• 1 Post
  • 24 Comments
Joined 1 year ago
cake
Cake day: July 2nd, 2023

help-circle




  • When you’re just shy of a three trillion dollar company big jumps are harder. Plus with the stock up 15% for the month and up 56% for the year people’s, once balanced, portfolios are likely already overly committed to Microsoft based on how it’s performed this year. It’s hard to ignore a stock when it’s performing this well, this consistently, but it puts an investor in a precarious position. If you haven’t already invested based on the AI boom this year has brought, this wouldn’t have been a tipping point. It was a plus, but only a small plus comparatively. The shake up of OpenAI has much more potential for a big sell off for Microsoft than a buying frenzy. This was more silver lining on instability of a big bet Microsoft made.










  • The way companies do it is a lot of napkin math. I worked next to a team that built a service to help other companies figure this out (I provided the sample code and docs they share with customers for onboarding). You plug in some basic info, as an example this building used X kilowatt hours of electricity that the power company says is 10% coal and 90% hydro, which, based on a lookup table that means Y tons of CO2 emissions per hour, add X*Y to your total and move onto the next building. It’s not an exact science measuring actual emissions, more looking for ballpark numbers trying to get rough estimates based on what sustainability consortiums agree is the emission rates for certain things/activities/events. It doesn’t matter if your X is slightly more efficient than your neighbors X, because your maintenance guy is better, both will get the “X” rate for emissions based on the agreed upon value for the thing being measured. The idea is to capture as many things/activities/events as possible to get an estimate of emissions, not a measurement.





  • To use a Reddit example, think of it this scenario:

    Last I recall the most downvoted comment of all time was when a representative from EA said paying $80 to unlock Darth Vader in their recently released Battlefront game was to give players a sense of pride.

    The fact the community was downvoting it with such fervor should have been important feedback to EA. If any platform were to blanket remove posts without review of an overly negative sentiment then EA wouldn’t have known they flew too close to the sun with greed on that comment.

    I do think the idea of downvote removal is a valid one to clear out a lot of garbage, but it removes the community’s voice and could result in easy suppression mechanism of types of content or information by those gaming the system which is why I would vote no on auto-deletion, but maybe leaning yes on triggering review.

    Though in that review there would have to be some guidelines from the mod team in the channel’s sidebar on what content would and wouldn’t be removed. Would a question people where people don’t agree with what is being proposed be deleted even though it’s not a stupid question? That could be an opportunity to learn more. Would a highly racist/sexist/etc question be removed? I’d vote yes, get rid of that troll.

    One last thing on auto-remove: sometimes timezones have differing viewpoints. I clarified some terms that I’m an expert on and most newbies confuse, but I did it at a weird hour of the day for me. I was downvoted pretty heavily for the first 12h of my post, but then it recovered back to zero, then positive. Despite being a worldwide sport different regions have different definitions for the same labels. This is not something any of us in the community had realized until the follow up comments of people coming to my defense which led to a back and forth discovering both sides of the debate were regionally grouped. I still thing my region is right. But that mod review would have been dependent on the mod region, and we may have never gotten an answer, which is why my vote on auto-review is a maybe. I’d want to see a more fully flushed proposal before saying yes.