Slashdot Mirror


Search Suggestions Causing Apple's Safari Browser To Crash on Many Devices (theverge.com)

An anonymous reader writes: According to the Verge (and my wife) Apple Safari browsers are crashing left, right, and center due to Safari's search suggestions feature. "Simply disabling this feature will stop Safari crashing, or using the private mode option in the browser as a temporary workaround. Not everyone is affected, and this could be because some have the search suggestions cached locally or they're still able to reach Apple's servers thanks to a DNS cache."

83 comments

  1. Safari users could be turned into a botnet by Anonymous Coward · · Score: 0

    If suggestions can cause code execution that crashes the browser, then some one can hack into the suggestions server to serve up malware. I hope someone does it to Firefox suggestions as well.

    1. Re:Safari users could be turned into a botnet by Anonymous Coward · · Score: 1

      But firefox autocompletion is pretty good, unlike the garbage that is chrome which is made to report as much personal data as possible while being slow as shit at local history autocomplete.

  2. Apple by stealth_finger · · Score: 0

    it just works.

    --
    Wanna buy a shirt?
    https://www.redbubble.com/people/stealthfinger/shop?asc=u
    1. Re:Apple by invictusvoyd · · Score: 0, Troll

      The company survives today because of the Apple II which actually promoted the hacker ethos. After they killed it their other products and Steve Jobs took them towards bankruptcy . After his quite dramatic return , Jobs ( read Apple engineers) gave the hungry audience an incomplete but good looking and well performing computer and saved the day.

    2. Re:Apple by Anonymous Coward · · Score: 0

      Better than anything else I've used before. You can be snarky but the proof is there.

      But I'm sure you're running a flawless system, eh?

    3. Re:Apple by U2xhc2hkb3QgU3Vja3M · · Score: 1

      Like it or not, it's true. But that's because most people always leave out the last part of that quote.

      Apple, it just works. Until it doesn't.

    4. Re:Apple by MobileTatsu-NJG · · Score: 3, Funny

      Yeah, no kidding. If this happens 99 more times then I might as well be using Windows!

      --

      "I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)

    5. Re:Apple by kuhnto · · Score: 1

      I am by no means an apple fan boy, but I think the most important aspect of what Apple did was to disconnect the cell phone from the carrier. It was their way or the highway.

      --
      "A 'person' is smart. 'People' are dumb, panicky animals and you know that."
    6. Re:Apple by squiggleslash · · Score: 2

      When did they do that?

      The original iPhone was so locked to at&t you weren't even allowed to use a different, non-iPhone, at&t SIM card with it.

      It took Apple years to sell unlocked cellphones, and that was largely pressure from the EU where they weren't even allowed to enter certain markets until they did.

      The only thing that's caused a disconnect of the cellphone from the carrier is the GSM family of standards, which finally, at last, has become a global standard that almost every carrier in the world is finally transitioning to (through LTE, GSM's 4G iteration.)

      Meanwhile, Apple's seeking to undo that by advocating for the removal of physical SIM cards, locking devices to carriers just like in the old days.

      --
      You are not alone. This is not normal. None of this is normal.
    7. Re: Apple by Anonymous Coward · · Score: 0

      Phone companies have no say in how the iPhone works or operates? Lol. Is that why app store downloads over 100mb only work on WiFi? Is that why Face time chats only work on WiFi?

    8. Re:Apple by stealth_finger · · Score: 1

      I am by no means an apple fan boy, but I think the most important aspect of what Apple did was to disconnect the cell phone from the carrier. It was their way or the highway.

      Umm what? You've always been able to get sim free/network free phones direct from the manufacturer or stores, Apple just charge more.

      --
      Wanna buy a shirt?
      https://www.redbubble.com/people/stealthfinger/shop?asc=u
    9. Re: Apple by Dog-Cow · · Score: 1

      FaceTime only works on Wifi? Guess someone forgot to tell my phone.

    10. Re:Apple by vtcodger · · Score: 1

      Perhaps the problem is one of people making rude suggestions to siri. Just because she's imaginary doesn't mean that she doesn't have feelings.

      --
      You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
    11. Re:Apple by stealth_finger · · Score: 1

      Maybe they're clicking it wrong.

