How to Fix the 'Black Dot' Unicode Bug Crashing iMessage

[coolfactor]

The Black Dot of Death syndrome?

If there has been a commonality between why these Unicode characters are crashing, and they've fixed it for past character sequences, I wonder why they can't implement a universe fix or catch-all?

 

[AndoAU]

Maybe an easier solution: Delete the actual message from Messages on your Mac.

It also says that in the article:

"It may also be possible to delete the problematic message on another device running the iOS 11.4or macOS 10.13.5 betas via Messages on iCloud."

 

[497902]

Oh, give me a break... if we're just going to say "well I don't believe them" then this entire conversation is pointless. Until Apple shows me otherwise, I believe them when they say it's tokenized. They have never given me any reason to assume otherwise, especially when they don't make their money on data mining. Facebook, on the other hand, has given us a LOT of reasons to distrust them and makes all of their money on data mining.

Alright then, put your blind faith into a closed source communication service.
[doublepost=1525898814][/doublepost]

"Pretty sure", based on what?

I already explained that.

 

[posguy99]

At this point, just throw the whole iOS 11 out the window! It was a compete dumpster fire.

You missed reading the article, where it is pointed out that the problem is multi-platform.

 

[Avenged110]

But not some others.

Immune to this, but vulnerable to 100 other exploits.

Eh, there's no security but at least my Messages app is stable. That's fine with me.

 

[Sevendaymelee]

Still on the last version of 10 and couldn't be happier.

 

[Markoth]

Article says it impacts Android as well as iOS, which suggests this is some very low level, very common library for handling unicode that likely impacts Mac and Windows as well.

When I write code, I just stay the heck away from unicode and stick with ASCII... millions of characters and trillions of two character combos, which go into millions of trillions of three character combos... just way too many edge cases that could go catastrophically wrong.

Of course, Apple doesn't have that luxury. They're making a messaging app for end users around the planet - it can't just ignore the massive issue of handling unicode.

A well designed unicode library should be able to handle it. I think I'm going to investigate which library is the culprit here out of curiosity.

EDIT:
That was quick: http://site.icu-project.org/#TOC-Products-from-Apple

Time to get out the torches & pitchforks.

 

[bighype]

Why are Apple's programmers so bad at Unicode? This is like 5th bug that crashes iMessage. With all those billions in the bank, can't they hire some competent engineers? Other messaging companies don't have these kinds of systemic problems.

 

[C DM]

Why are Apple's programmers so bad at Unicode? This is like 5th bug that crashes iMessage. With all those billions in the bank, can't they hire some competent engineers? Other messaging companies don't have these kinds of systemic problems.

Seems like other ones are affected in this case as well.

 

[bighype]

Seems like other ones are affected in this case as well.

Sure, but that doesn't excuse them for previous instances. And actually, that's even worse since Apple was bitten by this bug before. You'd think they'd overhaul the whole stack to make sure it doesn't happen again but nope... they just made band-aid fixes and called it a day. That's why these bugs keep coming back.

It reminds me of how bad iOS calendaring was/is. Remember all those embarrassing date calculation errors? iOS was plagued by them for years. I'm still not convinced it's been fully fixed... just like this Unicode fiasco.

 

[C DM]

Sure, but that doesn't excuse them for previous instances. And actually, that's even worse since Apple was bitten by this bug before. You'd think they'd overhaul the whole stack to make sure it doesn't happen again but nope... they just made band-aid fixes and called it a day. That's why these bugs keep coming back.

It reminds me of how bad iOS calendaring was/is. Remember all those embarrassing date calculation errors? iOS was plagued by them for years. I'm still not convinced it's been fully fixed... just like this Unicode fiasco.

But it's not the same thing. It sounds like it's some sort of a generic simplistic thing, but it's quite more likely almost the opposite of that in the sense of all the potential possibilities that are involved and how they can affect things. (Just like time and calendar calculations are actually quite a bit more involved with more places for pitfalls that most would think.)

None of that is to excuse anything or anything like that, but to bring a bit more reality into it.

 

[vmachiel]

Or just use another messaging app.

First of all, that won’t solve the problem. Second, that’s not a proper solution.

 

[carlsson]

Until we get iMessage in the cloud, deleting a conversation on one device leaves it intact on the others.

Aha, didn’t know that. Thanks for the clarification!
(Good then I wrote “maybe” )

 

[bighype]

But it's not the same thing. It sounds like it's some sort of a generic simplistic thing, but it's quite more likely almost the opposite of that in the sense of all the potential possibilities that are involved and how they can affect things. (Just like time and calendar calculations are actually quite a bit more involved with more places for pitfalls that most would think.)

None of that is to excuse anything or anything like that, but to bring a bit more reality into it.

It's the same subsystem: text and Unicode. How many times in the past has a text message been able to crash & block iMessage? 3-4 times? You'd think they would have rewritten the stack to sanitize and make it sure so no set of characters can crash the whole application. But no... they just patched the flaws without fixing the root cause.

 

[ROGmaster]

According to the artice it affects Android as well.

Maybe we shall throw all smartphones out the window?

