View-->text size: Control+/- : *Great* design choice

Blake · Post by **Blake** » November 22nd, 2002, 6:47 pm

Can you file a bug on hyatt about Ctrl+0 selecting the tenth tab and target to 0.6, please?

dean · Post by **dean** » November 22nd, 2002, 10:17 pm

ugggf · Post by **ugggf** » November 23rd, 2002, 2:37 am

Also ctrl+[mouse wheel] does the zoom in/out (if it happens to use it when you want to zoom). 10th tab bug exists only with keyboard zoom - not with mouse wheel.

:idea:

How about a full page zoom in/out feature? (both text and frames, images etc.) Perhaps with ctrl+shft+[+]/[-]/[0]...

laszlo · Post by **laszlo** » November 23rd, 2002, 8:54 am

ugggf wrote::idea:
How about a full page zoom in/out feature? (both text and frames, images etc.) Perhaps with ctrl+shft+[+]/[-]/[0]...

That would be Bug 4821 - Full zooming not functional (objects as well as text).

spaceman · Post by **spaceman** » December 9th, 2002, 10:32 pm

If I define the size of a text element in CSS to be in pixels, eg:

this text is 10 pixels in size

... then is it not correct that decreasing and increasing the text display size should _not_ touch the size of the text (in the above example)?

My copy of Phoenix 0.5 (WinXP) _does_ affect this text size (incorrectly IMHO), along with all the other text on the page (which is correct).

The visually impaired should be allowed to increase/decrease the text size for most or all of the text on a site, but my understanding is that the px size is useful for locking the text size for specific sections of text where changing the size of that text might 'break' the design (eg. in a text menu system).

Comments?

stephen · Post by **stephen** » December 10th, 2002, 5:35 am

spaceman wrote:If I define the size of a text element in CSS to be in pixels, eg:

this text is 10 pixels in size

... then is it not correct that decreasing and increasing the text display size should _not_ touch the size of the text (in the above example)?

My copy of Phoenix 0.5 (WinXP) _does_ affect this text size (incorrectly IMHO), along with all the other text on the page (which is correct).

The visually impaired should be allowed to increase/decrease the text size for most or all of the text on a site, but my understanding is that the px size is useful for locking the text size for specific sections of text where changing the size of that text might 'break' the design (eg. in a text menu system).

Comments?

If you're tying your design into the markup in this way, you deserve to have it break

<a href="http://www.w3.org/TR/WAI-WEBCONTENT/#gl-structure-presentation"> Use markup and style sheets and do so properly.</a>

Stefan · Post by **Stefan** » December 10th, 2002, 6:01 am

spaceman wrote:My copy of Phoenix 0.5 (WinXP) _does_ affect this text size (incorrectly IMHO), along with all the other text on the page (which is correct).

No, you incorrectly THINK that specifying font-size in Pixels should make them unchangable becuse that is how IE does it. That is a BUG in IE and has absolutely no support in any HTML or CSS spec.

On the contrary, CSS is designed to leave the USER in charge, not a random possibly unconsiderate webdeveloper. If someone is trying to make the text on a site larger, they probably have a very good reason to do so and what you think about it doesn't really matter.

spaceman · Post by **spaceman** » December 10th, 2002, 6:10 pm

Stefan wrote:No, you incorrectly THINK that specifying font-size in Pixels should make them unchangable becuse that is how IE does it. That is a BUG in IE and has absolutely no support in any HTML or CSS spec.

Fair enough. That's all I needed to know. I was using the example just to illuminate my question, not as an example of great CSS usage.

That said, IMHO, a web developer should have the _option_ to use a font that cannot be resized by user. Don't get me wrong, I'm a usability 'nazi' as far as many of my associates are concerned, but having a rule that says that all text font sizes on a web page should be infinitely scalable up or down by the user doesn't suddenly force us into being better web developers or necessarily more considerate to every single last user. There is a train of thought that developers should be allowed - if they choose to do so - to have their pages viewed they were intended to be viewed.

Like it or not, most web pages are a blend of text _and_ graphics. The 'message' a web site is trying to convey is a combination of both - one illuminates the other. If the user can upsize or downsize text - and the graphics don't upsize or downsize by the same proportion - then web developers are inevitably going to have a battle on their hands (which translates into longer development times and therefore less affordable web sites) in order to develop sites which can maintain the clarity of their message in an environment where the text font size is an 'x' factor. So, if ctrl +/- was actually a zoom feature (where text _and_ graphics were enlarged/reduced together), I'd be a bigger fan.

Sorry - went off a bit there. :roll:

This issues has rumbled around the web community since the dawn of web browsers. For example, check out the article 'Give Me Pixels or Give Me Death' over at http://www.alistapart.com/stories/fear4/. Perhaps the advice contained therein is a little dated, but the ability for px sizes to be resized by the user by Phoenix (albeit standards compliant, which I support wholeheartedly), may be a little 'utopian' in terms of practical web design at this exact moment in time. But hey, someone has to set a good example.

Thumper · Post by **Thumper** » December 10th, 2002, 11:08 pm

Having the option to make a web-page readable and unpretty as opposed to beautiful and impossible to read is essential. Any browser which doesn't allow zooming of text regardless of hardcoded values is making a big mistake. Hell, I've got perfectly fine close vision and I still have to zoom for those wonderful arty sites which consist of 1px high dark purple text on a black backround, designed for 2000x1600 resolution.

