Design of Hyperlinks

I’ve been converting one of my parser games into hypertext (Squiffy). Are there any guides on how to style a hypertext game? I’m interested in issues like those mentioned below.

Consider this example:

John studies the television. He hears something in the kitchen and leaves the room.

Emily wanders into the living room and sees you on the couch watching the television. “What’s on,” she asks, sitting beside you.

Let’s assume the nouns are to be linked, and selecting one will trigger its description.

  1. Does it matter if the hyperlink is a proper name, like John, or the pronoun, like he? It seems to make more sense to link the name, to reduce ambiguity.

  2. Should the television be linked in the first mention of it or last? When possible, is it better to group links at the beginning of a passage or at the end? Should links be grouped at all?

  3. Is there any reason not to hyperlink Emily in the tag “Emily says?” It does seem counter intuitive to me, and ambiguous. Some people may think you want to engage her in conversation.

Another question: Should hyperlinked text be a certain number of characters/words long so that they are easier to find?

And number 3 above is one thing that I think a lot of hyperlink games could improve upon. Clearly state what your links do. If “television” in the example is linked and clicking it turns it off, then suggest that in the text, otherwise I think it’s reasonable to assume clicking it will show a description of it and nothing else. I think a good idea would be to describe at the beginning of the game how the links work, i.e., clicking a noun describes it, clicking an action performs said action.


This doesn’t answer your main questions, but… Have you tried the recent game Thanksgiving by Hannah Powell-Smith? She uses different colors for expanding text, cycling text, and branching text, and I found it really useful.

When you make some word a hyperlink, you make this word stand out. It becomes an important word. So that’s the first rule to consider: what words are really important here?

Consider this example.

See what I did there? I put different emphasis on the characters and the objects (television). It implies these links are not the same, so character links might start a conversation. If you use the same style, it’s the first hint that these links are the same.

That’s why you need a “tutorial” section to show them that these links do NOT engage the characters. The UI is never 100% intuitive. Clicking a hyperlink is easy, much easier than typing a command, so user can quickly learn your UI by trial and error.

Why not both? You don’t have a hyperlink limit.

One reason not to link both is to avoid confusing the player about whether there are actually two different things they can do with the television. If both mentions are links, then the player doesn’t know whether they go to different places–unless the links both gray out when one is clicked, or something like that. (Are solutions involving different colors/formats/graying out accessible to screen readers? I guess the screen reader has to have some way of distinguishing linked words from non-linked words; does that extend to blue links versus red links?)

I’d just like to point out that, personally, if I see too many keywords - especially in cases like Oreolek’s example - my eyes get really tired and really bored really quickly, and I start looking for a way to turn the keywords off.

I thought the fewest links possible would be the best approach. Screen readers won’t, by default, pick up any changes in format, like color changes. They also don’t read out whether a link has been followed.


Neil, if a link in the middle of ambiguous text had a key label at the front of it -

eg “Jane says we should (A) open the skylight. I’m not crazy about her idea. I’d rather go outside.”

do you think it benefits screen reader users to have a closing label as well, to show where the link ends? This could be something like (/A), or ‘(end of A)’.

In the above example, you could put one after ‘(A) open the skylight (/A)’ to show that the action is actually ‘open the skylight’, and that there’s actually no go outside option – that the going outside part is just the character’s thought preference. This isn’t the most brilliantly written example but I still think it conveys my point.


In many hyperlinked games, most, perhaps all, screen readers say “link” when it finds one and reads it on its own line. Squiffy games don’t do the former, but they do the latter. For these, I don’t think any kind of tagging is necessary for screen readers.

For some games, unfortunately, links are indistinguishable from regular text. I think there was at least one hyperlinked game like this in this year’s IF Comp. In this scenario, tagging text for screen readers would be helpful. Tagging would also be a good idea if the meaning of the text depended upon font format, like colors.

I think the linked text need only be tagged at the start – no closing tag is necessary.

I do think that it can be important for authors to consider how a game would be interpreted if the player couldn’t see the screen. I remember being very confused with The Insect Massacre. I didn’t know that the dialogue was color-coded. I hesitate to rate or review games for fear of missing something important because I use a screen reader. I’m not sure whether it would be fair to rate a game low because it isn’t accessible.


Thanks for the ideas.

Btw, The Insect Massacre subtracted accessibility from everyone because the text took so bloody long to appear.