Hiding images in highlighted text

Originally published at: https://boingboing.net/2019/10/07/the-big-reveal.html

5 Likes

something along the lines of

<style>
.red::selection { background:red; }

.green::selection { background:green; }

...
</style>
4 Likes

So a picture really is worth a thousand words.

Yes, I’m a dad. Why do you ask?

12 Likes

I like this. Yet more data hidden in planning sight.

I’d rather call it inverse-steganography. I like the phrasing, even if it isn’t exactly correct.

See also Cypherella, who offered steagonographic tools for the “price”(free, but it would nice if one would send one) of a toy stegosaurus.

[Edited for content]

2 Likes

Well that was definitely the highlight of my day!

7 Likes

I’ll be impressed when it works responsively as the text reflows. Otherwise this is the CSS equivalent of tediously creating “pixel” art in Excel by changing the background color of cells.

1 Like

Oh, cool, Hooloovoo fan-art.

3 Likes

Getting SSL errors trying to connect to Dan’s domain.

I considered it the equivalent of ASCII art but in the highlights and I remember ASCII art really getting butchered when the text re-flowed. ya I think you right to doubt

1 Like

I’m trying to allow for the possibility that the real build is actually doing something like processing an image into character-sized chunks and mapping it dynamically onto a block of text, but with a video demo that simple it’s hard to determine whether there’s something interesting and reusable going on or if it’s just one-time flash.

(And if you think I’m being a party-pooper now, wait 'til you hear what I have to say about the accessibility of this little nightmare!)

2 Likes

I totally thought of her and Stego, which was my first exposure to steganography. I still have Stego on floppy.

1 Like

You call THAT “creative text encoding”…?

Pff. I present to you… ASCII Doom. Yes, really. It’s actually completely playable, as long as you avoid the entire right side of the difficulty scale (which is nearly impossible).

1 Like

It seems to use characters as pixels. While the effect is interesting, it could be even better if the algorithm was aware of character shape and choose the best character to represent pixels underneath it (for example by using the least squares method with some additional filters).

Being a dad, it is your duty to ask yourself, “Is this really the Antichrist, or am I overreacting again?”

This topic was automatically closed after 5 days. New replies are no longer allowed.