I can't see a single argument that allowing the page designer to completely bar text zooming has any befefits at all.

- Chris

Stefan · Post by **Stefan** » December 11th, 2002, 12:18 am

spaceman wrote:most web pages are a blend of text _and_ graphics. The 'message' a web site is trying to convey is a combination of both - one illuminates the other. If the user can upsize or downsize text - and the graphics don't upsize or downsize by the same proportion - then web developers are inevitably going to have a battle on their hands (which translates into longer development times and therefore less affordable web sites) in order to develop sites which can maintain the clarity of their message in an environment where the text font size is an 'x' factor. So, if ctrl +/- was actually a zoom feature (where text _and_ graphics were enlarged/reduced together), I'd be a bigger fan.

I'm not sure I agree. I would assume that a user would firstly set their screen resolution to something that works for them on their specific hardware platform. I also always leave the main text of a webpage at {font-size:100%}.
With that scenario I belive it will be quite rare to zoom text to other sizes then somewhere in the 75%-150%, or in extreeme cases 50%-200%.
If your page falls appart in such a scenario, there is something seriously wrong with your design approach IMO.

spaceman · Post by **spaceman** » December 11th, 2002, 12:18 am

thumperward wrote:Having the option to make a web-page readable and unpretty as opposed to beautiful and impossible to read is essential.

Couldn't agree with you more. The keyword there is 'option'.

Hell, I've got perfectly fine close vision and I still have to zoom for those wonderful arty sites which consist of 1px high dark purple text on a black backround, designed for 2000x1600 resolution.

That is a design error - not a browser/standards issue. Some designers are good, some designers are bad, and that's life.

I can't see a single argument that allowing the page designer to completely bar text zooming has any befefits at all.

I'm not advocated that. I'm saying that if a designer wants to exercise the *option* to lock certain specified sections of text, then this *option* should not be denied to him/her. If said designer chooses to exercise that choice poorly, that's their freedom of expression to do so.

I should point out that I'm being a bit of a devil's advocate here in that I'm a programmer first and a designer a very distant second. It would make my life a whole lot easier if I only had to worry about text layouts and compliant CSS. But reality tells me that web browsing should be allowed to be rich experience, in terms of both graphics AND text, without the less-creative usability nuts jumping up and down screaming "what about the blind people!?!". Take that argument to it's logical conclusion, and we'd live in black and white world of text.

But now I've really gone WAY off topic. Sorry everyone! I don't believe there's a strict 'right' or 'wrong' here - just some glorious, colourful shades of opinion. :wink:

lar3ry · Post by **lar3ry** » December 11th, 2002, 12:28 am

asa wrote:<code>javascript:factor=Math.sqrt(2); if(!window.scale) { scale=1; zW=[]; zH=[]; unitless=/^[0-9.]+$/; function r(N) { w=N.width; h=N.height; if (unitless.test(w)) zW.push([N,w]); if (unitless.test(h)) zH.push([N,h]); var C=N.childNodes,i; for (i=0;i<C.length;++i) r(C[i]); } r(document.body); } scale*=factor; for(i in zW) zW[i][0].width=zW[i][1]*scale; for(i in zH) zH[i][0].height = zH[i][1]*scale; [].v</code>

Here's a bookmarklet that Jesse wrote for zooming pixel values on a page. It scales images, tables, anything with hardcoded sizing.

Just tried it on THIS page, and it zoomed the images (avatars) correctly, but the text didn't scale. Ctrl-Plus DID scale the text size. Strange.

Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3a) Gecko/20021210 Phoenix/0.5)
(built from CVS with Xft, if that has anything to do with it).

Jesse · Post by **Jesse** » December 11th, 2002, 10:44 pm

lar3ry wrote:
asa wrote:<code>javascript:factor=Math.sqrt(2); if(!window.scale) { scale=1; zW=[]; zH=[]; unitless=/^[0-9.]+$/; function r(N) { w=N.width; h=N.height; if (unitless.test(w)) zW.push([N,w]); if (unitless.test(h)) zH.push([N,h]); var C=N.childNodes,i; for (i=0;i<C.length;++i) r(C[i]); } r(document.body); } scale*=factor; for(i in zW) zW[i][0].width=zW[i][1]*scale; for(i in zH) zH[i][0].height = zH[i][1]*scale; [].v</code>

Here's a bookmarklet that Jesse wrote for zooming pixel values on a page. It scales images, tables, anything with hardcoded sizing.

Just tried it on THIS page, and it zoomed the images (avatars) correctly, but the text didn't scale. Ctrl-Plus DID scale the text size. Strange.

The "zoom layout" bookmarklet does not alter text sizes. It's hard to zoom text with a bookmarklet, and the browser already handles text zooming perfectly well :) Also, images scale much better at integer ratios, whereas the most readable text size is often somewhere between 100% and 200%. When I wrote the bookmarklet, I was trying to make a way to fix the layout of sites that didn't handle text zooming well.

jsalmeron · Post by **jsalmeron** » December 12th, 2002, 3:01 am

what i really like is the CONTROL-mouse wheel up/down approach to do this, ala ie

Magus · Post by **Magus** » December 12th, 2002, 4:33 am

One thing I miss from Mozilla is seeing my current zoom % in the menu with Zoom In/Out

View-->text size: Control+/- : Great design choice

bug 181586

Re: Great feature, but could be better

Re: Great feature, but could be better