      --
      Wanna buy a shirt?
      https://www.redbubble.com/people/stealthfinger/shop?asc=u
    12. Re:Apple by macs4all · · Score: 1

      The company survives today because of the Apple II which actually promoted the hacker ethos. After they killed it their other products and Steve Jobs took them towards bankruptcy . After his quite dramatic return , Jobs ( read Apple engineers) gave the hungry audience an incomplete but good looking and well performing computer and saved the day.

      Idiot.

      Jobs didn't take Apple towards bankruptcy. That would be John Sculley.

      Jobs RESCUED them from the brink of bankruptcy.

      Get your history right, idiot.

    13. Re:Apple by Chris+Mattern · · Score: 1

      Apple: It Just Borks.

    14. Re:Apple by Anonymous Coward · · Score: 0

      What a load of crap!
      Do some research in IMEI whitelisting. Apple sells locked phones to carriers (for example Sprint USA). When you boot such locked iPhone with SIM card from different carrier you get unsupported SIM error.
      Apple hosts whitelist servers -- when you unlock your phone through your carrier (or shady ebay unlocker) your IMEI is entered in Apple's whitelist database. When you first connect to interenet or connect your iPhone to Internet-connected iTunes, it will sync with whitelist servers and unlock your phone if IMEI is in their database.

      Of course Apple also sells fully unlocked phones in their stores...

    15. Re:Apple by Applehu+Akbar · · Score: 0

      "The original iPhone was so locked to at&t you weren't even allowed to use a different, non-iPhone, at&t SIM card with it."

      No so. The original iPhone required AT & T as a carrier because it was world standard, GSM only, and AT & T was at the time the only GSM carrier in the US. When T-Mobile came along, it too supported iPhone.

      Later models added support for the old-timey CDMA cellular scheme, so that the other established carriers like Verizon and Sprint, with fully built-out service networks. could use iPhone too.

    16. Re:Apple by squiggleslash · · Score: 1

      No so. The original iPhone required AT & T as a carrier because it was world standard, GSM only, and AT & T was at the time the only GSM carrier in the US. When T-Mobile came along, it too supported iPhone.

      Absolute balderdash. And it doesn't address the charge.

      1. T-Mobile, formerly Voicestream, has been GSM since the mid 1990s. For what it's worth at&t/Cingular didn't switch until the early 2000s.
      2. If what you were saying were true, there'd be no logic in locking the iPhone in the US anyway.
      3. But even so, it doesn't explain why you needed a special SIM card for the iPhone, not merely an at&t one. Early iPhone buyers who were already at&t GSM users were not able to use their existing GSM cards in the iPhone - they had to instead activate the one that came in the box, giving AT&T unusually ridiculous control over their equipment choice.

      Thankfully the next version of the iPhone ditched the weird SIM card requirement - but nonetheless, iPhones were still carrier locked for a long time afterwards.

      --
      You are not alone. This is not normal. None of this is normal.
    17. Re:Apple by Applehu+Akbar · · Score: 1

      All smartphones are locked to a carrier during the contract period, if the contract includes a subsidy. Now that this arrangement is becoming unpopular, new iPhones are being sold without a contact. Pay upfront, but no locking.

    18. Re:Apple by squiggleslash · · Score: 1

      Indeed.

      Are you agreeing with me (that Apple didn't change the relationship between customers and carriers, and indeed used SIM locking in a way that was extreme relative to other manufacturers) or correcting me on something?

      --
      You are not alone. This is not normal. None of this is normal.
    19. Re:Apple by invictusvoyd · · Score: 1

      Get your history right, idiot.

      End-to-End control ?

    20. Re:Apple by kuhnto · · Score: 1

      YOU are correct. What I was trying to say is that "phone design" was separated from the carrier. From what I remember, I think Apple originally when to Verizon, but Verizon said "No, we want control of how the phone operates, features, etc" ATT was approached next and agreed that Apple would have 100% control of design of the device partially because Apple could bring such a highly demanded product to a carrier. I have to admit, the phone was revolutionaly from "almost" everything before it. It is similar to a situation where Apple goes to Comcast and says they want to build the next Set Top Box, but Comcast gets no control over any of the design and functional aspects for the box. It would be a win for Apple, and a win for the consumers.

      --
      "A 'person' is smart. 'People' are dumb, panicky animals and you know that."
    21. Re:Apple by macs4all · · Score: 1

      Get your history right, idiot.

