in reply to PM CSS and markup optimizations

That's a pretty short list, and a small sample of pages. What if some of these are used, just not that often? For example, spoiler tags are used much less often than e.g. code and readmore tags. And there are four different ways to present spoilers, so maybe it is only used for one of those. I think it's a good idea, but I'm concerned about the small sample size.

Elda Taluta; Sarks Sark; Ark Arks
My deviantART gallery

Replies are listed 'Best First'.
Re^2: PM CSS and markup optimizations
by kimmel (Scribe) on Jul 12, 2012 at 20:06 UTC

    I ran a new scan, this time with 444 pages instead of just 105 pages. The unused CSS selector count went from 35 to 34. So I am doing a much larger test right now to see if anything else gets picked up.

[reply]

      Isn't this even irrelevant?

      The PM CSS files are far smaller than most websites. In fact I think you'd be hard pressed to visit any website with smaller CSS files. And more importantly, they are cached by the user's browser never to be requested again until the cache expires. So unless you are a very first time visitor to PM, it plays no part at all in the bandwidth of a PM web page load.

      So it's a small chunk of bytes that rarely get transferred anyway. If we were discussing Perl code instead of CSS markup there would have been several pointers about the pitfalls of pre-mature optimization by now.

      Sweating over the size of a CSS file is misdirected effort IMO.

[reply]

        I cannot reasonable create a good CSS design for PM unless I know what ids, classes, and tags are used and allowed. I do this by running a tool over PM which tells me what is used and not used. I do not want to spend the time styling something that is left over cruft that no one uses. The size optimization is a side effect of that program, not the primary goal.

[reply]

In Section Perl Monks Discussion