I feel like a lot of these complaints are endemic to defining things in CSS using 'px' as units instead of more sensible (and repsonsive) modern techniques like defining type in 'pt' and widths in more complex units (percents, or mixed units).
It's possible to avoid a lot of these pitfalls with better design.
1. User far from big screen: You can zoom much faster and easier with your forearm than a pinch gesture
2. User far from screen, unusual viewing angle: how much time to you spend browsing at awkward angles versus dead-on? This seems like a small consideration that would affect some types of sites more than others. I doubt a banking website would feel the same way about this as a social media website project for example…
3. Classroom projection is decidedly not the use-case nearly any mobile website is designed to work under, unless the website is cloud-based projection software, but that's a specialty thing in itself.
4. This is why we define font sizes in PT and no PX, this isn't a mobile or responsive issue, but rather a design choice in CSS
5. Does your mobile browser include a 'reader' mode that isolates text content and provides the ability to make it more legible in these conditions?
6. Ah yes, and what's even more surprising for people is how differently Mobile Safari calculates screen dimensions between iPhone and iPad! Then the difference between how iOS and Android handle keyboards, and then the difference at how those OS's handle browser verus web-app mode. There is currently no good solution to keyboard control.
Anecdotally, we ran into an issue on Friday where an email field was autofocused and the keyboard was popping up and obscuring what you were entering. I didn't make it that way, but I did provide the solution: remove autofocus for touchscreen users: http://staticresource.com/autofocuser.html This is a way that we can have one feature for desktop users where it speeds up entry, but target mobile users separately and provide a better targeted experience for them.
<script>
// Block autofocus for touchscreen users
onload = function(){
if(('ontouchstart' in window)||(navigator.msMaxTouchPoints>0)){
var field = document.querySelectorAll('[autofocus]');
for(var i=0;i<field.length;i++) {
field[i].blur();
}
}
}
</script>
7. I hear you there, I've felt the same way when reading long texts. Any easy workaround would be to allow the user to adjust the font size of the article text on the site within reasonable limits!
8. That's usually a problem with the difference between 'pixels' on your screen and CSS 'px' units. Defining layout in other units in CSS can avoid this problem on HiDPI displays.
9. Testing, testing, testing. I make sure to supply well-made fonts, I use a special set of CSS rules that control the rendering of text display so that the same font renders the same thickness in all browser's and OSes, and with those differences out of the way I can move forward using only weights that work (usually no lower than 400 weight unless the size being displayed is very large) and be sure that they are legible everywhere. Here's my snippet (and honestly, this cuts out SO many other common text rendering bugs):
10. My job is to make sure that the experience of the desktop site, with all of its abilities, functionality, and information, are displayed in an equally recognizable, usable way on mobile. I don't cut features, I don't ever assume a mobile user or desktop user won't want anything. My job is to make it just as great everywhere.
11. Turn your brightness down, you only have one set of eyeballs.
12. I'm not sure panning and zooming will help you feel better when the room is spinning so fast you can't read.
I'm not dismissing these reasons, I believe many of them are small or fleeting and to change an entire project for one of these whims seems short-sighted. Other reasons listed here are 100% valid, and workaround can be found without enabling zooming.
For projects where zooming makes sense I wouldn't have it any other way, but I don't see a compelling case listed here to always have zoom enabled.
> I feel like a lot of these complaints are endemic to defining things in CSS using 'px' as units instead of more sensible (and repsonsive) modern techniques like defining type in 'pt' and widths in more complex units (percents, or mixed units).
What? Isn't px a fixed multiple of pt, with an actual size that varies by platform?
CSS 3 defines pixels that way (1/96 inch, pt is 1/72 inch), but the previous definition was rather different, and had no fixed relation to either inches or ems.
But I thought the change in CSS 3 was because pretty much every actual implementation of CSS 2.1 had taken the definition, chucked it out the window, and used 1/96 inch anyway. So, I don't see how any problems could arise in practice from using px instead of pt.
If the site can render responsively, is there a reason it can't be zoomed?
Both the reasons looking for zoom, and your reasons for not providing it seem reasonable. Is there some technical barrier for not enabling zoom?
Zoom is a first-class gesture on mobile, it is the second one after scrolling that you always expect to be there. On a desktop, you generally have more desktop than you need for a window, so it is possible to resize the window to adjust the size of the viewport. Responsive sites work nicely in the case. On mobile, there is no "window" and the device's dimensions themselves are the viewport. You can reorient the device to enable changing the width, but without zoom, you are taking away any ability to change the viewport.
Maybe zoom isn't a critical feature, but it is part of how the devices work. As a result, it is an expected interaction for those on mobile. Have you ever used a piece of desktop software where you couldn't resize the window - it drives you crazy. Ignoring expected behavior is contrary to good usability.
> Is there some technical barrier for not enabling zoom?
“Safari on iOS displays webpages at a scale that works for most web content originally designed for the desktop. If these default settings don’t work for your webpages, it is highly recommended that you change the settings by configuring the viewport. You especially need to configure the viewport if you are designing webpages specifically for iOS.”
Zooming is enabled by default on Mobile Safari as a feature to help mobile users navigate sites not designed for mobile.
Apple intentionally added many different ways to control the viewport, including the ability to disable zoom.
“Setting the viewport width should be the first task when designing web applications for iOS to avoid the user zooming in before using your application.”
So it very much sounds like Apple has added this functionality as a usability feature to help mobile users navigate sites designed without regard to their specific hardware, Apply strongly recommends setting the viewport specifically for the pages you design with iOS in mind. Also note, when looking through Apple's documentation that they refer to the gesture as 'zooming' and the feature in Mobile Safari as 'user scaling.