      End-to-End control ?

      WTF does that even mean?

  3. Not here by Ol+Olsoc · · Score: 2

    Not on Safari at the moment, but I've had no such problems.

    --
    The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
    1. Re:Not here by Anonymous Coward · · Score: 0

      "I don't use it therefore the problem doesn't exist"

    2. Re: Not here by Anonymous Coward · · Score: 0

      Firefox user here. Never had problems with Safari crashing either. Neither has my mom, dog nor my desk lamp (neither of which uses Safari...or computers for that matter).

      I think this is just a small issue that affects a minority of users that is being blown way out of proportion by Apple haters.

    3. Re:Not here by U2xhc2hkb3QgU3Vja3M · · Score: 2, Funny

      No idea. I'm using Safari here but I'm still on dial-up so maybe that's why I haven't had that probl{#`%${%&`+'${`%&NO CARRIER

    4. Re:Not here by MobileTatsu-NJG · · Score: 1

      I can tell you that this bug hit my wife's phone yesterday. I turned Search Suggestions off and now everything's fine. I can tell you I've witnessed it, but I have no idea how widespread it is. I am going to turn off that feature on my phone for now.

      --

      "I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)

    5. Re:Not here by ledow · · Score: 3, Informative

      400+ iPads here, a school.

      At least 100 affected. As soon as the keyboard should pop up to let you type in the search/address bar, it closes Safari. Doesn't matter what you do or what version of iOS you are on (which suggests the server is sending some junk instead of what it's supposed to send, but still bad programming).

      The only fix is to disable search suggestions.

      In fact, I linked all staff and pupils to the BBC News article this morning because it solved the problem we've been having with that all day.

    6. Re: Not here by Anonymous Coward · · Score: 0

      Given the statistics of browser use, Safari users overall are a small minority.

    7. Re:Not here by Anonymous Coward · · Score: 0

      Why is this modded up? Didn't this joke play out in the mid 90s?

      Slashdot; where people complain about old artists still making money from old art but continue to mod up the same nonsense like clockwork.

    8. Re:Not here by Ol+Olsoc · · Score: 2

      Thank you for that. Your hugely statistically significant data point will add so much to the discussion.

      Touchy this morning? I've been testing it out, now that I'm in Safari at home, and it still isn't crashing.

      And perhaps people's browsers that are not crashing might offer some insight into those that are. Just sayin'.

      --
      The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
    9. Re:Not here by macs4all · · Score: 1

      Thank you for that. Your hugely statistically significant data point will add so much to the discussion.

      Hey, that's like all the snarky "Works for me" "help" on Linux forums.

    10. Re: Not here by Anonymous Coward · · Score: 0

      I know, right. Get with the times and at least use the Candlehack meme, you ol

    11. Re:Not here by Anonymous Coward · · Score: 0

      Works for me too.

    12. Re:Not here by Anonymous Coward · · Score: 0

      Being modded funny doesn't cost money and doesn't give him money.

    13. Re: Not here by Ol+Olsoc · · Score: 2

      Firefox user here. Never had problems with Safari crashing either. Neither has my mom, dog nor my desk lamp (neither of which uses Safari...or computers for that matter).

      I think this is just a small issue that affects a minority of users that is being blown way out of proportion by Apple haters.

      Exactly.

      --
      The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
  4. this is why by Thud457 · · Score: 2

    you don't need a SQL database to track your browser search history.

    --

    the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff

    1. Re:this is why by buchner.johannes · · Score: 1

      Apparently you do, since that is what Chrome and Firefox do (Sqlite 3).

      --
      NB: The message above might reflect my opinion right now, but not necessarily tomorrow or next year.
    2. Re:this is why by vtcodger · · Score: 1

      Some people seem to experience a great feeling of disharmony when data is not stored in a relational database. Never made much sense to me, but what the hell. Maybe they know something we don't.

      --
      You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
    3. Re:this is why by Anonymous Coward · · Score: 2, Informative

      Well, you could implement a different form of data storage in each of your applications, reinventing the wheel each time and potentially opening yourself up to new bugs each time, and you could fix those bugs in different ways each time. That kind of insanity is fun in a large organization with lots of devs and apps.

