Reasons a player might not want to engage with Web-Based parser deployments

This community and its tools are on the whole refreshingly free of the general tendency to record and scrutinise every aspect of your interaction with a website for analytic or commercial gain; but in the exceptions (like IFComp transcript recording) I do think we could do better with notification and consent.

Don’t think I can check what it actually looks like for IFComp, since there isn’t an IFComp running currently. I think there’s one or two other isolated places that do transcript recording, but I can’t remember where </fud>. In technical terms, I think it is much easier for someone to enable online play without transcript recording than with. (I’m pretty sure iplayif.com, as used by IFDB’s “Play Online” links and hence probably a large proportion of total online play of parser games, does not record, for the avoidance of doubt. I wonder if it’s worth adding a notice to it saying so.)

(And yes, I avoid playing online where I know or suspect transcript recording is happening. Another reason I tend to play offline is that I like to keep a transcript for myself, and this is often hard-to-impossible with web-based players; and for the security of having saved games in my filesystem, rather than in my browser and possibly subject to being unsaveable/lost due to per-website local storage limits.)

4 Likes

That is a little disconcerting. If a game offers a White House location, and a stick of dynamite as a object, and I type in blow up Tr*mp. I really don’t want my name in an FBI file next to Jeffrey Epstein.

I believe Parchment has added the ability to download a transcript file fairly recently (in the past year, perhaps).

In the interest of full disclosure, Bisquixe does not support this.

Huh! I didn’t realize this was a concern for people, though it makes sense. I’ve been working on adding the “automatically send transcripts for IFComp” feature to Dialog, but now I’ll make sure to add a toggle to turn it off.

8 Likes

Wow! What a gratuitous invasion of privacy. Is there a way to switch this off?

1 Like

Download the game instead of playing it online via the IFComp website. (Or play via IFDB or the Archive; anything that’s not ifcomp.org.)

It sounds like it’s worth asking the IFComp staff to add a toggle for this, too.

5 Likes

For players, offline play is the only guaranteed way.

For authors, if you submit a raw parser file and IFComp creates a “Play Online" version (which it automatically does), there is no way to avoid the collection of transcripts. But I’ve had success bundling my parser games as websites within Inform 7, and then uploading those to IFComp. This disables the IFComp transcript collection. I haven’t received a single transcript from any parser game I have ever submitted to IFComp, which I consider a little private victory!

9 Likes

Thanks for the explanation. I think it’s outrageous that IFComp even thinks they should automatically collect transcripts. Basically spying on players.

This whole thing reveals the modern IFTF attitude;

If you want to play our games, you have to agree to our draconian ToS. Accede to all our demands! Bwahaha, maniacal laughter of glee. Our Games

@Giger_Kitty I really liked A Colder Light for it’s story and setting. It was an experiment in interface design and maybe that design aspect impacted the downloadable version.

That said, I have a strong desire to download games to play them and it has nothing to do with lack of internet access. It just revolves around my own sensibilities and I accept that I’m in the minority with that… so I also accept the friction that brings to the table for myself.

If they intend that I play the super glitzy Vorple version but curate/give me a version that runs offline, I’ll play that.

As @Encorm mentioned, you’re describing an accessibility need. Offline versions are an important form of accessibility… and when it’s looked at through that lens, it carries more weight. I can understand your frustration.

2 Likes

For what it’s worth this has been in place for quite a while and seems to be for the convenience of authors (who AFAIK are the only ones who can access the transcripts for their games).

I think it’s worth revisiting as modern feelings about Internet privacy have shifted, but it’s not a new feature by any stretch.

13 Likes

Yeah, this isn’t really about the IFTF. It reflects more on IFComp authors, in my opinion. Many of them seem to desire these transcripts, which I have found slightly disconcerting since I first entered the comp in 2015. If attitudes toward internet privacy are finally shifting, that’s good in my book!

4 Likes

A reminder to everyone that the IFTF doesn’t monitor this forum, so anyone with strong feelings about this issue is best off emailing them at ifcomp@ifcomp.org. It’s early enough that they’d likely be able to get in a change of this (small) scale in for the upcoming competition, should enough people bring it to their attention!

