing only ~1,000 games in that time frame, no where near the 11,000 games published since the archive launched.
IFDB is the organizational system for games on the IF Archive. It is uncommon to find links to the IF Archive that no longer work.
IFDB links to other third-party websites do break, and it’s not uncommon to find that those are broken.
7 Likes
Right. Now that both of these are both IFTF projects, we do our best to coordinate. The people who file games on the Archive also update IFDB to keep the links correct.
(Obviously it’s not a perfectly synchronized system. If we were starting from scratch, we’d be thinking about a single service that incorporated the metadata and the files. But we’re not starting from scratch, and building a whole new system is not in the cards.)
5 Likes
Well, we know CASA has far more games from the 80s than IFDB. That’s a discussion that has come up at least once; whether they should be copied across (conclusion: no). On a quick count they list 5,567 games from 1974 through 1992, with 828 games in 1984 (again a peak).
But yeah, still only about half of that 11,000 games, which we also know is mostly the IF from this particular community, which is a fraction of what has been published more recently…
8 Likes
Garry makes an important observation, though, around the preservation of early pre-IFDB work.
Such games are often difficult to find and play, with preservation of titles quite patchy depending on the specific micros or mainframes they were written for. For example, the ZX Spectrum archive is very comprehensive and titles are easy to access, download and play. Some other micros, less so. You usually need to have a knowledge of where each machine’s archive is and understand which emulators and procedures you need to use to play games.
CASA is not an archive of IF works. It’s a database of game titles, where the primary purpose is to store solutions, maps and tips.
2 Likes
This topic is about the state of IF preservation, not the state of IFDB or IF Archive. IFDB is a database of games. It does not preserve anything. That is the role of the IF Archive and, as everyone keeps emphasising, this primarily preserves modern games. IF Archive has a policy that prevents submission of games without the author’s permission, otherwise the number of preserved games would be far greater.
For the majority of early games and commercial games, you have to look elsewhere. The Internet Archive is a good place to start, but it’s very hard to find anything and it only covers the more popular platforms. You then have to start looking elsewhere. Having done this many a time, I can assure you that it is a nightmare that usually ends with nothing but frustration.
For those that doubt that IF even existed before IF Comp started, here’s a chart I did some time ago for another thread, but I couldn’t find that thread. Note that there was a peak in 1984 (long before choice-based IF became a thing). This was the golden era of IF and far exceeds any year since.
There was a steep climb around 2013 when choice-based IF became a thing. CASA’s games in the modern era are much lower than IFDB because CASA does not include choice-based games and that community is not so interested in ‘modern’ games, so no one submits them to the database. I should also point out that CASA is only a database, just as IFDB is only a database, and it does not preserve any games, but it’s good at identifying what was available over the 40-year history of IF, providing it wasn’t choice-based.
5 Likes
Maybe what’s missing is a renegade IF archive that doesn’t worry about consent or copyright.
6 Likes
I am searching in the internet for some years the games that Jason plays in Renga in Blue. Some people asked me for the files with preservation and online playing in mind, but I haven’t received any notice after downloading the files. By the way I tried to contact IFarchive in order to actualize the database with no answers. Notice that there is a big amount of games wich aren’t listed in IFarchive or IFDB.
2 Likes
I’d definitely agree that the primary issues facing video game preservation generally - and IF specifically - are social and, unfortunately, legal. There are definitely technical complications, especially for things like live service games, but by and large, we know how to preserve games.
The social piece, like Zarf mentions, is both that game developers aren’t necessarily thinking about preservation as a top concern (though, in my experience, lots of developers are deeply concerned about this but don’t have the time or energy to do too much about it), and there aren’t great connections between these communities and the library/archives communities who are both interested and have the infrastructure and expertise to preserve games as part of culture.
The legal piece is more complicated for commercial digital games distributed on platforms like Steam and less complicated for games that are already released for free, like a lot of IF. Really, these represent two different sides of the legal issue spectrum. Commercial games have tons of legal red tape that libraries need to cut through - and bigger publishers are not interested (yet) in talking to libraries. Freeware games don’t have that red tape, but they also don’t have clearly expressed permissions for what libraries can - or should - do with those games.
As I learned from talking to folks here, just because a game is released for free doesn’t mean that libraries should assume that they can snap it up without developing some kind of arrangement with the rights holder. IF authors will have very different attitudes, ranging from ‘sure, take it and do whatever you want with my game,’ to, ‘I never want anyone to see this game again! why is this still on the Internet!!’ Libraries need to build trust with game developer communities by talking with them about their preferences and expectations, even if games are essentially donated to the archives.
I’ll actually have the article talking about what I learned from my discussion with IF folks coming out in a couple months. So I’ll share that when it’s published. My plans for building on that research have, unfortunately, been cut short after my funding got DOGE’d. But I’m regrouping now and thinking about how to move forward to develop real tools and strategies (e.g. license templates for game developers who want to donate or sell their game files to libraries) for libraries to collect and preserve games better.
17 Likes
And differently complicated for games that were commercial in the 1980s but are not on any modern platform!
When was this? I don’t think I saw the message.
You’re referring to a collection of the games that Jason Dyer has played on his blog? (He’s going in chronological order, so these are all 1983 and earlier, to date.)
Some of these are commercial, so they have the same problem as the Infocom games. (And the earlier Infocom games are on the list, of course.) Maybe I should do the same thing I did with the Infocom games – host them on my personal website.
4 Likes