It would be very nice if there were an option to convert all pixel font-sizes set with CSS to point sizes. Using 12px or so to specify a default size for a document is becoming commonplace (bestbuy.com, netscape.com, go.com, hotmail.com, zdnet.com). At anything above ~120dpi, small px font sizes are nearly invisible. It would be nice for those of us with high dpi displays, which will likely be dominant by the time Mozilla is released, to be able to read text without scaling every font on every page, most of which are just fine thanks to relative sizes. The URL for this bug is just a long-winded argument for such a feature. Not necessary to read.

->gerv?

No :-) This would need backend support which I am not qualified to provide. rbs is in the area at the moment. rbs? Gerv

Dup of bug 30910? If so, my font changes in bug 74186 has a fix for this.

I'm not sure I'd call it a duplicate. Maybe an alternate solution to the same problem. One issue is that authors may use different sizes to illustrate various level headings. 14px for top level, then 12 and 10. I'd select 14px for minimum size and lose all hierarchical information. Whereas 8pt is probably acceptable and above my minimum pixel size (at high dpi).

OK I see what you mean. All the computations are done internally in a common base unit of reference called "twips". So 'pt', 'px', 'mm', etc, are converted internally to "twips". And there are functions for PixelsToTwips and TwipsToPixels conversions, using a factor derived by the system from the dpi setting. So what you are asking is a configurable factor for the 'px' case to override the factor would otherwise be computed from the dpi setting. Nice to have, but not something that I can look at as I have other things in my plate. Adding 'helpwanted' if that could make someone pick up the torch from the details I gave above.

This isn't one of mine :-) Gerv

Note that CSS "px" are not the same as screen "px". This should be fixed by a proper implementation of the CSS "px" unit.

Is this still an issue now that bug 177805 is gone?

While there are things we could do to improve users' ability to deal with poorly designed Web sites, I don't think the proposal in this bug is where we should focus our energy. I think there are other approaches that are likely to help users more by being less destructive to the appearance and behavior of the Web sites they're trying to use. So I don't think this approach is the way forward; I think if we want to do work here it should probably begin by studying problems that users have.