6 ms·

Then do as I: develop in Firefox and if it works there (and isn't a PWA where maybe you get in trouble with Safari?) then it works everywhere. Less testing, le

Then do as I: develop in Firefox and if it works there (and isn't a PWA where maybe you get in trouble with Safari?) then it works everywhere. Less testing, less bugs. Whats not to like? Contrast to Chrome first developers who often get caught by cross browsers incompatibilities just like they did back in the days when they were IE first developers : )

fabiospampinato 5y ago
Unless I'm missing something that doesn't fixes the issue, the problem is symmetrical, browser A is different than browser B, so browser B is different than browser A, testing in either of them doesn't guarantee a correct output in both of them.
- skinkestek 5y ago
  Chrome - like IE before it - has a number of "features" that only work/ed in Chrome/IE. Writing for a standards compliant browser like Firefox makes your code more well defined today just as it did back then because you won't get away with the same sloppyness. (This also makes you catch and fix problems in early iterations over the problem instead of after QA calls to complain so it saves you time and context switching too and if you are good you might look like a cross browser superhero almost for free ;-) Earlier on not every basic thing was supported everywhere, many people here will remember the ACID tests. Younger devs won't remember them as we stopped talking about them after every browser became compliant. Today every mainstream browser has comprehensive test suites to cover everything we need from CSS I think.
  - fabiospampinato 5y ago
    Testing on ~~Chrome~~ Firefox would save you from accidentally using Chrome-only features, but that's only part of the problem, caniuse.com kinda works better for that as you get data about other browsers too.
    - skinkestek 5y ago
      >Testing on Chrome would save you from accidentally using Chrome-only features, but that's only part of the problem, How? > caniuse.com kinda works better for that as you get data about other browsers too. caniuse.com is nice but unlike using a standards compliant browser it requires you to be mentally alert and aware of it. Using a standards compliant browser means you'll see the result of sloppy css immediately on the screen.
      - fabiospampinato 5y ago
        >>Testing on Chrome would save you from accidentally using Chrome-only features, but that's only part of the problem, >How? Sorry I meant "Firefox" and "Chrome-only" there. In general though it's not like Firefox is spec compliant and others aren't, pretty much every browser works a little different in some areas. Just for the sake of saying something verifiable: no browser is spec compliant because the spec mandates a precise maximum length for strings and each main browser has a different, arbitrary (= I have the RAM, they just won't let me use it), lower limit on that.
pineconewarrior 5y ago
I do this too, but still run into issues with Safari. Luckily, later Safari 14 versions and Safari 15 are not too bad.
spoiler 5y ago
> Contrast to Chrome first developers who often get caught by cross browsers incompatibilities just like they did back in the days when they were IE first developers : ) I'm not entirely sure how I feel about this statement (and comparison to IE). If anything, Safari is "the new IE" rather than Chrome. MOST (hopefully the nitpickers pick up the caps lock) stuff in Chrome are drafts or standards. Sure, Google pioneered/championed some of them, but that's kinda irrelevant if developers voted for those features. There's very little Chrome-specific stuff. Also, other browsers have vendor specific stuff in them too, yet people rarely fling shit in their direction? FWIW I also mostly use Firefox for dev because I prefer how some devtool features are designed/implemented. Most of my cross browser issues in FF were "brief" in the senses that they were bugs that got fixed eventual.
- skinkestek 5y ago
  > I'm not entirely sure how I feel about this statement (and comparison to IE). If anything, Safari is "the new IE" rather than Chrome. Some people who either don't know history or willfully ignores it keeps claiming that Safari is the new IE, at one point one even made a webside out of it. Don't fall for it. Chrome is the new IE: - technologically advanced? Check! - implement a number of things without asking or waiting for consensus? Check! - will be abandoned as soon as they have crushed every competitor? Well, it is produced by the worlds most famous company when it comes to killing its own software.
  - bananabreakfast 5y ago
    I can't believe you seriously just suggested Google is going to abandon Chrome. Next up, Microsoft is going to abandon Word. Also, Facebook is going to abandon Facebook. This is why no one takes these conversations seriously. All vitriol, no substance.
    - danielheath 5y ago
      IE still had released, but it stopped moving forward after 6 because there was no longer a good reason for ms to invest. I doubt google would spend much on chrome if no other browser were popular.
jitl 5y ago
There’s plenty of weird cruft in Firefox’s ~30 year old codebase that causes bugs and unspecified behavior. For example, Firefox’s ContentEditable code deletes things “backwards” compared to all other browsers and operating systems (bugzilla: https://bugzilla.mozilla.org/show_bug.cgi?id=1735608 https://bugzilla.mozilla.org/show_bug.cgi?id=1735608) - this behavior isn’t standardized in any spec, and Firefox is different from other browsers. There are plenty of gotchas in layout/rendering as well, where either the standard is under specified, or Firefox has some small bugs. Maybe Chrome has many-chrome only APIs, but the developer will always need to test in Chrome and iOS Safari 13 (or whatever your oldest supported iOS version is). Chrome and iOS are where the users are, and a good website or app should be usable and beautiful for everyone.
- dbrgn 5y ago
  Developing UI components based on ContentEditable is trial and error, with a lot of code held together by cross-browser integration tests where you always have to pray that a browser upgrade doesn't brake something new in a subtle and surprising way... 1/5, would not recommend. (But sometimes there's no way around it.)
- TheGoddessInari 5y ago
  > There’s plenty of weird cruft in Firefox’s ~30 year old codebase that causes bugs and unspecified behavior. Just wondering where you're getting "30 years" from? That asked, a Chrome-shaped monoculture doesn't help anybody. We need more competing implementations, not less. Anyone feel like collaborating on such?
  - hiptobecubic 5y ago
    Are you casually asking if anyone wants to start writing a rendering engine?
  - jitl 5y ago
    Netscape was founded in 1994, Firefox is derived from Gecko (1997) layout code code open-sourced from Netscape in 1998. See https://en.m.wikipedia.org/wiki/Gecko_(software) https://en.m.wikipedia.org/wiki/Gecko_(software) for some background. I was thinking 1994 when I wrote the comment, but it’s closer to ~25 years if you reckon from Gecko’s birth year of 1997.
nightpool 5y ago
This is absolutely not true. I wrote a simple flexbox layout in Firefox last year and it was majorly broken when I tried to test it on Chrome, since Firefox and Chrome disagreed on how to compute "min-width" for inline images and other replaced content. In this case, my layout worked in Firefox, but Chrome was the browser that was following the CSS standard correctly, and my code was broken.

