Hacker News Reader

Comment by ghostcluster | Hacker News Reader

How many gigantic software disasters can Apple have in one week?

It’s been a bad week for apple.

Before the next spate of “what happened to quality at Apple” articles come out... I would love to know objectively if there are actually more problems or if it’s just bigger impact or an unfortunate run of problems (happens...).

Before this year, I had never seen an Ars Technica review of a major OS X release with a GIF like this: http://cdn.arstechnica.net/wp-content/uploads/2017/09/Sep-23...

Which article is that from?

> Unfortunately, my experience with the new Metal-driven window server hasn’t been flawless. I’ve run into graphical glitching on both my MacBook Pro and iMac that have persisted into the GM build, and I’ve seen developers on Twitter posting about the same thing. Most of the glitches seem benign (blinking videos, webpages, or 3D graphics), but some seem serious (I can reliably make Finder windows flip out as in the GIF above, and older Nvidia systems have been running into kernel panics throughout the beta process and into the GM). Hopefully, these are issues that can be ironed out in the first couple High Sierra updates and are not ongoing discoveryd-type problems that Apple needs to walk back.

https://arstechnica.com/gadgets/2017/09/macos-10-13-high-sie...

Hence that is why i never liked or trust Graphics Hardware acceleration when you dont have access to the source for Drivers. I really wish Apple would just ship their own GPU in Mac as well, but many people dislike the idea because Apple is unlikely to provide an GPU drivers for Windows, meaning dual booting is no longer an option.

To be fair, I have difficulty imagining how you could practically test for this kind of bug ahead of time. You can't test every feature in every app in the app store with the clock set to every possible date and time.

It's not every app. It's a problem with the OS and notifications.

This is something you test with unit tests. It should have been pretty easy to discover it. I know most of us don’t have that much cover by these tests, but also we don’t have hundreds of millions of daily users. They ahould invest into this.

This is not a bug that should be caught by testing, but by simply looking at the code and thinking about it (and often getting others to do the same for your code.)

The current fashion of "write some code, write some tests, modify either until the tests pass" is not a way towards fewer bugs. All it leads to is the analogy of "throwing things on a wall and seeing what sticks". Tests should be for reducing the remaining uncertainty about the functioning of some code, and not a substitute for actually trying to get it right through careful reasoning.

Unit testing seems to frequently result in overconfidence.

Reply on news.ycombinator.com