Comment by howlgarnish

6 years ago

Insightful comment from the thread:

Yeah, I think part of the problem is that Facebook has never had an actual product which people pay for and expect to work. Their whole culture and set of priorities are counter to this.

Unfortunately I get the impression that this is how you get proper support from Facebook or Google. Make a Reddit or Hacker News post, get a few upvotes, and someone with influence inside the company files an "Oops Help My Friend" support ticket or contacts another person with the right authority to take a look.

  • In the Catholic theology, there's a concept of Saints, who are basically humans that intercede with God on the behalf of ordinary mortals. People will sometimes address prayers to saints in the hope that the prayer will be more effective. Probably similar in other religions too but I'm pulling from memories of catholic school

  • I wonder if there is a way to incentivize the behavior of "I work at a big company and have the ability to help out a stranger with a problem"? Perhaps people who help others with these issues (which are public by nature of being discussed on reddit/twitter/etc) can put it as an item on their resume when looking to switch jobs - something like "I care about the community and people who use the products I help build. I advocated for [this person (with link to public forum)] which ultimately resulted in [this solution (with link to release note if possible)]."

    Of course there is a discussion to have about incentivizing the _wrong_ behavior (the need to create publicity around issues that shouldn't exist in the first place / only people with "clout" become the ones who get their issues resolved).

    • There are groups inside Facebook to raise issues like this. I can guarantee this one has already been escalated.

      They wouldn't allow just any employee to "fix" these issues. Too often there is a lot of context that is omitted from the shared story, so it's not just a matter of hitting a button to remove a block.

    • It's incentivized like this:

      [At a party] A: What do you do? B: I work for $BIGTECH. A: You suck dude. I bought one of their products and it never worked once.

      2 replies →

    • I assume that quite often those are marketing & PR people who use special tools to track every mention of a product / company name. Many companies (especially niche ones) buy those services that basically track the whole internet - so their representatives can comment on issues.

      Bigger companies seem not to care that much, apart from maybe twitter? And those few thousand upvote posts on reddit.

    • Couple of pop psych thoughts:

      People on HN are not really strangers, we are part of a community. By helping one of us, they help one of their in groups. Not helping might devalue their investment in this place.

      HN users are the early adopters that small group of users who have much more influence early on in projects. Care of early adopters makes sense.

    • I wonder if there is a way to incentivize the behavior of "I work at a big company and have the ability to help out a stranger with a problem"?

      When the coin in the coffer rings / An account from bannination springs!

    • You don't want to incentivise this too much or people will start taking bribes.

      Also, "I helped a member of the public defeat my organisation's policy" is a negative on the resume not a positive. Sadly.

      1 reply →

  • Unfortunately for each of these cases going public there will be about a hundred poor souls being unnoticed.

  • This has happened with me re: Intuit, Microsoft and Adobe as well - all companies with proper support channels - issue resolution/escalation for people with the "right contacts" goes about 100x faster.

    What about the 99.9% of the rest who don't have those contacts?

  • Anecdotal, but.. I got dicked around by Wayfair for almost two months (stalled order, various excuses, well meaning but essentially powerless customer support). In the end all it took to resolve was one tweet at their account. Next day the order cleared.

That is a fantastic comment. I was reading through Google SRE stuff. Lots of good info. Lots of very, very bad ideas too. Lots of it is "at scale, you can't make everyone happy, so don't try." You end up with SLOs and SLIs based on percentages of availability, errors, etc. If 99.99% of requests are good, don't worry about the rest. However, at scale, 0.01 can be literally millions of real, breathing people with feeling and priorities and things to do. When you give away your product for free, these are good trade offs. When you sell your product, the same assumptions may not apply.

I cannot afford to ignore 0.01% of incoming requests because we charge per request. At a certain point, yes, money and pros and cons yadda yadda yadda. It just feels that, once again, people are saying, "but Google does it!" -- and that does not mean you should.

  • 100% is unachievable, especially at Google scale. Your car or coffee shop or whatever also fails 0.0x% of the time. Heck the energy net in western countries doesn't reach 99.999% uptime and I would call it very dependable.

    I'd say the biggest problem it the non-existent support. It's fine if 0.01% fails, as long as there is human troubleshooting or help available if necessary.

    • The distribution is important.

      If 100% of users are experiencing 0.01% random errors it probably isn't worth bothering about.

      If 0.01% of users are experiencing 100% errors than it is much more important.

      1 reply →

  • I think you are very confused about how availability is tracked. If a class of users is always getting an error this would be investigated. There is literally no way, at any scale, to guarantee 100% availability.

    • >I think you are very confused about how availability is tracked. If a class of users is always getting an error this would be investigated.

      Oh, please, you are confused. I worked at Google, most teams just don’t have metrics like this.

      While I don’t think it’s acceptable, reality is that it’s expensive to compute such metric, i.e. you need to slice you monitoring stream per each user, has privacy implications, and often it’s unclear what action should be taken.