[naasking]

> Looking at modern day web development we are seeing the opposite, where the program is many times bigger than it's output

You're seriously underestimating the size of output in a web browser. Consider that the CSS, JS and HTML describe the position of every pixel and its motion on the screen.

[dmreedy]

Not just the size of a given output, but the space of possible outputs. I am also of the opinion that webdev is bloated and miserable, but it is still a massive compression of the information it represents.

[titzer]

I don't really buy that acres of whitespace and gradients really describe much information.

Considered at the boundary of your awareness--in that you are mostly unaware of all the pixels on the page, then absolutely not true. A typical news article might be a few KB of text but MBs and MBs of images that you never look at.

[dmreedy]

I'm talking about Shannon Information specifically. In which case, acres of whitespace and gradients very much are information. Calling them "whitespace" and "gradients" is a significant act of compression in and of itself.

[titzer]

I understand your argument, but even then I don't think it's quite right; there is lots of room to improve on the size of a webpage vs its rendered pixels on the screen, even with the fancy fonts and gradients and animations. I still don't believe it's really megabytes. Have you seen 4k demo contests, btw? https://www.youtube.com/watch?v=0w_xEUoK79o

[abelhabel]

Yes, I agree. I also realised it wasn't really accurate of me of comparing an applied theory to the theoretical theory, as the author calls it.