      Or, you could bake a common data storage mechanism into the operating system as an API, name it maybe something like "CoreData," and then have all your devs use the same API to manage data in their applications by making calls to something like /System/Library/Frameworks/CoreData.framework/Versions/A/CoreData. If you run into problems with CoreData, you can upgrade once and fix it across all applications. It's almost like that concept of "libraries" and "shared code" or "code reuse" and "small programs that do one thing and do it well" that they used to talk about in the Linux world before SystemD.

    4. Re:this is why by vtcodger · · Score: 1

      There's surely zero chance of my changing your mind, but you might do that yourself eventually. I submit that The_Simplest_Tool_That_Will_Do_The_Job on average produces better results than One_Size_Fits_All. Think about it from time to time when you're stuck in an airport waiting for a flight that never left Albuquerque or in a traffic jam waiting for tow-trucks to clear the 97 car pileup in front of you on the motorway and have nothing better to do.

      --
      You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
    5. Re:this is why by Anonymous Coward · · Score: 0

      The idea is to break jobs down until The_Simplest_Tool also Fits_All. That's the philosophy behind small programs/libraries that do one thing and do it well.

      Data storage is the kind of field where this works well. Under CoreData, the program is able to hand off persistent data (like history that's saved across browser sessions) to an API that saves the data. The program doesn't know or care how that data is saved, it just knows that everything is cool and that the data will be there when it's wanted again. On the other end of the cycle, CoreData is responsible for accessing the data. The program (and thus the programmer) only needs to care about what to do with that data within the program's active state, now how it gets stored. The programmer never has to develop a file format, choose between CSV and XML and SQL, etc: CoreData takes care of that for him. Simplest_Tool, Fits_Most_or_All. That's how things are supposed to work.

    6. Re:this is why by Anonymous Coward · · Score: 0

      What's the simplest tool might have an answer that changes when one of them is already written and pretty well bug tested.

      I'm all for squeezing out performance and simplicity where it makes sense, but using sqlite for something like that to make the programming task simpler seems a reasonable tradeoff, and does allow for more flexibility.

    7. Re:this is why by BasilBrush · · Score: 1

      I was with you right up until the moment you mentioned CoreData. CoreData is shit, and even Apple don't tend to use it for their apps. And it's non-portable. A very bad thing for data.

      But for sure using some variety of SQL for record based data makes a lot of sense.

      (Of course Core data is hackable on other platforms, as it has SQL (or XML) underlying it. But that doesn't make it a good choice over using SQL directly.)

  5. autocompetition by invictusvoyd · · Score: 1

    The browsers autocompletion tried to compete with his wife ?

  6. Good old BBC by Viol8 · · Score: 0, Offtopic

    Nice to know they always have the most techno literate reporters on speed dial when they need insightful comment on a technology issues like this one with Safari:

    http://www.bbc.co.uk/news/tech...

    "This would suggest that the problem is caused by a process happening at Apple's data centres rather than a coding error in Safari itself."

    Yes, thats right, the apple data centres are sending special CRASH codes to the browsers, its nothing to do with their being a bug in the browser software, no no no. Fecking idiot whoever wrote this piece.

    1. Re: Good old BBC by Anonymous Coward · · Score: 0

      Other people have reported the crash logs report an uncaught exception that is the problem.

    2. Re:Good old BBC by ledow · · Score: 3, Informative

      Please explain why yeseterday it worked, today it doesn't.

      On 1/4 of 400 iPads. On every level of iOS from 7 to 9. On iPads updated either today or last month, rebuilt today or last month, restored today or last month, no matter how old the backup restored from. Simultaneously. Suddenly. Today. And only on search suggestions.

      Because, as a programmer, the only thing I can think of is that they are sending some unexpected junk in the search suggestion reply from the Apple server that isn't handled properly by the browser causing a crash.

      Literally, this morning, a load of our pupil's iPads all started crashing on Safari search suggestions no matter how old, how long ago they updated, what iOS level, what apps were installed, or anything else. But they were all working yesterday. And 3/4 of them still work today.

      It's currently suspected that some Apple server from some kind of round-robin response system has flaked out and produced bad responses that are being cached by those iPads. Restore from known-good-working-backup does not fix the problem and the first search suggestion can crash them again.

      So stop being a smart-arse and research the problem first.

    3. Re: Good old BBC by Anonymous Coward · · Score: 0

      You are one stupid motherfucker, and a fanboi.