It affects Whatsapp on Android specifically.
Whatsapp will be updated and this issue will go away.
[doublepost=1525942549][/doublepost]

True but both offer end to end encryption by default nonetheless and iMessage is baked into iOS. The hardest hurdle for messaging apps is getting people to use them.

What's so hard about simply installing a cross platform app on a smartphone?
The fact that iMessage is limited to Apple's ecosystem is not a disadvantage?
[doublepost=1525943060][/doublepost]

How so? iMessage is totally encrypted and tokenized. Add the fact that it's already integrated into every iOS device and auto-sends as SMS when the receiver needs it, I don't see how anything else could possibly be better. Any service that requires me to ask someone to download the app so I can message them isn't worth using IMO.

So how do you communicate with somebody that has an Android phone? You ask him to buy an iphone? or you simply refuse to talk to people that own Android phones(like I've seen various iphone users state online)? Or you use the insecure and non-encrypted SMS?
[doublepost=1525943342][/doublepost]

Alternative facts and sensationalism, as they say.

https://www.androidpolice.com/2013/...oid-app-google-is-aware-and-working-on-a-fix/

Using the word never was quite of a strech but your link is from 2013. That's aeons in tech years.

 

[Lonak75]

The article says that Android is also impacted…

According to Forbes, only affects Android minimally. Just got to close the window.

Confirmed: iPhones Suffer Serious New Problem
https://www.forbes.com/sites/gordon...pad-mac-macos-ios-problem-apple-tv-watch/amp/

 

[I7guy]

Or just use another messaging app.

Other than WhatsApp? This is a cross platform bug, not confined to iOS.
[doublepost=1525956172][/doublepost]

..:So how do you communicate with somebody that has an Android phone?...

SMS. Been doing it for years.

 

[Bacillus]

The Black Dot of Death syndrome?

If there has been a commonality between why these Unicode characters are crashing, and they've fixed it for past character sequences, I wonder why they can't implement a universe fix or catch-all?

As it seems to be a display problem, I’d expect a fix using an error trap & question mark symbol.
(which might require a core OS specialist 15 mins and Craig 2 mins to mobilize him)

 

[C DM]

Using the word never was quite of a strech but your link is from 2013. That's aeons in tech years.

But certainly easily addresses the sensationalistic absolute that was being presented as fact.

 

[iAdam1n]

Still on the last version of 10 and couldn't be happier.

It also affects iOS 10 by the way.

 

[NoahK17]

Maybe an easier solution: Delete the actual message from Messages on your Mac.

It crashes Mac OS too.

 

[ROGmaster]

But certainly easily addresses the sensationalistic absolute that was being presented as fact.

Actually it shows that his statement was not that far off.
[doublepost=1526020518][/doublepost]

SMS. Been doing it for years.

So the insecure and unencrypted SMS.
I thought security was very important for you iphone users.

 

[chucker23n1]

I'd love to hear a technical explanation for why these issues keep coming up.

Since it impacts both Android and iOS, I assume this is something low-lying in a common library for handling Unicode?

I don't believe so. Instead, it's inherent in how Unicode glyphs are expected to be rendered.

What exactly is that doing which ends up with such bad results when it messes up... and like... couldn't you write some unit tests which just generate hundreds of millions of random strings to throw at that function during testing?

Yes, fuzz testing probably helps, but there's a crazy amount of permutations due to stuff like combining codepoints.

Apple may want to move text rendering out of process, but is probably shying away for performance reasons.

 

[I7guy]

Actually it shows that his statement was not that far off.
[doublepost=1526020518][/doublepost]


So the insecure and unencrypted SMS.
I thought security was very important for you iphone users.

What? The “I’ll be there in 10” has to be encrypted? iPhone users understand the limitations of technology and use it appropriately.

 

[chucker23n1]

How long is the string in question?

Given the ease with which these bugs can be sent to others, could they run an automated test covering the entire Unicode character set? Obviously if the bug requires a multiple character string it becomes significantly more difficult to achieve full test coverage.

Unicode 10.0 has >136,000 character codepoints plus additional non-character codepoints. That by itself isn't so bad, but some codepoints can (and will) get combined, sometimes even recursively. For example, ö can be expressed both as its own codepoint LATIN SMALL LETTER O WITH DIAERESIS, or by combining the two codepoints for ¨ COMBINING DIAERESIS and o LATIN SMALL LETTER O.

For emoji, it gets a lot more complex than that — there are the skintone modifiers such as EMOJI MODIFIER FITZPATRICK TYPE-3, and various other combining codepoints that let you, say, create a family emoji by combining three different humans.

What you end up with is a practically unlimited amount of permutations, and the best you can do is to hard-limit the amount of recursion allowed. Presumably, Apple and Google already do this in some places.
[doublepost=1526035927][/doublepost]

What? The “I’ll be there in 10” has to be encrypted?

Sure. My personal schedule is nobody's business, and there are plenty of reasons you wouldn't want someone snooping on where they can find you at what point in time.

iPhone users understand the limitations of technology and use it appropriately.

Maybe, but that doesn't mean technology shouldn't strive to be better.