12 Likes

I suppose in retrospect this attitude makes sense, but from the perspective of an author, having ifComp automatically collect transcripts is a huge boon. I do agree it should be ‘opt-in’, but there’s literally no better way to debug an obscure problem in your game than having a transcript. Literally the first two commands I ever type into any parser game are >SCRIPT ON, and then >RESTART. Every time! And of course that is literally opt-in, but it never even crossed my mind that anyone would object to an author having access to your transcript of their game. These are volunteers and hobbyists, just trying to share cool things, and transcripts help them do that better. These aren’t nefarious marketers, carefully calibrating how many Monty Python references they should insert to make you nostalgic for Coke.

I mean, don’t get me wrong; if you don’t want an author to see your transcript, that’s fine, and if you don’t want to see player transcripts, that’s fine. But there’s a weird assumption of bad actors here that I don’t think is warranted.

10 Likes

Well, I call my own opinion “paranoia” for a reason. But it’s based on many conversations I had back around 2015-2017 with fellow authors who were never too sympathetic with my concerns re: transcripts. I have heard so many arguments to justify the quiet collection of transcripts. So I just gave up. My own trust is broken. Now I play parser games offline, and likely always will. Then I can opt-in to transcript collection of my own volition if I want to; and if I don’t want to, I don’t have to wonder whether I’m secretly being observed.

4 Likes

To be constructive, I’m trying to work out what my ideal user experience change would look like.
(Possibly the topic of IFComp transcript collection should be split to a new topic…)

Starting point: here’s what the ballot page looked like last year. At the head of the game list there’s a note (presumably the one that @CMG caused to be added):

Please note that some entries, when played online through this website, collect anonymous transcripts of player input for the benefit of those entries’ authors. You can opt out of this by downloading and playing these entries offline. Read this note for more information.

with the explanatory note (which is available all year) saying:

A note about transcripts

Interactive fiction works created with Inform offer the ability to record and store a complete transcript of play-sessions, including all player input and story output. Through the use of certain add-on features, this ability extends to Inform-based work played via web-based interpreters. As a service to competition entrants, this website (ifcomp.org) enables this style of transcript-recording for the Inform-based entries that it hosts.

Therefore, authors of Inform-created works – including Z-code and Glulx entries – can read full, anonymous transcripts of sessions played via the ifcomp.org-hosted copies of these entries. If you, as an IFComp judge, wish to play these entries without having your input recorded, please download the entries and play them that way, rather than playing them directly on the ifcomp.org website.

If you have any questions about transcripts, please contact the organizers.

but, I think, nothing visible in the path from looking at a game on the ballot page to entering your first command into the parser.
(It’s more difficult to explore this via the Wayback Machine. I’m taking A Smörgåsbord of Pain as an example where the author used IFComp-provided infrastructure, since what ended up on the Archive was just a .gblorb file. But it and another similar example aren’t playable via the Wayback Machine, so I’m not 100% sure what a player sees.)

Very easy to miss that it’s happening; no way to opt out while still having the convenience of web play; and no way for authors to opt out of transcript collection, which at least one author wants.

My personal position: in general, on the web, I minimise the data that’s collected about me as much as possible (and accept quite a lot of inconvenience to do so); but in this case, I trust that transcripts are going only to authors, who are using them to learn and improve their work / craft, and find them super useful; and in many circumstances I’d happily participate in that.

First, opt-out or opt-in?

  • There’s an argument that unless it remains an opt-out, authors won’t receive enough transcripts to be worth having the feature at all. Many authors have said how helpful access to transcripts is.
  • On the other hand: back in the day, some players went to the effort of emailing transcripts to authors. So maybe enough players would still click on a low-friction opt-in version of the same for it to be worthwhile? (Particularly if the choice was presented when the player might just have run into a problem, like when they end the session / close the window? Dunno if that’s practical.)