    4. Re:Good old BBC by Viol8 · · Score: 0

      "So stop being a smart-arse and research the problem first."

      Not handling a response properly and crashing is a coding bug you clueless fuckwit.

    5. Re:Good old BBC by Anonymous Coward · · Score: 0

      It's both. It's caused by a bad server response. But the browser should not be crashing on that response.

    6. Re:Good old BBC by chmod+a+x+mojo · · Score: 0

      And a shit response from the home server is still the primary problem, since if the proper response is sent the application doesn't crash with an unhandled exception.

      Both pieces of software MAY have bugs, but the application being sent shit data before crashing means that the primary problem is on the side sending the shit data. The fact that the client application then crashes exposes a SECONDARY issue, while simultaneously exposing the primary issue, I.E. the shit responses being sent to the client.

      --
      To err is human; effective mayhem requires the root password!
    7. Re:Good old BBC by Anonymous Coward · · Score: 0

      So to sumamrize: the problem is the browser isn't tolerant of malformed HTML from the server. And that's the server's fault, not the browser. Got it. I guess when you get ass-raped it's your fault, not your attacker's fault.

    8. Re:Good old BBC by Anonymous Coward · · Score: 0

      No! No matter the source you always have to verify input/received data. If the client was properly doing that, it wouldn't have crashed. That is the primary bug. Software only has control of itself, trusting other software to be 100% correct 100% of the time is an absolute failure from the programmer. If you make mistakes like this, you're a very crappy software engineer/designer/developer but probably on par as an average coder/programmer.

    9. Re:Good old BBC by ledow · · Score: 1, Troll

      So if the browser didn't crash, but just produced an error that satisfied your pedantry, there's no problem with that? Because the server sent data?

      And the browser doesn't necessarily crash. It's an unhandled exception, there's nothing to suggest it's exploitable or dangerous, it's just unexpected. The correct response is to fatal error and then get out of there. There's nothing you can really do. I suppose you could just throw away the error and carry on regardless, but that's hardly the point - and all the user would see is a broken search still.

      And the BBC article was just updated. The servers were sending junk between two times. That got cached in the suggestions cache on the browser. They recommend you clear the cache now and try again.

    10. Re:Good old BBC by Anonymous Coward · · Score: 0

      lol theyll be feeling like the victim if they try and touch this unruly hole

      captcha: afraid

    11. Re:Good old BBC by Anonymous Coward · · Score: 0

      Really, I'd expect my software to be more robust than to fatally die when one data value gets screwed up. If search suggestions ends up not being able to suggested something due to an internal error to that module then fine, but that shouldn't bring down the entire program.

      Only exploitable bugs matter? No wonder quality in the software industry has taken a nose dive. It's a basic software design concept that you can't trust user input, no matter if that user is an actual human user or some other section of your system.

    12. Re:Good old BBC by chmod+a+x+mojo · · Score: 2

      Yes! The bad response and garbage data from the search server is the PRIMARY issue here that uncovered a SECONDARY bug in the application.

      It's simple fucking logic: Bad response from server triggers bug in the software, therefore the primary problem that is at hand is the servers barfing and sending bad data.

      The secondary issue discovered is that the application crashes with an unhandled exception when sent bad data. It HAS to be a secondary problem since the program won't crash without being sent bad data, and there is no way for the program to CAUSE the server to send bad data ( in this case ) even after crashing.

      Both are bad things, but as long as the search servers are sending good data the secondary issue isn't as bad, since it won't crop up at all. That isn't saying it shouldn't be fixed ( of course it should ), or that it is a trivial bug, just that as long as everything else works it won't be a problem.... just like it wasn't a problem up until now.

      Damn, it is really stupidly simple cause and effect chains. How can people not see it plain as day?

      --
      To err is human; effective mayhem requires the root password!
    13. Re:Good old BBC by Viol8 · · Score: 1

      If you're a teacher it explains why UK education is in such a poor state. Get a bloody clue and don't ever teach IT classes.

    14. Re:Good old BBC by Anonymous Coward · · Score: 0

      I was going to post the second paragraph, but then I realized we're using different definitions for primary and secondary. I'm using "primary" as meaning the more important bug, not meaning the originating cause of the issue. English sucks. We're arguing about different things.

