5 ms·

How was this not catched during QA?

by montenegrohugo 8y ago

jve 8y ago
What's /e/? Could you provide a link? Because that something non-googable, non-duckable, non-bingable.
- godelmachine 8y ago
  Same question
  - DavidPiper 8y ago
    /e/ is an open, privacy-conscious mobile OS currently under development - https://e.foundation/ https://e.foundation/
    - akuji1993 8y ago
      What the hell is going on with the scrolling on that page? You're just skipping backward and forward instead of the usual consistent scroll..
      - FreeFull 8y ago
        And there is a custom scrollbar that you can't actually click or drag. And the website changes the cursor for no clear reason, too.
      - ratsimihah 8y ago
        They sure don't care about UX.
        akuji1993 8y ago
        Since I started working for a UI/UX design company, I notice so much of this. It gets annoying really fast if you developed an eye for it.
    - singularity2001 8y ago
      a hundred 1000 Dollars crowdfunding sure will take them a long way /s/ https://e.foundation/crowdfunding/#single/0 https://e.foundation/crowdfunding/#single/0 .foundation?
    - mikorym 8y ago
      Why not use a Linux derivitive?
- ritchie_13 8y ago
  Looks to be this https://e.foundation/mobile-phone-os/ https://e.foundation/mobile-phone-os/
- ratsimihah 8y ago
  They sure don't care about SEO.
- treizecisidoi 8y ago
  /e/ is LineageOS + microG + a custom launcher (which is kind of rudimentary ATM) + a few apps from f-droid.org with minimal UI changes to give it a "personalized" look. They provide 50MB cloud storage a mail account (50MB) and that's all there is. Plus a lot of PR. They imply to be the only Google-free Android OS (considering the multitude of roms based on LineageOS this is BS). In all their published material that I've seen there is no mention of LineageOS which is like 95% of their product. So relax, you didn't miss much.
philjohn 8y ago
Because it doesn't happen on all devices - my XS is unaffected for instance.
- zaarn 8y ago
  QA would normally test hundreds of devices with different settings for each test depending on the test. Testing a single or a low number of samples is a good way to get a bad or insignificant result.1
  - saagarjha 8y ago
    Apple at the very minimum tests many thousands of devices before shipping hardware.
    - zaarn 8y ago
      Yes and it makes me curious how they missed such obvious bugs despite thousands of tests.
      - pwinnski 8y ago
        There are many tens of thousands of third-party and no-name chargers, and Apple only tested hundreds, or maybe thousands. The vast majority of chargers, including all chargers for which Apple is responsible, work fine.
        zaarn 8y ago
        Actually no, Louis Rossman has uploaded a video where an Apple-branded charger observes this behaviour. So no, this doesn't work with a vast majority of chargers, including the ones Apple is responsible for.
        ratsimihah 8y ago
        Many they forgot this particular test case while "many thousands of devices."
ratsimihah 8y ago
I know right? This reminds me of work, when I sometimes send over builds for testing, and they have major edge case bugs. Although "plugged in and locked" isn't really an edge case. Maybe Apple should make its new slogan "it mostly works."
dogma1138 8y ago
I’m not sure there is a real issue behind this this doesn’t happen with the original charger or any good charger it seems without tampering I’ve found 2 devices that don’t auto charge on my XS both have pass through or secondary functionality capabilities one is a cheap charger at work that comes with an SD card reader for some reason and another is when I use my USB OTG dongle with charge passthrough that I use to connect my hardware password wallet (mooltipass) to my phone. All the demonstrations I’ve seen so far were with either non-Apple and no brand at all chargers, often coupled with a USB MITM power meter with data passthrough or some other dongle. This might just be working as intended feature.
- zaarn 8y ago
  Here is a demonstration with an apple-brand charger: https://www.youtube.com/watch?v=r7tf4JASzlI https://www.youtube.com/watch?v=r7tf4JASzlI A USB Power meter does not inhibit the function of a charger, they only tap the power lines of the USB line with a low value resistor to measure the voltage drop. You can do it with a hall sensor too in which case you don't interrupt the voltage line at all. I see no reason why a current sensor should in any way inhibit a charger's function unless it's deliberate.
  - dogma1138 8y ago
    It doesn’t matter if it inhibits or not but how it actually negotiated with the charger, charger only works anything that adds stuff like data or OTG requires an unlock to approve. To me this looks like working as intended.
    - zaarn 8y ago
      The current sense should not add data or OTG, these devices are completely out-of-band for the USB bus. This definitely does not work as intended. Again, all this does is insert a small resistor on one side of the voltage line. It does not tap the data lines, they go straight through to the charger as normal.
      - 8y ago
endorphone 8y ago
You complain about hardware issues (that generally happen to a tiny percentage of users, and are inevitable with a consumer electronic that sells in mass quantities) and then talk about switching to a software platform. It does not follow.

