Hacker Newsnew | past | comments | ask | show | jobs | submit | imperfectcats's commentslogin

That's rather ironic, because I read your comment you linked to, and it is pure speculation.

The Goodreads dev said "the site was originally built as a giant pile of Rails spaghetti with views mixed with business logic and such and then a fuck ton of weird features built and left to sit there". I'm not sure why you think that is or ever was "the Rails way", but it seems more like the classic "move fast and break stuff" startup engineering than anything resembling rails to me.


Until they ant to retire. Then what do you do?

If you are happy to walk away and hand the "business they work at" for free to someone, flame on. Thing is, unless that is family, who would you give it to and why?


Ditto - and the same with the claims of solar and wind companies because of - [Goodhart's Law - https://en.wikipedia.org/wiki/Goodhart%27s_law

      - "When a measure becomes a target, it ceases to be a good measure"
But even if we take that view, how many deaths would nuclear have to have caused to break even with fossil fuels? https://ourworldindata.org/grapher/death-rates-from-energy-p... puts nuclear at 0.03 and hydropower at 1.3. So it is ~50 times less than hydro. That's a lot of wiggle room!


I wish regulation worked that way! Regulations change all the time - if not in law certainly in practice/interpretation.

That is what makes industries with heavy regulation such risky investments.


I searched for info on the cost overruns and it appears that management failures contributed significantly. An anti-nuke lawsuit didn't help but appeared to be $1B in impact.

We see consistently that big projects never come in on time and budget, which is why SMRs are the only economical way forward.


Goodhart's Law: https://en.wikipedia.org/wiki/Goodhart%27s_law

"When a measure becomes a target, it ceases to be a good measure"

Any ESG system that uses a calculated number composed of many areas will quickly fall prey to Goodhart's Law. With an E, an S and a G to work on, the areas for change are numerous enough that Goodhart's Law holds even in a non-cynical world.

When an edict that "We need to increase our X score this quarter" is handed down, the real, living people who have to improve the ESG rating are likely limited in what they can do. A combination of pre-existing contracts and commitments, difficulty in changing large systems and problems with multi-department co-ordination means that most changes will be small and isolated, i.e. things where one department or small team can implement the entire change.


I'm working adjacent to the sustainability space, and I think Goodhart's law applies to a good degree, however a bad measure is often still more useful than no measure at all. It at least enables us to show flaws, develop better indicators and metrics.

When it comes to human rights in the supply chain and environmental impact, there was a complete lack of data in many cases. With the ESG, we see a major shift where companies are now scrambling to gather said data, analyse it and often also publish it.


> are likely limited in what they can do.

If they aren't suffering under these rules, which will almost be a certainty.


Obviously KPIs should be (and probably are) chosen specifically with this effect in mind.


I wish articles like this had prominent dates!


They don't do they?

By my count all the non-front end devs do back end, so Client Applications (2), Core Services/Platform (1) & Dev Ops/Infra (1)


Can you point me to the section that says that: https://assets.documentcloud.org/documents/3914586/Googles-I...

Being as generous with the interpretation as possible, I guess suggesting new programs to replace existing ones could be interpreted as "programs that are aimed at a helping women and people of color should be ended", as ending one and replacing it is kind of ending. Saying "repeatedly" seems a stretch though.

Can you explain what I'm missing that lead to your beliefs?


Under the harms section first bullet for things to end: "Programs, mentoring, and classes only for people with a certain gender or race"


OK, so you read that section and see it as "repeatedly said programs that are aimed at a helping women and people of color should be ended"?

There is another section with programs Damore claimed would be more effective. I am confused why replacing programs equates to ending them. Unless the goal is only to have good intentions, shouldn't the effectiveness of programs matter?


My interpretation:

He uses that to suggest the programs increase racial tensions, not to suggest that we shouldn't be trying to get more people of color involved in tech. He cites this article [0] as his reasoning.

Is there something I'm missing?

[0]: http://archive.is/00yb5


Is there a name for that, where whatever the limit is, people set things right at it? The same is true in progressive tax brackets, and the number of cancer cases that people get right when people are eligible for Medicare in the USA.

Seems like it is common enough to have a name.


> The same is true in progressive tax brackets

I might misunderstand you, but progressive tax brackets mean you pay extra money only on money earned after the limit. So if that line is $1000, under which the tax rate is 10%, you pay $100. Then if the tax rate above that is 20%, I pay that much only on the amount above $100. So if I make $1020, I pay 10% * $1000 + 20% * $20 = $102. (Not 20%*1020 = $204)

In other words, you can’t take home less money overall just because you crossed the line into a higher tax bracket. I felt it was worth mentioning because it’s an easy misconception to have.


Then that is tomorrow me's problem!

I honestly think a lot of thinking works that way, where we'll let tomorrow me worry about it. To be fair, a lot of the time it makes sense. The problems a solution creates are things we need to accept when alternative is worse (think global warming vs starvation circa 1770).


I honestly think that if you are involved in an operation like this and you bring this attitude to work that you'll be fired before the day is out.

This is the kind of job you want to get right the first time.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: