When I joined my manager had a few onboarding tasks, I did all of those, and started protoyping my first PR. It was actually a pretty meaty open-ended project and in retrospect it was sort of high-effort, low-reward. I iterated on that several times, wrote proposals, solicited feedback, but ultimately nobody really gives a shit about it, so it hasn't gotten shut down or moved forward.
At this point I've kind of accepted the project is dead but my manager is gone. I've put together a few small PRs but they're similarly just in review hell.
3. They go through a panel to be peer-reviewed and compared to other proposals (similar to how grants are allocated by the government). I'm fairly certain this process also includes review by instrumentation experts. The process is double-blind and the first cycle has already been allocated: https://www.stsci.edu/jwst/science-execution/approved-progra....
They have a tool called APT that lets the researcher define the parameters of the experiment (target object, instrument settings, time, etc) and the software will find an open time slot which can meet those parameters. Then it goes through a double blind scientific review process. The researchers that worked on the telescope hardware get some amount of guaranteed time slots while the rest is used for general observers.
Or we continue to do that ev transition thing but continue to do it with everyone were it is a no brainer first and on the way we solve all the upcoming problems?
And if we continue burning fossil fuels it might never be cold enough for snowstorms any more, problem solved! /S
Who the fuck is this guy writing that stupid ass article... How about you solve the problem of terrible American driving, and lack of winter tyre regulations. Hah, he takes Norway as an example, the country where you have to learn ice driving conditions to get your license..
Tilling and fertilizing gives you a great yield today, but damages the long term fertility of the land, and pollutes nearby waterways. The best analogy is buying things on credit. You might feel rich at first, but when the debt comes due and you can't afford it bankruptcy is a bitch.
It does require letting some land lie fallow to regeneration, so you need to devote more total land to agriculture, and it also requires fencing and for someone to go move the livestock to a new paddock every few days (though you save on tilling/herbicide application costs). On the plus side the unit cost of everything you produce is lower, and meat is typically much more profitable than vegetables so you've diversified.
Yeah, it's not like they will label something a train just because a single person says so. But if you have 10k responses with 95% confidence saying it's a train, it's very likely to be the case.
For unambiguous images almost all humans will label them the same way. For ambiguous ones humans will differ. Presumably they'll accumulate stats on each image and will be able to detect cases like this.
unless a properly obfusicated bot net has seeded the data set with -everything is a train- responses to the tune of >>10k responses with 95% confidence saying it's a train<<
I don't think that anyone who cares about specific CPU features would not know that it shouldn't work and therefore wouldn't risk getting it removed later
Everyone else would not know that this might have a performance impact.
Do you don't work in a Team?
Or aren't you building up a team or org?
Either it's your position to see issues and solve them for the org or it's your job to execute.
If you can't come up with things to do, but it's your job, you should search for a team or product and not for a company.