fabiospampinato 5y ago

Unless I'm missing something that doesn't fixes the issue, the problem is symmetrical, browser A is different than browser B, so browser B is different than browser A, testing in either of them doesn't guarantee a correct output in both of them.

skinkestek 5y ago
Chrome - like IE before it - has a number of "features" that only work/ed in Chrome/IE. Writing for a standards compliant browser like Firefox makes your code more well defined today just as it did back then because you won't get away with the same sloppyness. (This also makes you catch and fix problems in early iterations over the problem instead of after QA calls to complain so it saves you time and context switching too and if you are good you might look like a cross browser superhero almost for free ;-) Earlier on not every basic thing was supported everywhere, many people here will remember the ACID tests. Younger devs won't remember them as we stopped talking about them after every browser became compliant. Today every mainstream browser has comprehensive test suites to cover everything we need from CSS I think.
- fabiospampinato 5y ago
  Testing on ~~Chrome~~ Firefox would save you from accidentally using Chrome-only features, but that's only part of the problem, caniuse.com kinda works better for that as you get data about other browsers too.
  - skinkestek 5y ago
    >Testing on Chrome would save you from accidentally using Chrome-only features, but that's only part of the problem, How? > caniuse.com kinda works better for that as you get data about other browsers too. caniuse.com is nice but unlike using a standards compliant browser it requires you to be mentally alert and aware of it. Using a standards compliant browser means you'll see the result of sloppy css immediately on the screen.
    - fabiospampinato 5y ago
      >>Testing on Chrome would save you from accidentally using Chrome-only features, but that's only part of the problem, >How? Sorry I meant "Firefox" and "Chrome-only" there. In general though it's not like Firefox is spec compliant and others aren't, pretty much every browser works a little different in some areas. Just for the sake of saying something verifiable: no browser is spec compliant because the spec mandates a precise maximum length for strings and each main browser has a different, arbitrary (= I have the RAM, they just won't let me use it), lower limit on that.

