Parser game conventions?

Clarification: Convention as in “how things are usually done” not as in “anime convention”.

I’m looking into parser games as a player and possibly as an author, but it feels like there isn’t good introductory material.

  • What kind of sentences do parser games usually use? “verb object”? Can I use filler words like articles? If something is described as “a blue pen” is “blue pen” allowed as an object, or do I have to leave out adjectives?
  • What kinds of verbs are supported? Without even knowing what’s allowed, it’s hard to start.
  • Other things I should know?

Short answer is it depends quite a lot, but a couple hopefully-helpful general points:

  • there are different approaches to this that largely correspond to authoring systems. More “new school” games, which often are written in Inform or TADS, will accept more complex sentences, though articles are usually purely optional; you can refer to objects by their full name, or just the noun, or sometimes just their adjective too (e.g. it’ll recognize BLUE as the blue pen). Some older-school games, often written in Adventuron or ADRIFT, stick to a stricter two-word VERB NOUN approach. But individual authors can change these defaults relatively easily, so sometimes a bit of experimentation is helpful.
  • often games will have a VERBS command or something similar; this introductory postcard also has a lot of the most-common actions. But again, sometimes playing around is needed as authors often introduce new commands or alter existing ones.
  • if it seems helpful to get your feet wet before jumping all the way in, reading transcripts of others playing parser games and seeing what they type in and what works might be something to consider. Club Floyd has a giant archive of these (with player commentary).
  • Have fun and definitely post more if any questions or roadblocks come up!

This is the image Mike linked to, but a high resolution PNG instead.

Something else is giving NPCs commands. Not too many games require this, but I think the convention is addressing the character, adding a comma, and then issuing the command…


Oh, and this could be important. Some games apparently think LOOK is different than EXAMINE or SEARCH, from what I’ve seen talked about here (I don’t have a lot of experience with parsers). Apparently, LOOK is treated like a cursory glance and EXAMINE/SEARCH is treated like a thorough examination that uncovers new information. Not all games do this, I believe, but it’s important to know that it exists. I think they should be treated as the same, for what it’s worth.

(Maybe someone can verify this claim.) Update: Thanks, Mike! Double Update: Thanks, Phil! :slight_smile:


Yeah - look is usually “describe the overall environment”, examine object is usually “look at this closer.” LOOK OBJECT will often be understood as examining, though.


Some games treat EXAMINE as a surface examination, whereas SEARCH is something that is meant to uncover hidden contents. This duality is frowned upon in some quarters (including mine).


@rileypb I would be so pissed if I had examined an object, only to discover later after racking my brain, that I had to search it instead.

You may have just saved many a monitor from being thrown across their respective rooms. This should be a standard public parser service announcement.


There is! It’s just not easy to find if you aren’t already aware of it. Which is ABSOLUTELY RIDICULOUS, especially for a group of people who really want to attract more parser players.

Aside from the helpful card Mike and Hal posted about, there’s a whole competition-- the Text Adventure Literacy Project (TALP) every year for writing introductory games with tutorials, mostly aimed at children, although they can be fun for adults too. This year’s will begin in 10 days, but there are links to previous competitions, and I think you can search using the keyword TALP on IFDB.The idea here is to teach conventions.

I have no idea why IFDB doesn’t list beginner resources and games right up front on the home page. It kinda seems like maybe that should happen if we want people to join our little cult.


In the Book of IF, Crowther 4:56, it clearly states that the mothership can only support 100 human beings for the cryosleep journey back the IF homeworld where (we’ve been repeatedly told that) we will not be used as hosts for our parasitic alien overlords.

I’m sorry, but I’m not giving up my ticket for the spaceship just so you can have more IF buddies to hang out with. Fuck that!



Thank you for posting the link to TALP. I need to add that to my personal website.



Exactly! I read multiple posts here now that talk about making parser games more accessible to new users, but that was geared towards authors of parser games so the ideas were more abstract and general, not concrete like the verb cheatsheet.

You’d think being a Linux user and programmer would make playing parser games easy, but at least shell commands and libraries have manuals and documentation.


We do too, but they’re mostly for the various IF programming languages.

Adding on to what others have said, sometimes beginner-friendly games implement a VERBS command, where it’ll just list all (or most) of the possible verbs in the game. Most will also have a basic response to HELP or HINT (sometimes HELP is general while HINT is contextual, but sometimes they’re synonyms) or at the very least ABOUT.

Personally, I think the best way to learn parser conventions is through playing one (with a walkthrough if you need help). My first (completed) game was Violet, which is nice because it introduces various puzzle mechanics but is contained in one room, so everything you need is right in front of you. I then played Lost Pig, which is longer than Violet and has more difficult/less hinted puzzles.