OK, so you're suggesting that we not enable a "usability feature" because Apple's guidelines suggest that there are alternative ways to specify the correct viewport?
I get the design angle, and I don't disagree that setting the correct viewport is a good idea, but you shouldn't assume that good design and correctly setting the viewport will work for everybody. The fact remains that the zoom gesture is critical on mobile, and disabling it does not help anybody use your site.
On responsive sites, the idea behind them is that the site does its best guess of being responsive based on the size of the viewport. However, sometimes it doesn't work out, sometimes the websites behavior isn't as intended, sometimes the designer didn't test with my exact device, size, eyesight, etc. It is nice in those cases that I can make up for the site not being designed for my mobile by using the zoom feature as intended.
> K, so you're suggesting that we not enable a "usability feature" because Apple's guidelines suggest that there are alternative ways to specify the correct viewport?
No that's the opposite of what I'm saying. Apple has created a usability feature called 'User Scaling' and enabled it by default for usability purposes. Apple also added the <meta> tags that control the viewport, including disabling User Scaling, and strongly urges designers & developers to do two things:
1) Design your site with iOS users and hardware in mind
2) Use the correct viewport tag that gives the best experience
> On responsive sites, the idea behind them is that the site does its best guess of being responsive based on the size of the viewport.
I'm not sure what you're trying to say here. There's not much guessing going on here, you are either able to define the viewport, or it will render the page at the widths specified in the page styles. If you have defined the viewport there is no guessing involved and it will render the same way from device to device.
> sometimes the websites behavior isn't as intended, sometimes the designer didn't test with my exact device, size, eyesight, etc.
Chances are I didn't test with your exact device, but when I style things I have a set of rules and tools that help ensure that the same content renders the same way anywhere. If I tested only against released hardware then my sites wouldn't be future-proof. I design my stuff to work on all screen sizes smaller and larger than would be in use today so it not only shows up the same everywhere, but will continue to show up the same on devices released years from today.
I also regularly test on quirky hardware like PS Vita, and anything else with a browser that I can get my hands on just to make sure there aren't any weird bugs or edge cases.
Here's a real-world case study: a friend had a sign up funnel and was tracking 900 different mobile devices on his site. On my main gig we have 1,300+ different devices a month! I put my responsive and mobile polish on top of his existing site and we moved the needle on signups 2% since then, and that change has lasted. Clearly, whatever I'm doing it making it easier for mobile users to get in the door with their dollars, and I have the numbers to vouch for that!
> Clearly, whatever I'm doing it making it easier for mobile users to get in the door with their dollars, and I have the numbers to vouch for that!
The flaw in this argument is that you are measuring across the total population, and totally ignoring that while your overall changes may help some proportion, some parts of it may actively prevent others from using the site at all.
Great for you if all you care about is the money. Not so great if you want to e.g. ensure people with disabilities have a good experience too.
It's possible to avoid a lot of these pitfalls with better design.
1. User far from big screen: You can zoom much faster and easier with your forearm than a pinch gesture
2. User far from screen, unusual viewing angle: how much time to you spend browsing at awkward angles versus dead-on? This seems like a small consideration that would affect some types of sites more than others. I doubt a banking website would feel the same way about this as a social media website project for example…
3. Classroom projection is decidedly not the use-case nearly any mobile website is designed to work under, unless the website is cloud-based projection software, but that's a specialty thing in itself.
4. This is why we define font sizes in PT and no PX, this isn't a mobile or responsive issue, but rather a design choice in CSS
5. Does your mobile browser include a 'reader' mode that isolates text content and provides the ability to make it more legible in these conditions?
6. Ah yes, and what's even more surprising for people is how differently Mobile Safari calculates screen dimensions between iPhone and iPad! Then the difference between how iOS and Android handle keyboards, and then the difference at how those OS's handle browser verus web-app mode. There is currently no good solution to keyboard control.
Anecdotally, we ran into an issue on Friday where an email field was autofocused and the keyboard was popping up and obscuring what you were entering. I didn't make it that way, but I did provide the solution: remove autofocus for touchscreen users: http://staticresource.com/autofocuser.html This is a way that we can have one feature for desktop users where it speeds up entry, but target mobile users separately and provide a better targeted experience for them.
7. I hear you there, I've felt the same way when reading long texts. Any easy workaround would be to allow the user to adjust the font size of the article text on the site within reasonable limits!8. That's usually a problem with the difference between 'pixels' on your screen and CSS 'px' units. Defining layout in other units in CSS can avoid this problem on HiDPI displays.
9. Testing, testing, testing. I make sure to supply well-made fonts, I use a special set of CSS rules that control the rendering of text display so that the same font renders the same thickness in all browser's and OSes, and with those differences out of the way I can move forward using only weights that work (usually no lower than 400 weight unless the size being displayed is very large) and be sure that they are legible everywhere. Here's my snippet (and honestly, this cuts out SO many other common text rendering bugs):
10. My job is to make sure that the experience of the desktop site, with all of its abilities, functionality, and information, are displayed in an equally recognizable, usable way on mobile. I don't cut features, I don't ever assume a mobile user or desktop user won't want anything. My job is to make it just as great everywhere.11. Turn your brightness down, you only have one set of eyeballs.
12. I'm not sure panning and zooming will help you feel better when the room is spinning so fast you can't read.
I'm not dismissing these reasons, I believe many of them are small or fleeting and to change an entire project for one of these whims seems short-sighted. Other reasons listed here are 100% valid, and workaround can be found without enabling zooming.
For projects where zooming makes sense I wouldn't have it any other way, but I don't see a compelling case listed here to always have zoom enabled.