Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Apple Pulls iOS 16.5.1 and macOS 13.4.1 Rapid Security Response Updates Due to Safari Bug

Apple is slowly whittling away our trust in their software engineering quality. I thought the other bugs in Apple applications was just sloppiness, but it is becoming apparent that Apple has a general poor software quality infection.

Maybe someday we'll get Apple leadership that understands that Software Quality is essential to the Apple brand.

Edit: Correct Apple's spelling correction.
 
Last edited:
  • Like
Reactions: mxrider88, Pinkyyy 💜🍎 and srknpower
ifxf said:
Just proves you should never immediately install any update. Allow the testers (customers) discover the bugs. I wait a month for before installing a update and four months or 2 updates before installing a base release.
Click to expand...

I rather have problems with some websites (have none encountered) than a browser that is vulnerable to an exploited security vulnerability.

Therfore I would have preferred that Apple had not revoked the update, as one of my Macs is not updated.
 
  • Like
Reactions: Pinkyyy 💜🍎 and mr_jomo
djplong said:
Websites - Are you THAT fragile that an extra byte in a VERSION NUMBER causes your site to fail COMPLETELY?
Click to expand...
When it comes to checking user-agent strings? Oh, yes, they're all that fragile and have been for decades.

Have a look at the user-agent string that fully-patched Safari sends:

Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.5.2 (a) Safari/605.1.15

See how "Safari" is actually the last thing it sends? That's because most of this string is fake data that's sent to appease older websites. For instance, websites that want to see a Mozilla-based (like Firefox) browser, as opposed to Internet Explorer (and I mean Internet Explorer 1.0). My Apple Silicon-based Mac still identifies as Intel to websites because a check for Intel is hard-coded into many of them (as opposed to Motorola 68000 or PowerPC, which had a different byte order). KHTML is in there because it's part of WebKit's ancestry from the UNIX side, and some web sites behaved differently for KHTML... but some didn't know KHTML but looked for "Gecko" in the user-agent string to identify behavior available in Netscape Navigator 6.0 and later.

There are a lot of websites doing a lot of incredibly naïve text-string checks on the user-agent.

It's a mess, and it's not surprising at all that the version number change broke things. In fact, the space between the 16.5.2 and the "(a)" probably didn't help at all...
 
  • Like
Reactions: freedomlinux, tomnavratil, Alwis and 1 other person
nt5672 said:
Apple is slowly whittling away our trust in their software engineering quality.
Click to expand...
When you say our please don’t include me.
nt5672 said:
I thought the other bugs in Apple applications was just sloppiness, but it is be common apparent that Apple has a general poor software quality infection.
Click to expand...
It’s not common apparent. Bugs will always be present. Ask Microsoft who has been doing this for decades. It’s a nature if the industry. However a bug that clobbers a system is less preferable than one that doesn’t.
nt5672 said:
Maybe someday we'll get Apple leadership that understands that Software Quality is essential to the Apple brand.
Click to expand...
Maybe, but not today. Don’t know when there will be a change of leadership but thankfully for consumers there are alternatives if one doesn’t really like apple.
 
AJTC said:
I wonder why Apple don’t just call the version 16.5.1.1. We use that version number format in work all the time for emergency fixes.
Click to expand...
Yeah why the weird format? When I first saw it I thought something was wrong or I got malware as I never seen macos version with (a) before.
 
  • Like
Reactions: Pinkyyy 💜🍎
SoYoung said:
In respect, your comment is also biased. You pretend having tons of problems with macOS over the years, it's possible, but you're truly an exception. Everyone I know have little issues with macOS. And then, you pretend I have cheap PCs just because I have Windows problems...so you also completely ignoring that Windows 10 can have issues as well. I just compared my own experience between the two OS. I use macOS all the time and I honestly not having very big issues for years. On another side, I only use Windows for gaming, so it's pretty much a clean install with nothing else except legit games and I have some problems on the OS. I didn't say Windows is bad, I'm just saying every companies can have some issues, and the security update that Apple provide here isn't a big deal. Its just a little issue that can easily be avoided until a second fix is coming, but some users talks about this like the issue is completely break their computer when the real problem isn't just an "apple" one. Every website I visited works fine, except meta ones...and again, even in the past, FB never really allowed changing user agent on Safari so FB have also a blame here.

And yes, Android is objectively not as secure as iOS. They never pull rapid security updates, major Android versions takes age to release on non-google phones, most users aren't at the latest Android version, many companies will not support their phones as long as Apple and its a fact Google play don't have the same quality controls over the apps released there compared to the App Store.
Click to expand...
Yep. Just out of the blue randomly twice now, both Windows 10 and 11 just corrupt my Xbox controller drivers. Forcing me to dig deep and do a hard uninstall of THEIR OWN DRIVER AND OS!