There are of course numerous other beginner-friendly games. Many people here will suggest The Dreamhold because it does a good job at slowly introducing new features. In my experience, though, it was still too overwhelming. So, You’ve Never Played a Text Adventure Before, Huh? is a pretty good introduction, too, with lots of references. Galatea is not your traditional parser game, and more like an interrogation where you ask an animate statue about a ton of different topics, with different endings. You may also find 77 Verbs helpful for all 77 standard verbs in the Inform [1] library. It is less an introduction to how to play parser games and more to get you familiar with all the verb options (not all of these will be used in all games).

All of the above games are well-hinted and most have walkthroughs as well. And most of them are short (except Dreamhold and Galatea (but this one is basically infinite)). Read the blurbs, pick a few, and try them out. They’ll give you a good idea of what to expect from most parsers.

  1. one of the many programming languages for parser games ↩︎


I guess if we (on this forum) can’t agree on what makes a good tutorial, or what one should include, or what is a good beginner game (e.g. I’m not a beginner and bounced off Dreamhold, but it self-describes as for beginners and everyone else says it worked for them) it makes it equally hard to agree on a list of such games or resources.

Maybe the creation of one requires a combination of nominations and self-nominations. Anything anyone suggests will usually be “right” for a lot of circumstances, unless it’s screamed down. e.g. Graham Nelson arrives and says “Curses”.

If nothing else, IFDB could potentially home the poll Games for Beginners on the front page. I already look at it and go ‘Hm,’ but I should ignore that instinct to place my money in the vicinity of my mouth. Part of that isn’t the overall range of games, but that newer school games that are probably Really Good for beginners are in a poor discovery situation (way down the list or not on it) because of the demographics, duration of the poll and cementing of the older games at the top. The first newest game is Counterfeit Monkey at position 11 (2012). My memory is the intfic -based outreach/accessibility push came around 2010.

But it also depends on – do we want games that are good for beginners who basically know what do do but haven’t done it much, or games that actually tell them how to begin? Or both? Or list them separately?



But we DO generally agree on what belongs in a tutorial: the basic commands. Different authors have gone about this different ways with varying degrees of success in TALP, but mostly no other games do it other than a “New Players/Basic Commands” reading section at the beginning of games. The reason they don’t do this is because writing a tutorial is just a horrible experience. I swear one day I’m going to write a simple tutorial, get it tested extensively, and put it out there for people to stick into their games like an extension. One day when I’ve screwed my balls on tight enough.

But it shouldn’t be too taxing to come up with a ranking of games for beginners that teach players well. TALP is a great place to start. I see people recommending things like Curses or Counterfeit Monkey for beginners and I’m like, what? I’m guessing those people cut their IF teeth in the Age of Cruelty.

Freshman, sophomore, junior, senior. Like a curriculum. A big NEW TO IF? CLICK HERE! link above the fold on the homepage might not go amiss. I personally do not think Dreamhold is a freshman game-- but maybe it’s a sophomore one.


The command syntax for the vast majority of games is:

verb_phrase [noun_phrase] [[preposition] [noun_phrase]]

The verb phrase is a verb (e.g. LOOK) or a verb followed by a preposition (e.g. LOOK AT).

The first noun phrase is usually the direct object. This may include a definite or indefinite article (A, AN, SOME, THE), but most parsers will strip this off in most circumstances. This may optionally be followed by one or more adjectives (e.g. BIG, BLUE) and one or more nouns (e.g. BALL). The looser parsers (such as Inform 6 and 7) will allow the adjectives and nouns to occur in any order and you can even omit the noun and use only an adjective. Hence article, adjective(s) and noun(s) are all optional, but you must provide at least one adjective or noun to uniquely identify the object you’re referring to.

If there are multiple objects in scope (usually meaning able to be seen and/or able to be touched, depending on the verb being used), then you need to provide enough information to identify the object you want or you’ll get a disambiguation question. For example, if there’s a green ball and a blue ball in scope and you say GET BALL, the parser will ask you which one you mean, the green ball or the blue ball and you can respond with BLUE or BLUE BALL without a verb.

The second noun phrase is usually the indirect object and it abides by the same rules as the first noun phrase re article, adjective(s) and noun(s). The preposition separates the direct object and the indirect object. If the preposition is omitted, then the direct and indirect objects are swapped. For example, GIVE AXE TO DWARF and GIVE DWARF AXE mean exactly the same thing, where AXE is the direct object and DWARF is the indirect object.