How was this not catched during QA?

jve 8y ago

What's /e/? Could you provide a link? Because that something non-googable, non-duckable, non-bingable.

godelmachine 8y ago
Same question
- DavidPiper 8y ago
  /e/ is an open, privacy-conscious mobile OS currently under development - https://e.foundation/ https://e.foundation/
  - akuji1993 8y ago
    What the hell is going on with the scrolling on that page? You're just skipping backward and forward instead of the usual consistent scroll..
    - FreeFull 8y ago
      And there is a custom scrollbar that you can't actually click or drag. And the website changes the cursor for no clear reason, too.
    - ratsimihah 8y ago
      They sure don't care about UX.
      - akuji1993 8y ago
        Since I started working for a UI/UX design company, I notice so much of this. It gets annoying really fast if you developed an eye for it.
  - singularity2001 8y ago
    a hundred 1000 Dollars crowdfunding sure will take them a long way /s/ https://e.foundation/crowdfunding/#single/0 https://e.foundation/crowdfunding/#single/0 .foundation?
  - mikorym 8y ago
    Why not use a Linux derivitive?
ritchie_13 8y ago
Looks to be this https://e.foundation/mobile-phone-os/ https://e.foundation/mobile-phone-os/
ratsimihah 8y ago
They sure don't care about SEO.
treizecisidoi 8y ago
/e/ is LineageOS + microG + a custom launcher (which is kind of rudimentary ATM) + a few apps from f-droid.org with minimal UI changes to give it a "personalized" look. They provide 50MB cloud storage a mail account (50MB) and that's all there is. Plus a lot of PR. They imply to be the only Google-free Android OS (considering the multitude of roms based on LineageOS this is BS). In all their published material that I've seen there is no mention of LineageOS which is like 95% of their product. So relax, you didn't miss much.

philjohn 8y ago

Because it doesn't happen on all devices - my XS is unaffected for instance.

zaarn 8y ago
QA would normally test hundreds of devices with different settings for each test depending on the test. Testing a single or a low number of samples is a good way to get a bad or insignificant result.1
- saagarjha 8y ago
  Apple at the very minimum tests many thousands of devices before shipping hardware.
  - zaarn 8y ago
    Yes and it makes me curious how they missed such obvious bugs despite thousands of tests.
    - pwinnski 8y ago
      There are many tens of thousands of third-party and no-name chargers, and Apple only tested hundreds, or maybe thousands. The vast majority of chargers, including all chargers for which Apple is responsible, work fine.
      - zaarn 8y ago
        Actually no, Louis Rossman has uploaded a video where an Apple-branded charger observes this behaviour. So no, this doesn't work with a vast majority of chargers, including the ones Apple is responsible for.
        ratsimihah 8y ago
        Many they forgot this particular test case while "many thousands of devices."

ratsimihah 8y ago

I know right? This reminds me of work, when I sometimes send over builds for testing, and they have major edge case bugs. Although "plugged in and locked" isn't really an edge case. Maybe Apple should make its new slogan "it mostly works."

dogma1138 8y ago

I’m not sure there is a real issue behind this this doesn’t happen with the original charger or any good charger it seems without tampering I’ve found 2 devices that don’t auto charge on my XS both have pass through or secondary functionality capabilities one is a cheap charger at work that comes with an SD card reader for some reason and another is when I use my USB OTG dongle with charge passthrough that I use to connect my hardware password wallet (mooltipass) to my phone. All the demonstrations I’ve seen so far were with either non-Apple and no brand at all chargers, often coupled with a USB MITM power meter with data passthrough or some other dongle. This might just be working as intended feature.

zaarn 8y ago
Here is a demonstration with an apple-brand charger: https://www.youtube.com/watch?v=r7tf4JASzlI https://www.youtube.com/watch?v=r7tf4JASzlI A USB Power meter does not inhibit the function of a charger, they only tap the power lines of the USB line with a low value resistor to measure the voltage drop. You can do it with a hall sensor too in which case you don't interrupt the voltage line at all. I see no reason why a current sensor should in any way inhibit a charger's function unless it's deliberate.
- dogma1138 8y ago
  It doesn’t matter if it inhibits or not but how it actually negotiated with the charger, charger only works anything that adds stuff like data or OTG requires an unlock to approve. To me this looks like working as intended.
  - zaarn 8y ago
    The current sense should not add data or OTG, these devices are completely out-of-band for the USB bus. This definitely does not work as intended. Again, all this does is insert a small resistor on one side of the voltage line. It does not tap the data lines, they go straight through to the charger as normal.
    - 8y ago

endorphone 8y ago

You complain about hardware issues (that generally happen to a tiny percentage of users, and are inevitable with a consumer electronic that sells in mass quantities) and then talk about switching to a software platform. It does not follow.