4 ms·

As with most system failures, there is no (one) "root cause": The intern developer should not have committed the middle finger cursor. A code reviewer should

As with most system failures, there is no (one) "root cause": The intern developer should not have committed the middle finger cursor. A code reviewer should not have approved the code change. If there was no code review (and there probably wasn't during the Windows 3.1 era), especially of an intern's code, then that's another problem with the system. QA should not have approved the beta build with a bug that could impact the company's reputation. A bug report was filed, but whoever triaged the bug report didn't escalate the issue before the beta release. Finally, the engineering manager probably should not have rehired the intern as a full-time employee.

watwut 4y ago
This was not root cause. This is analysis of contributing factors. Root cause of middle finger cursor is individual who made middle finger cursor. That review could have caught it does not make the above less of actual root cause.
- gweinberg 4y ago
  No, that's backwards. Sloppy qa was the root cause of this getting to the beta testers, the fact that it was a prank rather than a mistake is irrelevant. Microsoft was lucky their sloppy qa was reveled by something so innocuous instead of by a genuinely harmful commit.
  - UnpossibleJim 4y ago
    >>If you read the post, you'd see the QA org saw that Easter Egg, took it in the right spirit (like they should), and signed it okay for beta release (mistakenly, I'd presume) followed by everyone else in the bureaucratic chain that might have okay'd it as well.<< As mentioned above, there were a string of failures. Don't just blame QA for finding the bugs that don't get fixed. If it gets signed off, it gets signed off, there are a bunch of departments that can look at the bug database as well. And, in the end, 9/10 of the people who found this people probably loved the fact that Microsoft finally showed a little bit of soul.
  - watwut 4y ago
    Sloppy QA does not causes pranks and bugs. And given that this is forum for developers, developers dont get to blame QA for pranks originating in development. Seriously.
ok_dad 4y ago
> Finally, the engineering manager probably should not have rehired the intern as a full-time employee. In your org there is no room for mistakes, like humans will routinely make; sounds like hell.
snickerer 4y ago
I disagree with the last sentence. Why should I not hire a person that does pranks (as a general rule)? For a creative job, like an engineering job, I want creative, humorous, witty, and interesting people. Someone who does funny pranks at the right time gets a pro-hiring indication mark from me. And another argument for the prankster is that we need fun to stay sane in the gray industrial business world.
Dylan16807 4y ago
> should not have committed the middle finger cursor. > should not have approved the code change Disagree. > impact the company's reputation In a bad way? > A bug report was filed, but whoever triaged the bug report didn't escalate the issue before the beta release. Yeah probably a problem. > Finally, the engineering manager probably should not have rehired the intern as a full-time employee. Jesus, learn to handle a joke.
- ufmace 4y ago
  I wouldn't necessarily fire an intern for not knowing it, but they should have been more closely watched by someone with more wisdom. Anyone who's been around the block a few times knows it's not a good idea to put potentially offensive or unprofessional things into the product, including in test environments, code comments, or other things that aren't supposed to ever make it out of the company. There's a lot more ways for that sort of thing to end up in front of a customer than you could ever predict, and it's not worth the reputation hit to the company. I would agree that they should still re-hire him. This poor fellow probably learned the lesson that there's a time and place to joke around plenty well enough now.
alpaca128 4y ago
> If there was no code review (and there probably wasn't during the Windows 3.1 era) Just recently an ex MS employee talked about this topic on YouTube in the context of possible backdoors [0]. Some developers were each responsible for a component of the system and would indeed check code changes, which led to an intern not getting a job at MS after he decided to integrate an easter egg. [0] https://youtu.be/CR7i1UfBtQM?t=10 https://youtu.be/CR7i1UfBtQM?t=10
ignoramous 4y ago
Fair. > Finally, the engineering manager probably should not have rehired the intern as a full-time employee. Slap on the wrist? Sure. Fix the qa and bug triage process? Be my guest. Fire the intern? A bit extreme for my taste.

watwut 4y ago

This was not root cause. This is analysis of contributing factors. Root cause of middle finger cursor is individual who made middle finger cursor. That review could have caught it does not make the above less of actual root cause.

gweinberg 4y ago
No, that's backwards. Sloppy qa was the root cause of this getting to the beta testers, the fact that it was a prank rather than a mistake is irrelevant. Microsoft was lucky their sloppy qa was reveled by something so innocuous instead of by a genuinely harmful commit.
- UnpossibleJim 4y ago
  >>If you read the post, you'd see the QA org saw that Easter Egg, took it in the right spirit (like they should), and signed it okay for beta release (mistakenly, I'd presume) followed by everyone else in the bureaucratic chain that might have okay'd it as well.<< As mentioned above, there were a string of failures. Don't just blame QA for finding the bugs that don't get fixed. If it gets signed off, it gets signed off, there are a bunch of departments that can look at the bug database as well. And, in the end, 9/10 of the people who found this people probably loved the fact that Microsoft finally showed a little bit of soul.
- watwut 4y ago
  Sloppy QA does not causes pranks and bugs. And given that this is forum for developers, developers dont get to blame QA for pranks originating in development. Seriously.

ok_dad 4y ago

> Finally, the engineering manager probably should not have rehired the intern as a full-time employee. In your org there is no room for mistakes, like humans will routinely make; sounds like hell.

snickerer 4y ago

I disagree with the last sentence. Why should I not hire a person that does pranks (as a general rule)? For a creative job, like an engineering job, I want creative, humorous, witty, and interesting people. Someone who does funny pranks at the right time gets a pro-hiring indication mark from me. And another argument for the prankster is that we need fun to stay sane in the gray industrial business world.

Dylan16807 4y ago

> should not have committed the middle finger cursor. > should not have approved the code change Disagree. > impact the company's reputation In a bad way? > A bug report was filed, but whoever triaged the bug report didn't escalate the issue before the beta release. Yeah probably a problem. > Finally, the engineering manager probably should not have rehired the intern as a full-time employee. Jesus, learn to handle a joke.

ufmace 4y ago
I wouldn't necessarily fire an intern for not knowing it, but they should have been more closely watched by someone with more wisdom. Anyone who's been around the block a few times knows it's not a good idea to put potentially offensive or unprofessional things into the product, including in test environments, code comments, or other things that aren't supposed to ever make it out of the company. There's a lot more ways for that sort of thing to end up in front of a customer than you could ever predict, and it's not worth the reputation hit to the company. I would agree that they should still re-hire him. This poor fellow probably learned the lesson that there's a time and place to joke around plenty well enough now.

alpaca128 4y ago

> If there was no code review (and there probably wasn't during the Windows 3.1 era) Just recently an ex MS employee talked about this topic on YouTube in the context of possible backdoors [0]. Some developers were each responsible for a component of the system and would indeed check code changes, which led to an intern not getting a job at MS after he decided to integrate an easter egg. [0] https://youtu.be/CR7i1UfBtQM?t=10 https://youtu.be/CR7i1UfBtQM?t=10

ignoramous 4y ago

Fair. > Finally, the engineering manager probably should not have rehired the intern as a full-time employee. Slap on the wrist? Sure. Fix the qa and bug triage process? Be my guest. Fire the intern? A bit extreme for my taste.