I think ideally I’d want:

  • To be told that transcript collection can happen, sometime after I’ve clicked on a specific “Play online” link. (With a brief but clear explanation of what that means – like, one sentence – and a link/expand for full detail.)
  • To be asked whether I’m okay with that, probably, at least once a year (because I might forget this is a thing by next year’s comp).
  • To be reminded, on an ongoing basis, (a) that transcript collection is a thing here, and (b) whether it is currently happening for the game I am playing.
    For Parchment… I could imagine a red “recording” dot in the corner of the window, without getting in the way too much? Which can be hovered over or clicked on to explain what it means and to allow changing that. (And similarly for e.g. the Dialog aastory web player.)
    Anyone doing layout fancy enough for this to get in the way is probably using their own Vorple/Bisquixe/whatever setup, so none of this applies. (edit: oh, huh, maybe that’s not true, judging by this)
  • To be able to express “don’t record ever” or “I’m fine with any recording”, and have it remembered (for at least the duration of a comp).
    (Does ifcomp.org already use cookies, or is all its statefulness dependent on being logged in? You shouldn’t need to be logged in to be able to control transcript collection.)
    (But not suppress the display of the previous point, in any case.)
  • To know when the recording will be submitted and at what granularity. (Is it live? Or if I turn it off during a play session, will that prevent the submission of commands I’ve already typed?)
  • (If this version of Parchment-or- whatever supports recording transcripts for the player) not to have my transcript-collection preference interfere with my ability to save transcripts for my own purposes.

(I don’t think allowing authors to say they don’t want transcript collected has many UI implications for the play interface, it’s presumably a Small Matter Of Programming™. Except I suppose for the recording indicator always being off and showing something like “This game never collects transcripts” when poked.)

4 Likes

My assumption—since I wasn’t around—has been that the practice emerged from a time when the community had a different shape. Perhaps there were fewer people who interacted more. It might have been safe to assume that community members were universally interested in workshopping their texts, and that this went without saying.

New questions and perspectives emerge with the passage of time. I don’t fault the organizers or think that they are up to no good. E: I’ll strengthen my language here: they are performing an invaluable service to this community.

I do wonder how I might reassure players that I am not gathering transcripts myself. I had not previously considered the possibility that a player might have privacy concerns with my games. Perhaps that is just an inherent danger with web technology.

9 Likes

If people don’t want the transcripts of what they type into a game to be shared without their knowledge or consent, then that’s pretty much it. Merely “wanting for your input to be private” should be enough without having to justify it. I don’t necessarily want to know that others will see my playthrough, and potentially judge me or my personality or my skills. I’m playing a game privately, just me and myself.

I have had people watch me while I play other games. It is an unpleasant experience for them, apparently, because I take forever to do things. I no longer play around those people. I’ll play how I want to play…

4 Likes

Do you remember where/when was this discussed? Was it like the authors’ forum or such?

It looks like transcript collection has been a thing since at least 2011 – this (not working) archived play-online page has “Anonymized transcripts of play sessions are saved for authors’ benefit” (on the play page itself, which I think is better than today’s situation), and this thread is authors that year talking about how helpful that was.
I’ve not thoroughly researched the history (in this forum, and maybe r*if archives).

(So I appreciate that my previous post may be reinventing an already-much-discussed wheel.)

It’s a sad thing that living in an environment where being spied-upon and scammed is an increasingly pervasive norm opens up even good-faith and previously-unremarked interactions to suspicion. (I say as someone who is super suspicious online.) Commercial surveillance is corrosive to society.

To be clear: I second this.

4 Likes

I’m struck by this conversation because I’ve seen authors laud how IFComp uses transcripts, and I’d assumed that the prevailing community opinion was pro-transcript, and this is the first I’ve heard of the opposing view, and I’m a little surprised, though of course I get it, being a privacy advocate myself. I think there’s a simple solution though - I mean, opt-ins, obviously - but also, just add a visual indicator whenever transcripts are being recorded, like the green light on a laptop camera, is a simple thing.

I’m working on a new parser system; transcripts is on my to-do list specifically because I’ve seen people here treat them as a must-have; but now I’ll be sure to add a constant visual indicator as well as an opt-in. This seems like it should be a relatively easy thing for other folks to work into existing systems that handle transcripts.

8 Likes

On the private forums, yeah, but especially in the defunct euphoria chat room, where many authors used to congregate.

I should say that I’m not really interested in digging up these old conversations and rehashing the discussion for what would be, for me, the umpteenth time. I end up having to excavate my personal life to explain my opinions, and nobody is ever convinced anyway. It’s exhausting!

The summary is simply that “privacy” is one more “reason a player might not want to engage with web-based parser deployments,” so people can take that into consideration – or not!

6 Likes