dewme

About

Username
dewme
Joined
Visits
932
Last Active
Roles
member
Points
15,799
Badges
2
Posts
6,115

Reactions

10.5KLike59Dislike864Informative

  • Latest macOS Ventura update brings a location services bug

    dewmedewme
    mayfly said:
    dewme said:
    Sounds like Apple’s regression testing missed this one. It happens. Humans still have their place.
    As public beta testers! The couple thousand real alpha, beta and gold testers can in no way predict the different ways hundreds of millions of consumers are going to use their devices!
    I agree with your point but it’s not what I’m saying. 

    One of the primary uses of regression testing is to verify that a set of features and functions that are known to be working prior to a new build are not broken by the latest (new) build. Because these test cases are so well established and serve as reasonable indicators that the build is stable from the standpoint of existing functionality, they are very often automated. In software organizations that follow continuous integration principles the regression test suite is often run at least daily, on every build, or even at every check-in. 

    As you may easily imagine, automated regression tests cannot cover every possible code path or feature. But as time goes by the regression test coverage increases for things that can be automated, like checking the results of a menu selection with a given set of system conditions, all of which can be established within the scope of automated testing, which the feature in question likely exhibits. 

    I don’t see the broken feature in question being dependent on how different users end up activating the code path that surfaces the broken feature. But I totally agree that there are plenty of features and functions that aren’t amenable to test automation and that’s where human testers and human plus automation testing comes into play. 

    Alpha, beta, and other testers definitely provide another layer of test verification. But they also provide a much more valuable layer of product and feature validation, I.e., an assessment that the features and functions that the product delivers are actually useful for how customers in different problem domains use the product. The developers specifying, creating, and verifying the features rarely have a complete picture or global understanding of how the product will be used. 

    Customer driven validation testing that covers a wide swath of representative customers is about as good as it gets, which is why Apple is so supportive of developer and public beta testing. But Apple’s first line of defense is their internal verification testing. The more often and more inclusive they can make their test coverage, the better off the product will be. Testing a product of the size and complexity of iOS more often definitely requires test automation. A commitment to improving test coverage and avoiding breakage requires the regression test suite to be revisited and updated as the scope of the product’s functionality expands. 
    muthuk_vanalingamAlex1N

  • Latest macOS Ventura update brings a location services bug

    dewmedewme
    Sounds like Apple’s regression testing missed this one. It happens. Humans still have their place.
    FileMakerFellerwatto_cobraAlex1N

  • Latest Intel and AMD vulnerabilities a gentle reminder to switch to Apple silicon

    dewmedewme
    whatisgoingon said:
    The thing is, there's nothing magic about Apple's processors, that somehow prevent them from having these kind of vulnerabilities.  They just are unlikely to have these specific ones.  Apple's processors are in widespread enough use, with lots of high-value targets, where people will find these kind of vulnerabilities in Apple silicon.

    If anything, Apple will (based on history) be much much slower at fixing/patching and reporting these vulnerabilities, when Apple becomes aware of them.
    Someone outside of Apple will undoubtedly discover vulnerabilities in Apple’s processors at some point. Anything as complex as the M1 is bound to have anomalies just do its shear complexity. Apple has the advantage of going into their design process from day one with security as a design imperative. Some of Intel’s issues have been due to legacy designs that have been carried forward/inherited from an era when security was less of a concern than, for example, performance and efficiency. I suppose the same can be said of the design components that Apple gets from ARM. Hopefully the wider use of ARM has subjected it to greater scrutiny because more eyes and test cycles are on it.

    I don’t think iBiCCC understands the difference between a programming language and a compiler.
    muthuk_vanalingamravnorodomFileMakerFellerwatto_cobra

  • Latest Intel and AMD vulnerabilities a gentle reminder to switch to Apple silicon

    dewmedewme
    It’s true that Apple doesn’t have to worry about x86 vulnerabilities. But it’s only a matter of time until Apple Silicon vulnerabilities are discovered. 
    Fidonet127ITGUYINSDdanoxwilliamlondonbala1234mobirdappleinsideruserFileMakerFeller

  • Apple still doesn't need RCS, but the latest update brings it closer to being suitable for...

    dewmedewme
    gatorguy said:
    auxio said:
    gatorguy said:
     Isn't support for SMS falling by the wayside, and deprecation set to accelerate? 
    • Microsoft no longer supports SMS for some sign-in types, including new devices and multi-factor authentication
    • Facebook/Meta Messenger will no longer support SMS as of September 28, 2023
    • Signal is removing SMS and MMS support to improve user safety and data protection
    Apple can cling to the insecure SMS as a backup and cross-platform messaging standard as long as they want, and for solely competitive reasons, but they're not doing Apple users any favor by doing so.

    My guess is that, for purely profit reasons, Apple will refuse to make any iMessage protocol changes until law or regulators mandate it. But change they will, and probably sooner rather than later. Any takers on a friendly wager of within 12-16 months (probably less but I'm being generous)?
    I love when people try to bring in the profit argument. We live in a capitalist world, every company needs profit or they go out of business. So let's analyze Google's profit model.
    I 100% agree with you, but I think you miss or ignore the point: Apple stubbornly clinging to SMS, and refusing to make iMessage cross-platform, has little to nothing to do with the reasons Apple states, essentially "we're doing it for our customers". It's for profit reasons, just as what drives most companies to do what they do.
    I concur. Apple is very good at maintaining its competitive “stickiness,” which is really just a soft form of lock-in. Yeah, you “can” move your stuff to another platform if you “really want to,” but it’s not going to be easy. Therefore, sticking with what you’ve already got is the easiest and bumpless way to go for most current customers. Every company I’ve worked for tried very hard to build stickiness and especially when open standards based alternatives were available.
    gatorguywilliamlondonLuJohnson