So you build a service that doesn’t properly validate or gracefully handle invalid file types and it’s my fault as the end user when I try to upload something and the system fails?
If you build a service that can’t support one of the most commonly used platforms in the world, that’s on you the developer, not me the user. This is software engineering 101 right here, and was completely avoidable on the part of the developers.
It does gracefully handle invalid formats. The problem that students did not follow the directions provided on how to take pictures in an appropriate format. Instead they tried their own hacks, such as changing the file extension.
Another user here claims their daughter was informed repeatedly before the test that HEIC was not supported and given instructions on how to change their phone settings to ensure comparability. This is confirmed in the article:
> though it does link to the College Board’s website, which instructs students with iPhones to change their camera settings so that photos save as JPEGs rather than HEICs.
So why should a student with an Apple product be punished for hostile moves Apple has made? What if the school issued the device the student took the test on? You're punishing the wrong party here.
And knowing that these big companies do stupid stuff around standards is part of building software. This isn't any different from the browser wars. It's completely unrealistic to expect developers to not have to deal with these kinds of issues.
I think the anger around Apple's hostility towards standards is valid, but completely out of scope for this kind of issue.
This, last time i encountered it even windows 10 didn't display them and wanted me to buy a decoder from windows store. This was probably due to licensing fees.
This story shows what happens when you 'think different' from established standards.
Take two seconds to understand the situation before you comment. iOS automatically converts images to JPEG from HEIC if you use a standard input. College Board didn't, and then somehow either failed to do QA testing with iOS devices entirely, or did a poor job in not catching this huge bug. The failure here is definitely theirs.
This is not what happened, as per the article. The problem was with people who were downloading images from their phone to their desktop in HEIC format, then either failing to convert the images to JPEG, or complaining that the conversion process took too long.
>He Airdropped an iPhone image of his responses to his Mac and tried to convert it by renaming the HEIC file to PNG.
Everly Kai had the same problem with Computer Science A last week — she attempted to rename the file to JPEG and received the same email a few hours after submitting her test.
> Sean S. used OneDrive to port a photo to his Windows desktop from his iPhone, then attempted to convert the file with Windows Photo. Due to the photo’s size, the conversion took over five minutes.
The AP testing website has instructions for changing iPhone settings to the appropriate format and a user here confirms that their daughter was repeatedly informed of this matter before the AP test.
Color me stunned that people are so dissatisfied with their touchpads in Linux. 0-3 lines of options in my config have made me perfectly content with every touchpad I've ever used in Linux.
I am away from my laptop for the time being, unfortunately. But it's generally a bit of pointer acceleration and enabling two finger scrolling, if I have to configure anything at all.
IF you count a request for continued examination a "disposition" sure. However, many (most?) applications include one or more such requests. And so, there is much more back and forth than your post implies. Filing such a request typically gets you at least two more back and forths.
I agree with your point, and I don't consider RCE a disposition. Still, the average application has about two office actions (CTFR, CTNF, CTRS, CTEQ) before an allowance or abandonment.
Is all of that really easier to understand than exponential notation? It's a great tool to visualize floating point precision, but it's lot more circuitious to get to an understanding of what a floating point number actually means IMO
I run it, along with other sketchy garbage proprietary software for wook, in a QEMU VM. Or I just dial in and let people suffer through me being on a phone connection owing to their choice of software.
Agreed. I find the whole experience to be horrible. Especially having to install some garbage executable rather than using my browser. It does nothing for me that Google Meet doesn't do better.
We've found Google Meet / Hangouts too unreliable for even small-scale calls (e.g. our daily standups) whereas Zoom has pretty much perfect quality (and the lowest CPU footprint).
I mean, so do all the major WebRTC video chat programs. Google Meet is absolutely painless, and I can use it in the browser instead of downloading Zoom's very sketchy application.
But Google made Hangouts/Meet G Suite only. My university is not using G Suite beyond student email (employees have outlook).
I only started using Zoom now and it solves a lot of my problems: virtual backgrounds while talking to students in my bedroom, recording my lectures, handling large live streams (50+ people), painless set up for non-tech-savvy users.
The only problem I had is that it would corrupt sound from my mic about every hour in a 4 hour stream.
They will be successful, but in part because other chose not to be.
