5 ms·

That thread has hundreds of legit replies with bugs (that are impossible to read on desktop Twitter from what I can tell). I saw a few cases of "I published an

That thread has hundreds of legit replies with bugs (that are impossible to read on desktop Twitter from what I can tell). I saw a few cases of "I published an article about our challenges supporting Safari, the team reached out and then ghosted us", like http://blog.apps.npr.org/2021/08/31/joy-generator.html http://blog.apps.npr.org/2021/08/31/joy-generator.html. Search for Safari on that page... oof. Dear Safari team, the first step in fixing any problem is to admit that you have one. Please internally acknowledge that Safari has _lots_ of issues like these and has a reputation problem among web developers. It's not just people hating on Safari and being ignorant, I would love to support Safari better, but it's just so frustrating in practice and when filing bugs they seem to go into a black hole. I think a big problem is that bugs seemingly get duplicated into some internal system where there might be some discussion going on, but none of that is visible to the rest of the world and it creates that black hole perception and lack of back and forth discussion with the people who are reporting these issues.

kevingadd 5y ago
The Safari team has always been pleasant to work with and they clearly care about the web, but something about their resourcing or schedules just stops them from delivering the best browser. They were great partners for the webassembly spec process, but they also frequently were unable to give any commitments or even talk about their perspective. Then more recently at my current gig we were basically unable to get our application to work in Safari - after a lot of debugging, I figured out that their wasm compiler had a bug that was producing massively bloated stack frames. In the end, getting that fixed required us to leverage personal contacts on the Safari team. Once we got in touch with someone they came up with a fix pretty quick! Back when I was still doing frontend work I frequently found that Safari had crippling bugs, but if you were able to work around the bugs it had good performance and a good feature set. I simply couldn't justify buying a Mac or an iPad to support a small % of the overall web audience. Bring back Windows Safari please :-( (Apple's patent/trademark battles with Khronos have also been a problem for WebGPU, but that would be a troubled standard even without them)
- ksec 5y ago
  >Bring back Windows Safari please :-( Another vote here.
- Steltek 5y ago
  I've long believed that Apple's focus on user privacy was more driven by lack of engineering ability/bandwidth to exploit data themselves, rather than intrinsic belief.
- Androider 5y ago
  It's 100% a culture / organization problem. Check this out. The original poster of the tweet thread being discussed here retweeted: "The bug where background audio stops playing for PWAs seems to be fixed in iOS 15.4 Beta" https://twitter.com/wes_goulet/status/1491238784020905986 https://twitter.com/wes_goulet/status/1491238784020905986 Job well done, ship it! Hang on, let's check out that bug's public tracker at https://bugs.webkit.org/show_bug.cgi?id=198277 https://bugs.webkit.org/show_bug.cgi?id=198277 - Reported in 2019 - In 2021 got a "Thanks for further reporting this bug... This bug is tracked internally". - Lots of excitement when someone finds it's fixed in 15.4 (of course no feedback from Apple that a fix was made anywhere or released, bug is still in original "New" state) - Disappointment strikes: "(on 15.4) it sometimes keeps playing, but sometimes stops when you pull down Notification Center or switch apps. Sometimes the Play button doesn’t work. It also stops when you get a push notification with a sound" :( Now, how much do you want to bet that the "internally tracked" bug is marked fixed, the developer is not aware of these corner cases, but could probably fix them if they only knew. No worries, another developer (the original having long moved on) will probably take a crack at it again in 2024 when this information makes it back to them. I've worked in organizations just like that, where every bug report is a game of telephone. Everything takes exponentially longer and is perpetually broken because the developer and the user having the problem can't just talk to each other. Throwing bugs (and fixes!) over a wall and hoping for the best just doesn't work.
- boogies 5y ago
  > I simply couldn't justify buying a Mac or an iPad to support a small % of the overall web audience. Bring back Windows Safari please :-( What about the other, cross-platform WebKit browsers?

kevingadd 5y ago

The Safari team has always been pleasant to work with and they clearly care about the web, but something about their resourcing or schedules just stops them from delivering the best browser. They were great partners for the webassembly spec process, but they also frequently were unable to give any commitments or even talk about their perspective. Then more recently at my current gig we were basically unable to get our application to work in Safari - after a lot of debugging, I figured out that their wasm compiler had a bug that was producing massively bloated stack frames. In the end, getting that fixed required us to leverage personal contacts on the Safari team. Once we got in touch with someone they came up with a fix pretty quick! Back when I was still doing frontend work I frequently found that Safari had crippling bugs, but if you were able to work around the bugs it had good performance and a good feature set. I simply couldn't justify buying a Mac or an iPad to support a small % of the overall web audience. Bring back Windows Safari please :-( (Apple's patent/trademark battles with Khronos have also been a problem for WebGPU, but that would be a troubled standard even without them)

ksec 5y ago
>Bring back Windows Safari please :-( Another vote here.
Steltek 5y ago
I've long believed that Apple's focus on user privacy was more driven by lack of engineering ability/bandwidth to exploit data themselves, rather than intrinsic belief.
Androider 5y ago
It's 100% a culture / organization problem. Check this out. The original poster of the tweet thread being discussed here retweeted: "The bug where background audio stops playing for PWAs seems to be fixed in iOS 15.4 Beta" https://twitter.com/wes_goulet/status/1491238784020905986 https://twitter.com/wes_goulet/status/1491238784020905986 Job well done, ship it! Hang on, let's check out that bug's public tracker at https://bugs.webkit.org/show_bug.cgi?id=198277 https://bugs.webkit.org/show_bug.cgi?id=198277 - Reported in 2019 - In 2021 got a "Thanks for further reporting this bug... This bug is tracked internally". - Lots of excitement when someone finds it's fixed in 15.4 (of course no feedback from Apple that a fix was made anywhere or released, bug is still in original "New" state) - Disappointment strikes: "(on 15.4) it sometimes keeps playing, but sometimes stops when you pull down Notification Center or switch apps. Sometimes the Play button doesn’t work. It also stops when you get a push notification with a sound" :( Now, how much do you want to bet that the "internally tracked" bug is marked fixed, the developer is not aware of these corner cases, but could probably fix them if they only knew. No worries, another developer (the original having long moved on) will probably take a crack at it again in 2024 when this information makes it back to them. I've worked in organizations just like that, where every bug report is a game of telephone. Everything takes exponentially longer and is perpetually broken because the developer and the user having the problem can't just talk to each other. Throwing bugs (and fixes!) over a wall and hoping for the best just doesn't work.
boogies 5y ago
> I simply couldn't justify buying a Mac or an iPad to support a small % of the overall web audience. Bring back Windows Safari please :-( What about the other, cross-platform WebKit browsers?