      What I was going to post but is incorrect with your definitions:
      I don't know how you can't see why it isn't the opposite. The claim you're making is no SQL server needs to check for SQL injection because it should never receive bad inputs. Thus if an SQL injection occurs, it's the fault of the user and we don't need to care about it because the user shouldn't do that. You're looking at program design backwards. The primary fault is not properly validating input. It doesn't matter want anybody is sending you, your program should survive it all.

    15. Re: Good old BBC by Anonymous Coward · · Score: 0

      From this AC's point of view, it's semantics about the cause (cause is server, effect is browser), but you lose the argument because no amount of handling on Safari would get a valid search suggestion when the server is the one shitting the bed.

      It sounds like you just want to feel superior for being a better programmer than the Apple programmer.

      After goto fail, I also think they are shit programmers, but your personal attack on the teacher for a valid different point of view makes you a tool.

    16. Re: Good old BBC by Anonymous Coward · · Score: 0

      Umm, your logic is wrong. The ass raper is the server in your example. Without the attacker, you don't get raped. Whether you can cock block or not, the rape attempt happens from attacker. You're still going to have a bad time.

  7. Not here by Anonymous Coward · · Score: 0

    Not here.... Is this just a media hype?

  8. Bring back Little Steve! by Anonymous Coward · · Score: 0

    The decline (along with the where's-the-button-gone aesthetic) started when Scott Forstall left. Everyone knows he is the anointed one. Cook needs to let the past go and do what he knows big Steve would have done.

    1. Re:Bring back Little Steve! by Anonymous Coward · · Score: 0

      you mean he should do illeagal wage fixing deals?

      not go to proper doctors?

      pretend to innovate?

    2. Re:Bring back Little Steve! by amicusNYCL · · Score: 1

      Cook needs to let the past go and do what he knows big Steve would have done.

      Deny the parentage of his daughter? Park in handicapped spaces? Screw Wozniak over?

      --
      "Our two-party system is like a bowl of shit looking at itself in a mirror." - Lewis Black
    3. Re:Bring back Little Steve! by dgatwood · · Score: 1

      No, take the license plates off of all the security vehicles. It's the only way to be sure that they aren't recognizable for the next step in the plan.

      --

      Check out my sci-fi/humor trilogy at PatriotsBooks.

    4. Re:Bring back Little Steve! by Anonymous Coward · · Score: 0

      Do you deny your poor diet? Eating your words against apk != good nutrition AmicusNYCL http://slashdot.org/comments.p... it was good food for laughter for all of slashdot though.

  9. Search suggestions is it Google by Anonymous Coward · · Score: 1

    I use DuckDuckDo and haven't experienced this on iOS or OSX! Is the issue Google sabotaging this browser?

  10. Weird by Anonymous Coward · · Score: 0

    My Android and Windows devices don't have this problem. Weird.

    1. Re:Weird by Applehu+Akbar · · Score: 1

      "My Android and Windows devices don't have this problem."

      Under Windows, crashing is built into the operating system.

  11. Clearly A Fabricated Story by Anonymous Coward · · Score: 0

    No swish-period user would have a wife.

    Safari has been Binging me search suggestions all morning.

  12. The other Safari problem by Applehu+Akbar · · Score: 1

    I have never had Safari crash on me, but I've always been irked by the lack of ability to suppress autoplaying videos. Apple will not give us a Preferences checkbox for doing that, and it it won't support plugins that keep autoplaying videos from starting unless you specifically click on them. I'm assuming that the advertisers are insisting on this as a condition for deigning to give Apple TV some content. In Safari you can prevent all videos from playing by unchecking JavaScript and WebGL, but then you can't run videos when you want to .

    But switch to Chrome, and there's a handy plugin Disable HTML5 Autoplay, which does exactly that. Right-click if you want a video to play.

    1. Re:The other Safari problem by BasilBrush · · Score: 1

      A lot of people seem to find Chrome better than Safari. But for me, I don't get any further than seeing it's UI and thinking no thanks. It looks like a Windows app.

  13. My data point. by Anonymous Coward · · Score: 0

    FWIW, my son's MacBook Air suffered this last night. Sympton first was thar ENTER key wouldn't start searching and eventually couldn't even get focus into the address box. Fixed by turning off Safari suggestions, Google did not have to be disabled.

  14. Re:earth is egg shaped by Anonymous Coward · · Score: 0

    Worst. Haiku. Ever.