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

Doordash could just employ delivery drivers to solve this problem. No need for a 100% at will temp work force. They could just hire people part time or full time. And if work is really that surgy they could just have workers on call.

Why does “innovation” have to go hand in hand with fucked up labor practices?


> Why does “innovation” have to go hand in hand with fucked up labor practices?

While still losing a billion dollars a year, or more, for years.


Because we live in a capitalist society and profit is the only objective.

Not exploiting your workforce as much as you can is an almost guaranteed road to failure.

Imagine taking investor money and saying "I'll pay more to my workers so they can have decent lives". They'd oust you in a second. Same if you're in a publicly traded company.

The only possible way this doesn't happen is if you're running an already profitable private company, you're willing to grow at slower rates (which carries its own risks), your market allows it and you're not a sociopath.

That's a very rare combination.


Indeed, it would be better if doordash would just hire full time employees. And vet them before hiring with a regular interview process. Instead of constant punitive surveillance with the threat of instant unemployment.

Can you imagine if your dev job was like this? Picking up jobs to write a single function. And if youre too slow, or one too many tests fail you get sacked.


The “problem” is that many (most?) delivery cyclists in NYC aren’t legally allowed to work, so this option would be jeopardizing a large part of their workforce (bad for DoorDash) while also being bad for the workers themselves (who aren’t exactly swimming in job options). I’m not making a value judgement here, just weighing in with a tradeoff to consider.


    many (most?) delivery cyclists in NYC aren’t legally allowed to work
Woah. Are you saying these people are illegal immigrants, or don't have a work visa? I never heard this before. That is a bold claim. Do you have any evidence to share? I tried Googling, but I couldn't find anything.



>Can you imagine if your dev job was like this?

Mine is. We have stack ranking and it's a living nightmare.


Weekly or monthly stack ranking? I doubt it. Probably annual or twice-yearly reviews. And how much of the stack is released for poor performance during each review? 5/10/15%? Still: Nothing compared to being a DoorDash delivery person.


So does it include health and sick pay for when you get injured or older and you’re less mobile? Cycling all the time could be way more taxing than driving.


Of course it doesn't include any of that. It's "independent contracting".

Though keep in mind they're throttle e-bikes, so there isn't a lot of actual pedaling going on. And as far as accumulated workplace injuries go, merely riding a bike is pretty easy compared to a lot of blue collar work. No, I think what'd get you here is the crashes.


I wouldn’t discount the long term health costs of breathing exhaust fumes all day either. Even as a recreational cyclist or occasional commuter in NYC I noticed it.


Tbh... It's likely still has a better balance then your average office job where you are sitting all day every day.


Sitting in a chair in a metal box is definitely worse for health than cycling all day


Than (e-)biking on roads alongside metal boxes? This ain’t biking through the woods. Just think about the pollution and chance of getting hit.


How often are people on e bikes getting mowed down where you live? What kind of hellscape do you live in where cycling causes you to ingest more polluted air than sitting in your bioweapon-proof, hermetically-sealed, fart-fermenting 3000 kilo Tesla? Sheesh


Are you joking that everywhere is a hellscape, or did you make a typo, or what?

Because anywhere that has any air pollution, you'll ingest more when you're breathing harder and have no air filter.


I’m kinda joking, but I genuinely feel bad for people who live in places where cycling makes their health worse because of the air pollution. I had to deal with that situation a few summers ago when wildfires caused ash to rain from the sky and made the air quality shit. It felt like a low-level hell.


I'm not sure if you noticed that the GP wrote "e-bike". Yes, even riding a e-bike would be more taxing than driving for many, I would say that an e-bike is much less taxing than a regular bicycle.


The act of cycling shouldn't be any sort of problem. The accidents though are another story. The e-bikes are as silent as regular bikes but much faster.


outdoor climbing is also usually a completely different style. I've found that often outdoor routes have way more focus on leg work. and that find a good hand hold is way more determined by how well you can read the route rather than how good the holds are in practise. Also people making the routes outside are often way better climbers so they often don't have a good pulse on grading the lower grades. Plus adjusting a routes grade based on feedback isn't really a thing because lowering the grade of a route after it has been set is quite the insult to the credibility of the person who made the route.


Anyways the whole discussion is moot, tracking of users is illegal. Im sure slavery versus freedom of employment also has a huge impact on the amount of revenue a farm might have. Protections and regulations are usually created for a reason.


not a lot usually. "popular" places like the guardian are way less visited than you might think. In a similar vein a best seller book is only 5000 - 10000 books.


Indeed, though that wasn't AI, that was direct discrimination and racism hidden behind a thin layer of policy.

https://en.wikipedia.org/wiki/Dutch_childcare_benefits_scand...


Indeed to me that's not a ban, but a restriction in the use of possibly risky software until they had the time to draw up guidelines. Honestly everyone working in a company or large organisation aught to have a similar stance. Until we know for sure, don't use it.


It's not that hard to just not stick your sensitive data in. If find it absurd that people would even consider it.

I had a whole presentation today on how important it was to 'scrub' your data before sticking it in a generative model, and I was just like 'hell no', I'm not doing that because data that's supposed to be scrubbed shouldn't go anywhere near the model.


You know since this assignment is basically a boss asking for a single report once about "loyal customers" complexity doesn't matter. let's say worst case scenario it takes 30min to run. who cares about complexity the business value is in the answers from the data.

If you're consistently getting reliable answers and finally decide to build a system for these types of reports, clearly this guy's real world experience at Amazon's Clickstream product is going to be far more valuable than what ever anyone who is brand new to the problem can come up with, even if they choose the "correct" algorithm from the start.

Because I bet you that for most real world products that create more than a single fixed format report you actually want your data setup in a completely different way than what you initially thought. You'll probably learn for example that you want to aggregrate data per week instead of per day. or perhaps you want to link this data to an internal users database, or perhaps your boss wants a notification when new data is added. Or perhaps you'll learn that loading it into a single 1GB SQLite DB solves your problem without even needing to think about any algos.


I have a similar feeling.

My team builds out a lot of one-off projects for customers, and a lot of them don't need to be especially performant. That has always been the case for reporting/analytical projects.

You know what does make a difference though? Amount of development effort spent on the solution. The most performant solution can very easily cost the company much more in terms of development time spent. So in that sense, it's sub-optimal


Its a good metaphor also because if you have an appointment in 20min you would never go to the shop 15min before. Yet managers will force you still out and to the shop because of your time estimate.


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

Search: