5 ms·

Hmmm... .accordion .content { padding:0 10px; overflow:hidden; border:1px solid #fff; -webkit-transition: all .5s ease-out;

by harryf 10y ago

Hmmm... .accordion .content { padding:0 10px; overflow:hidden; border:1px solid #fff; -webkit-transition: all .5s ease-out; -moz-transition: all .5s ease-out; } -webkit-transition, -moz-transition... it's like we're at 180 degrees to where the web was in the days of IE4-6

ricardobeat 10y ago
transition has been standardized and available in all browsers for years already, that is only necessary as fallback for very old browsers.
- talmand 10y ago
  The only reason I use the webkit prefix on transition is for Android, otherwise I would say it isn't necessary at all.
  - ry_ry 10y ago
    Still a suprising number of old safari builds out in the wild, from a very quick look at our GA. Largely elderly iphones I assume? Either way, I'm reticent to pull the WebKit prefixes from our mixins for now. I wonder if my employers would be interested in releasing (& aggregating with other relatively large sites) our browser usage stats to create a categorised set of current 'real world' stats?
clarry 10y ago
Remember when Good Developers would hate on framesets because they break the back button and linking and everything? Now it's totally OK to break all that with AJAX and "text-as-apps"... I much prefer the days of the old web.
- nkassis 10y ago
  Days of the old web where flash was pervasive and best viewed in [IE6|Netscape|Mosaic|...] pages were everywhere. Every generation of the web has had really crappy devs building bad experiences. The current generation is not making anything worse than it's predecessor.
  - JustSomeNobody 10y ago
    >... is not making anything worse... That's a pretty low bar.
    - jdmichal 10y ago
      Maybe, but it's an accurate assessment to someone pining for the "old days"...
  - TorKlingberg 10y ago
    "best viewed in Chrome" is getting annoyingly common.
    - inanutshellus 10y ago
      Bank Of America recently told me it only supported Safari and Chrome on MacOS... But... supported Firefox on Windows. Whaaaa....
      - clarry 10y ago
        And my bank keeps constantly nagging me that my browser (Firefox LTS) is no longer supported and that I must update immediately.
      - jdmichal 10y ago
        "Support" means they are on the hook if something is broken. Every OS/browser combination is a cost center and potential liability.
      - nommm-nommm 10y ago
- Hamcha 10y ago
  I have to maintain a web interface with framesets and it's much worse than the React stuff I build. Yes it's "one huge js-dependant page for everything", but it can keep track or where it is (and go back to it) using location.hash or path entirely (if the webserver agrees). So I can fix single page apps, I can't fix framesets.
  - clarry 10y ago
    From a user's perspective, what you [as a developer] can or cannot fix doesn't really matter. What's broken is what matters. I find that with the JS craze we've been living for a few years now, things break a lot in the same way they did back in the frameset days. Except that now it is worse in some ways; with framesets, I could link to a specific frame if I wanted. And I could navigate back and forth too, as long as I understood the mechanics behind it. Now with the JS crapps, it's no longer something I can work around as a user.
    - sanderjd 10y ago
      > JS crapps Just as a pointer, using this "portmanteau as diss" discourse style is generally a bad idea if you want to be persuasive to people who aren't sure yet whether or not they agree with you, rather than just "preaching to the choir" who already share your views. The rest of your comment made some decent points, but name-calling like this just comes off as childish.
- oliwarner 10y ago
  I would strongly disagree with "totally OK". A crappy, lazy developer won't care about state history. A good developer will. You can do things right if you want to. It's more work but not a huge amount more.
- ceejayoz 10y ago
  Every modern framework I'm aware of has either built-in or popular third-party routers that use HTML5's History API to make the back button and links work just like they'd work in a static HTML-only site.
- SquareWheel 10y ago
  That's a very outdated view on webapps. https://developer.mozilla.org/en-US/docs/Web/API/History_API#The_pushState()_method https://developer.mozilla.org/en-US/docs/Web/API/History_API...
iamtyce 10y ago
Autoprefixer and prefixfree allow for writing 1x declaration and as support improves over time the prefixes are automatically dropped from your build.
zeveb 10y ago
I think we learned from X- headers and x- media types that vendor prefixes are a mistake: inevitably one must support vendor-prefixed keys forever. That's why that practice is no longer recommended: just use the right key, and specify it properly so others may support it as well.
nostrademons 10y ago
The browser prefixes haven't been necessary on transition for almost 3 years now: http://caniuse.com/#search=css%20transition http://caniuse.com/#search=css%20transition This is just a webdev with old habits that isn't staying current.