These are the basics that apply to pretty much every parser-based game and every authoring language. There are more advanced features like referring to all objects using ALL or EVERYTHING, use of multiple objects using AND, excluding certain objects using BUT or EXCEPT, and stringing multiple commands together, but new players don’t need to know all that stuff.

Regarding common verbs, these are pretty much standardised, but the range of verbs varies slightly from language to language, even Inform 6 standard library, Inform 6 PunyInform library and Inform 7 have a slightly different set of default verbs and default grammar.

Each game will be customised by the author to add or delete verbs and modify the grammar so that it all makes more sense within the context of that particular game. Similarly, the default responses to the default verbs can be customised.


Oh, crap, as this year’s organiser, I should have included a plug for Text Adventure Literacy Jam 2024. (Thanks for already doing so @AmandaB.)

Despite the name, this is a competition, not a jam, with some nice prizes. The aim of the competition is to encourage games that are suitable for beginners and include a tutorial. Games become available for playing and judging on 1 May 2024 and I would encourage everyone (not just beginners) to play and rate these games, as people are always asking about beginners’ games and the TALJ has produced some of the best examples of beginners’ games in its first three years. Links to the past competitions are on the home page.

I’ve been testing a few of the games for this year’s competition and I can assure you that you’re in for a treat. It’s also interesting to see how authors have handled the tutorial, as tutorials are friggin’ hard to write and to get right.

This year, we had an optional theme of ‘Fairy tale’. The games I’ve seen so far have been inspired by Japanese mythology, Greek mythology, fairy tales and nursery rhymes. It’s certainly a mixed batch that can be savoured by beginner and expert alike.


I completely agree. My personal go-to game for this is Glowgrass by Nate Cull. Easy-to-medium difficulty puzzles, classic exploration and unlocking scenario, and a cool VR machine.

The IFDB page has a walkthrough by David Welbourn (always a plus), and a ClubFloyd transcript.

Glowgrass - Details (


I can’t remember if Glowgrass actually teaches the player how to play. Does it?

I would tend to think that a short (less than an hour) game, made for adults, with a non-intrusive tutorial, polished until gleaming, with no parser weirdness, would be the best bet (I’ll refrain from my usual wheedling for modernized parsers here. You’re welcome). Perhaps some “lite” versions of some of the best games that would train people and make them want to play the full thing? That’s how many games hook people-- by offering free lite versions that then induce people to want to play the whole thing. Except ours would still be free!

I don’t think that available walkthroughs/transcripts are the best way to hook newcomers. Why should they have to go to a spoiler-y, separate entity just to learn how to play? To my mind, those are resources for players who get stuck, not training tools.

Someone with the oomph to do it should run a Tutorial Comp, with the aim of producing short, tight, interesting tutorial extensions for authors to put in their games and making those freely available.


My question though is, what attracts new people in practice?

I think someone (maybe Aaron Reed) may have done a survey on this, but it could be worth doing again. Some polished games like you indicate do example with tutorials in them, especially Bronze and Dreamhold, and I have heard from several people that those games got them into IF.

On the other hand, I also know several people who got started with (like you said) Curses or Zork and who got hooked because of the challenge.

It would be great to get a survey on what games made people interested in parser games and also to survey those who tried parser games but couldn’t get into them to see what games they tried (I know there’s a group on here in the latter group). Because theorizing is good but data can be useful! Emily Short made many attempts to introduce people to if through short easy games (like When In Rome) that weren’t as popular as her other harder and longer tutorial attempts like Bronze.

The demographics have changed in the last few decades so I don’t think we have to rely on old results, but I do think new data would be useful.

As an anecdote mostly unrelated to the above, things that confused me early on were “inventory” (I had no idea that command would show what in holding), conversation in older games (TALK TO or ASK ABOUT or PERSON, HELLO) and the idea of using ENTER DOOR or GO INSIDE as directions that were possible.


Amen to that. That’s why I proposed a research article in the “What articles would you like to see in The Rosebush” thread. All I have to go on is the anecdotal evidence I have from relentlessly and annoyingly grilling my peeps, who aren’t exactly a representative cross-section of player demographics since most of them, to my great sadness, don’t play any games at all.

Introducing people to anything-- even if it’s done right-- can’t happen without some pretty serious exposure, which means advertising and media coverage by well-known game reviewers. So even if we did hit on a surefire, data-driven way to assimilate folks, how would they know about it?

Edit: …which leads me to another thing I think we should all be thinking about (except @HAL9000, who wants to keep competitors off the mothership): many people, myself included, find and try new games by reading reviewers we like and taking their advice. Why aren’t we courting these people and trying to recruit them to tell their readership about IF?