pineconewarrior 5y ago

I do this too, but still run into issues with Safari. Luckily, later Safari 14 versions and Safari 15 are not too bad.

spoiler 5y ago

> Contrast to Chrome first developers who often get caught by cross browsers incompatibilities just like they did back in the days when they were IE first developers : ) I'm not entirely sure how I feel about this statement (and comparison to IE). If anything, Safari is "the new IE" rather than Chrome. MOST (hopefully the nitpickers pick up the caps lock) stuff in Chrome are drafts or standards. Sure, Google pioneered/championed some of them, but that's kinda irrelevant if developers voted for those features. There's very little Chrome-specific stuff. Also, other browsers have vendor specific stuff in them too, yet people rarely fling shit in their direction? FWIW I also mostly use Firefox for dev because I prefer how some devtool features are designed/implemented. Most of my cross browser issues in FF were "brief" in the senses that they were bugs that got fixed eventual.

skinkestek 5y ago
> I'm not entirely sure how I feel about this statement (and comparison to IE). If anything, Safari is "the new IE" rather than Chrome. Some people who either don't know history or willfully ignores it keeps claiming that Safari is the new IE, at one point one even made a webside out of it. Don't fall for it. Chrome is the new IE: - technologically advanced? Check! - implement a number of things without asking or waiting for consensus? Check! - will be abandoned as soon as they have crushed every competitor? Well, it is produced by the worlds most famous company when it comes to killing its own software.
- bananabreakfast 5y ago
  I can't believe you seriously just suggested Google is going to abandon Chrome. Next up, Microsoft is going to abandon Word. Also, Facebook is going to abandon Facebook. This is why no one takes these conversations seriously. All vitriol, no substance.
  - danielheath 5y ago
    IE still had released, but it stopped moving forward after 6 because there was no longer a good reason for ms to invest. I doubt google would spend much on chrome if no other browser were popular.

jitl 5y ago

There’s plenty of weird cruft in Firefox’s ~30 year old codebase that causes bugs and unspecified behavior. For example, Firefox’s ContentEditable code deletes things “backwards” compared to all other browsers and operating systems (bugzilla: https://bugzilla.mozilla.org/show_bug.cgi?id=1735608 https://bugzilla.mozilla.org/show_bug.cgi?id=1735608) - this behavior isn’t standardized in any spec, and Firefox is different from other browsers. There are plenty of gotchas in layout/rendering as well, where either the standard is under specified, or Firefox has some small bugs. Maybe Chrome has many-chrome only APIs, but the developer will always need to test in Chrome and iOS Safari 13 (or whatever your oldest supported iOS version is). Chrome and iOS are where the users are, and a good website or app should be usable and beautiful for everyone.

dbrgn 5y ago
Developing UI components based on ContentEditable is trial and error, with a lot of code held together by cross-browser integration tests where you always have to pray that a browser upgrade doesn't brake something new in a subtle and surprising way... 1/5, would not recommend. (But sometimes there's no way around it.)
TheGoddessInari 5y ago
> There’s plenty of weird cruft in Firefox’s ~30 year old codebase that causes bugs and unspecified behavior. Just wondering where you're getting "30 years" from? That asked, a Chrome-shaped monoculture doesn't help anybody. We need more competing implementations, not less. Anyone feel like collaborating on such?
- hiptobecubic 5y ago
  Are you casually asking if anyone wants to start writing a rendering engine?
- jitl 5y ago
  Netscape was founded in 1994, Firefox is derived from Gecko (1997) layout code code open-sourced from Netscape in 1998. See https://en.m.wikipedia.org/wiki/Gecko_(software) https://en.m.wikipedia.org/wiki/Gecko_(software) for some background. I was thinking 1994 when I wrote the comment, but it’s closer to ~25 years if you reckon from Gecko’s birth year of 1997.

nightpool 5y ago

This is absolutely not true. I wrote a simple flexbox layout in Firefox last year and it was majorly broken when I tried to test it on Chrome, since Firefox and Chrome disagreed on how to compute "min-width" for inline images and other replaced content. In this case, my layout worked in Firefox, but Chrome was the browser that was following the CSS standard correctly, and my code was broken.