If you’ve never seen this kind of comment on StackOverflow (also this), I’m going to break it down for you, because I’m frustrated.

Where I see this

It happens all the time:

• vi.stackexchange
• piazza (which my CS classes at UNC use)
• github
• class readings—specifically, scanned images of books instead of actual text

Why it bothers me

Images absolutely ruin the value of the text they contain. They prevent me from doing all of the things that I naturally do with text: namely, text processing.

No, I’m not talking about Microsoft Word-esque “processing” (which is more accurately called “typesetting”).

I’m talking about transforming, editing, searching, copy-pasting, googling, analyzing, etc. I’m talking about the power of textual interfaces. I’m talking about your images ruining my ability to help you or read for your class.

Have you ever noticed that most programs you deal with on a daily basis are text-driven, even if they display images to you?

You wouldn’t send me a screenshot of your essay to read and edit, would you? No?

Then stop sending me images of anything else you want me to read!

Things I can’t do in an image that I could do if it were text

• edit the words
• search for a quote
  • have you ever tried to ⌘-f a PDF that’s actually just a scanned image of the text you’re supposed to be citing?
• copy-paste your error message into google to see if it’s a known issue
• ask a program to read the text to me
  • what if I’m blind?

Other problems with images

There are a number:

1. Foreground and background colors affect readability in a big way (and I’m not even colorblind…).
2. Resolution is wonky. Have you ever considered that between the actual typeset text of a book and the scanned PDF on my hard drive, there are no less than 4 stages of transformation? The scanner, the compression on upload, online storage, download, and then finally my actual display? What happens if I zoom the image? Proper text would be electronically scalable, like a vector graphic, meaning that I can read the text comfortably (without visible aliasing or pixelization).
3. Sometimes, your image is just blurry to begin with.
4. OCR is hard, and I don’t have the time to do it on your cruddy images.

Please stop

For the professors assigning readings: please find actual electronic copies of the text, if we’ve got to read digital things.

For the students and others who need help: please copy-paste your code or error messages.

You’ve been warned.

Post-scriptum

Images are O.K. Sometimes.

But only if the what you’re describing is visual!

That could be color issues, or a painting relevant to the reading, or what have you. But please make sure the content stands on its own, without the image in the worst case. Images are hard.

Categories: Blog

Previous Next

Back to posts