Also four times now both with AMD and NVIDIA two cards each my graphics drivers got corrupted out of the blue. Resulting in me using DDU. Happened on both Windows 10 and 11. Two different computers too.
 
djplong said:
I've started writing code professionally in the 1970s and I have to admit, I'm of two minds about this..

Apple - What, did NONE of your testers log on to Facebook with Safari?

Websites - Are you THAT fragile that an extra byte in a VERSION NUMBER causes your site to fail COMPLETELY?

My god, if a website that *I* wrote crashed because of an extra tag on the END of a version number.... I'd be embarrassed beyond belief. I've had some stupid bugs in my code over the course of almost 50 years but... Really???
Click to expand...

In other words, Apple’s mistake was that it assumed Facebook coders knew what they were doing.
 
User agent stuff has been a problem since the web began. That's why chrome adjusts its user agent string occasionally - which forces lazy web developers to either check features instead of user agent strings or get a library that does.

Apparently lots of websites didn't get that memo.
 
  • Like
Reactions: Pinkyyy 💜🍎
OhMyMy said:
That’s the whole point of separating the security fixes from the OS. Quick design and delivery of security patches.
From your response it’s obvious that you’re just a typical user unaware of how software design works. So I suggest you educate yourself with the fundamentals before making statements like these.
These OS’ have grown quite a bit in complexity in recent years and mistakes do happen as a result. And rewriting the whole thing would take considerable resources and isn’t a viable option even for a company as big as Apple.
Click to expand...
And you think they're good about the regular updates that goes through rounds of testing, haha. Get your head out of the sand. No one in their right mind can argue that Apple's updates have gone down the hill. Look at all the numerous updates to correct the updates over the past few years because the original update broke something.
 
  • Like
Reactions: Pinkyyy 💜🍎
doboy said:
And you think they're good about the regular updates that goes through rounds of testing, haha. Get your head out of the sand. No one in their right mind can argue that Apple's updates have gone down the hill. Look at all the numerous updates to correct the updates over the past few years because the original update broke something.
Click to expand...
I think you’re the one with their head in the sand so far down that you can’t even understand what I said. I was trying to enlighten you on how the RSRs work and that them installing quickly is by design and in no way a reflection on their function.
My response doesn’t concern their software QA at all. Please read and re-read before you respond or you risk looking ignorant.
 
Well it looks like Facebook have sorted their sloppy code as it working fine for me now and I’ve got the rsr installed
 
OhMyMy said:
I think you’re the one with their head in the sand so far down that you can’t even understand what I said. I was trying to enlighten you on how the RSRs work and that them installing quickly is by design and in no way a reflection on their function.
My response doesn’t concern their software QA at all. Please read and re-read before you respond or you risk looking ignorant.
Click to expand...
No duh. I know that's how the RSR is "supposed" to work but it didn't since they had to pull it and now ppl have to do it all over again making it not so rapid. You may not have commented on the QA part but it's all related because if the RSR doesn't work correctly, what's the point.
 
  • Like
Reactions: Pinkyyy 💜🍎
doboy said:
No duh. I know that's how the RSR is "supposed" to work but it didn't since they had to pull it and now ppl have to do it all over again making it not so rapid. You may not have commented on the QA part but it's all related because if the RSR doesn't work correctly, what's the point.
Click to expand...
What about Meta’s poor QA that caused this issue? If they used user agents properly, then this update wouldn’t have been seen to cause issues. It just highlighted bad design by other companies in order to “save time”.
 
  • Like
Reactions: Pinkyyy 💜🍎
doboy said:
No duh. I know that's how the RSR is "supposed" to work but it didn't since they had to pull it and now ppl have to do it all over again making it not so rapid. You may not have commented on the QA part but it's all related because if the RSR doesn't work correctly, what's the point.
Click to expand...
You’re too condescending for someone who’s totally unaware of a thing about software development. It’s not a good look.
RSRs didn’t exist before last year and only started going out to the actual public a couple months ago. There’s still some refinement needed and it may take awhile when there’s a lot of tech that needs to work together. The primary focus is on security patches and odds are while low something may break. And the rapid part refers to the installation time for one update and it was rapid for me as I don’t use anything Fb. QA at Apple isn’t what it used to be back in the day but regardless no software is perfect and is still much better compared to the competition.
 
  • Like
Reactions: Pinkyyy 💜🍎
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back