The Double Edged Sword of the Web
ponyfoo.com
ponyfoo.com
It's especially unnerving for me as a firefox user since these days I come across a lot more sites that make stupid mistakes that just works fine in their Chrome.
This mindset makes me angry. The same applies when developers think that if your browser blocks google tracking it is ok to just fuck up the whole website.
The web should be
- progressive
- standards-compliant.
Some years ago this was widely believed to be right way forward. Nowadays it seems that developers think that Chrome on OS X is the world we all should live in.
Loosely related: I am not going to switch to Chrome. I would never use a browser from an Ad Company like Google, as I don't trust them at all.
On top of that he disabled accessibility on Firefox entirely in favour of design preferences. Many developer might not care, but it's an attitude which needs to stop.
> As such, visitors are mostly privileged web developers using macOS and Google Chrome.
There's a lot of this going on in the tech community. Far cry from when <geeks|nerds|smart people> got pushed around... Now they're (we? I hope I'm not like this) the ones doing the pushing.
No. There's a lot of this going on in Silicon Valley and among its camp followers.
A lot of the hate comes from having to deal with the later group, not the former.
Besides that, Windows 10's Bash on Ubuntu on Windows is a very interesting step in this direction. It's going to increasingly be a "real" option.
ETA: Also, Virtual Machines and Remote Login are blurry edges of "real" that aren't dual-booting.
Lets be honest, Cygwin is a joke if you're doing *nix development.
Windows 10's Bash on Ubuntu is months old, we're talking historically on the order of 5-10 years.
VMs and remote systems are fraught with compromises.
Additionally he refers to it as macOS which pedantic and yet incorrect unless the majority of his users are running a beta OS.
Using the term macOS is fine. It’s the new name, and we can start using it before the official release of the next version in the fall. I think you are being pedantic here.
> > As such, visitors are mostly privileged web developers using macOS and Google Chrome.
Yeah, mine too. I wouldn't think of using macOS & Chrome as a privilege. A cost of doing business, perhaps, but not a privilege.
I'm privileged to have a job in which I'm running Linux and Firefox, and I really do count that as a privilege.
That said, I think what the parent comment is saying is that Chrome too has bugs and a web developer who exclusively tests in Chrome forgives those bugs as "just the way HTML rendering works." This makes the developer effectively blind to Chrome's defects.
A developer who uses Firefox exclusively will have a similar feeling of privileged disdain for Chrome's bugs: "How can Google ship a browser with this problem? This is unacceptably broken!"
I know I felt that way for ages with Chrome's font rendering [1]. Chrome's font rendering was so poor compared to Firefox that it was difficult for me to use even for short bursts of testing. Although admittedly—and in a way confirming my point—those who used Chrome regularly were conditioned to its rendering and in some cases didn't even perceive the difference when comparing them side-by-side.
[1] https://bugs.chromium.org/p/chromium/issues/detail?id=25541
I don't know about its bugs, but that's pretty much how I feel about Chrome's address bar when I'm testing apps in it :)
From what I’ve understood, the thesis is about the issue of browser inconsistencies, and how it sometimes may be sensible to resort to user agent detection and hacks.
Maybe if the site had a cleaner, simpler design it would be easier to make it work on all browsers.
It took me finding the bug on the Chrome bug tracker to convince my team that Firefox was rendering things correctly, but because we code/design on Chrome it "looked wrong" because it had been designed to look right in Chrome, which was actually wrong. I can no longer remember the exact bug it was unfortunately - had something to do with positioning elements and flexbox though.
Oh - and font rendering. Fonts are always "too thick" in Firefox because Chrome has a poor implementation of font rendering since like Version 21 (22?) and renders fonts thinner than they should be. :)
Also someone who will never switch to Chrome, but for different reasons. I hack the UI and UX of Firefox to be almost unrecognizable as Firefox and Chrome doesn't allow for that level of freedom. There is no userChrome.css to modify, addons are more limited in what they can modify, etc.
If a large majority of your visitors use chrome, you are going to have to prioritize their experience. While ideally you will make it work perfectly for every browser, sometimes you have to make compromises.
It isn't about thinking that Chrome is the world we all SHOULD live in, it is trying to design a web page for the world we DO live in.
No way - The only way we'll end up with compatibility across browsers is to have unambiguous (as much as possible) specifications that define how they should behave and for browser users to insist their browser vendor aligns with the standard. I lived through that past and have no desire to repeat it.
If you think about it from a network theory perspective, each of the browsers is only trying to be compatible with one other party (the specification) in this mode. The alternative is that each browser is trying to be compatible with "n - 1" other parties where n is the number of browsers in the pool. And the total effort is the classic (n * (n - 1)) / 2 "connections", so the browser industry would be wasting a lot of effort.
Worse yet, there's no single release point. Consider that browser A makes changes to be compatible with browser B (because that behavior seems correct compared to browser C). Unbeknownst to them, browser B is about to change their behavior to be compatible with browser C. Now the A-team's work is either wasted or yet another incompatibility.
Browser-focused standards are the best thing that happened to the web in the last 10+ years.
- http://html5readiness.com/ (now a bit dated)
Sadly, it's usually the predominant browser (i.e. Chrome these days) that they're trying to maintain compatibility with and not the specification.
I saw that quote too, and thought that's a weird way to look at it. The other way to look at it, which makes quite a lot more sense to me, is that Chrome is creating the inconsistency by being flat out wrong...
The article text on the far left, and the side column on the far right with a LOT of white space between. The white space in the middle is as wide as the article text itself.
BTW, I'm on a 2560 width screen, using Firefox.
And this is what it looks like:
http://i.imgur.com/bWwMuNH.png
If this is considered good web design, I'm going back to Lynx ...
Jokes aside, virtually no sites are tuned for large displays, and there's widespread consensus that you shouldn't simply use the full width of the window for text because lines would be too wide to read comfortably. Still, I think my expectation is that most sites just end up centered within large windows.
The layout seen here, with columns anchored on the left and right sides, is unfamiliar and looks weird to my eye as well.
Most good sites, I think, (strangely not Hacker News though) have a width limiting, centered layout which makes it easy to read and browse.
I find it strange how many sites seem to be more horizontal gutters than content space these days due their fixed width containers recreating a square virtual monitor inside my widescreen monitor. Admittedly it is good for ad space, I guess, because those gutters can be and often are filled with plenty of ads.
let's see. they blatantly ignore most browsers to focus on the one they use. and then cames up with excuses ("its over 50% of my visitors") just like IE6.
then they get confortable with things that browser does that are proprietary, just like with IE6.
then they blame all the other browsers in the word for not dropping everything else and go imitate the proprietary behaviour. they even points out that chrome is the only one not following the standard, and still gets praised, just like IE6.
"What should be important is that browsers put consistency across themselves first" is what they say on the article about other browsers not ignoring standards and following the proprietary browser they like.
google arrogance over standards and this developer lack of memory and abundance of hypocrisy will give us IE6 all over again.
damn you all
> Humans will seldom use different browsers. Unless there are gross differences across browsers, like using entirely different sets of font faces, humans are not going to care. We need to learn to let go.
Sometimes, the cost/benefit calculation comes out so squarely in the "let's just not" column that even though it may feel wrong to discriminate so blatantly, the economics are just not there. Important to consider whether this discrimination amounts to technical debt which proves costly over time, or whether they are in fact tailored experiences that makes sense to keep around. In any event I agree with the authors sentiment, that chasing the identical look everywhere unicorn often just isn't worth the bother.
At the limit of this kind of "logic" I could equally say that 100% of google search visitors use TorBrowser (out of 1: me in a very precise time frame).