by harryf 10y ago

ricardobeat 10y ago

transition has been standardized and available in all browsers for years already, that is only necessary as fallback for very old browsers.

talmand 10y ago
The only reason I use the webkit prefix on transition is for Android, otherwise I would say it isn't necessary at all.
- ry_ry 10y ago
  Still a suprising number of old safari builds out in the wild, from a very quick look at our GA. Largely elderly iphones I assume? Either way, I'm reticent to pull the WebKit prefixes from our mixins for now. I wonder if my employers would be interested in releasing (& aggregating with other relatively large sites) our browser usage stats to create a categorised set of current 'real world' stats?

clarry 10y ago

Remember when Good Developers would hate on framesets because they break the back button and linking and everything? Now it's totally OK to break all that with AJAX and "text-as-apps"... I much prefer the days of the old web.

nkassis 10y ago
Days of the old web where flash was pervasive and best viewed in [IE6|Netscape|Mosaic|...] pages were everywhere. Every generation of the web has had really crappy devs building bad experiences. The current generation is not making anything worse than it's predecessor.
- JustSomeNobody 10y ago
  >... is not making anything worse... That's a pretty low bar.
  - jdmichal 10y ago
    Maybe, but it's an accurate assessment to someone pining for the "old days"...
- TorKlingberg 10y ago
  "best viewed in Chrome" is getting annoyingly common.
  - inanutshellus 10y ago
    Bank Of America recently told me it only supported Safari and Chrome on MacOS... But... supported Firefox on Windows. Whaaaa....
    - clarry 10y ago
      And my bank keeps constantly nagging me that my browser (Firefox LTS) is no longer supported and that I must update immediately.
    - jdmichal 10y ago
      "Support" means they are on the hook if something is broken. Every OS/browser combination is a cost center and potential liability.
    - nommm-nommm 10y ago
Hamcha 10y ago
I have to maintain a web interface with framesets and it's much worse than the React stuff I build. Yes it's "one huge js-dependant page for everything", but it can keep track or where it is (and go back to it) using location.hash or path entirely (if the webserver agrees). So I can fix single page apps, I can't fix framesets.
- clarry 10y ago
  From a user's perspective, what you [as a developer] can or cannot fix doesn't really matter. What's broken is what matters. I find that with the JS craze we've been living for a few years now, things break a lot in the same way they did back in the frameset days. Except that now it is worse in some ways; with framesets, I could link to a specific frame if I wanted. And I could navigate back and forth too, as long as I understood the mechanics behind it. Now with the JS crapps, it's no longer something I can work around as a user.
  - sanderjd 10y ago
    > JS crapps Just as a pointer, using this "portmanteau as diss" discourse style is generally a bad idea if you want to be persuasive to people who aren't sure yet whether or not they agree with you, rather than just "preaching to the choir" who already share your views. The rest of your comment made some decent points, but name-calling like this just comes off as childish.
oliwarner 10y ago
I would strongly disagree with "totally OK". A crappy, lazy developer won't care about state history. A good developer will. You can do things right if you want to. It's more work but not a huge amount more.
ceejayoz 10y ago
Every modern framework I'm aware of has either built-in or popular third-party routers that use HTML5's History API to make the back button and links work just like they'd work in a static HTML-only site.
SquareWheel 10y ago
That's a very outdated view on webapps. https://developer.mozilla.org/en-US/docs/Web/API/History_API#The_pushState()_method https://developer.mozilla.org/en-US/docs/Web/API/History_API...

iamtyce 10y ago

Autoprefixer and prefixfree allow for writing 1x declaration and as support improves over time the prefixes are automatically dropped from your build.

zeveb 10y ago

I think we learned from X- headers and x- media types that vendor prefixes are a mistake: inevitably one must support vendor-prefixed keys forever. That's why that practice is no longer recommended: just use the right key, and specify it properly so others may support it as well.

nostrademons 10y ago

The browser prefixes haven't been necessary on transition for almost 3 years now: http://caniuse.com/#search=css%20transition http://caniuse.com/#search=css%20transition This is just a webdev with old habits that isn't staying current.