4 ms·

Imagine waiting 16 years for a PR :/

pengaru 4y ago
> Imagine waiting 16 years for a PR :/ Imagine writing bug-free code put in continuous, mass population-critical use for the better part of a century.
- robertlagrant 4y ago
  That would be the analog of a fully autonomous power station. Plenty of people in current ones to watch and patch things!
- tchaffee 4y ago
  > Imagine writing bug-free code Can't possibly go tits up. These new reactors are surely infallible, right?
  - Turing_Machine 4y ago
    Nuclear power has been in use in the United States for over 70 years without a single fatality to a member of the general public. Better than any other energy source. A few plant and supply chain workers have been killed, but again, fewer than with any other source (mining coal and climbing on roofs and towers to install wind and solar is dangerous, yo). Since I've been muzzled again: > Climbing on the roof of a two story house to install solar panels is considerably less dangerous than building a nuclear power plant. Yo. In terms of workers killed per watt-hour generated? No, it isn't. It's not even close. Yo.
    - tchaffee 4y ago
      The claim was not "relatively safe compared to other forms of energy". The claim was bug free. Which -- if you want to use past history -- is false. > climbing on roofs and towers to install wind and solar is dangerous, yo Climbing on the roof of a two story house to install solar panels is considerably less dangerous than building a nuclear power plant. Yo. > without a single fatality to a member of the general public. No immediate fatality is the only accurate claim you can make. Over 150,000 people were evacuated from the Three-Mile Island disaster. Due to the stress, there were some increases in consumption of alcohol, cigarettes, and tranquilizers immediately following the accident. The eventual long term effects of those behaviors surely resulted in shorter life spans for at least some of the people. https://www.sciencedirect.com/science/article/abs/pii/0038012184900181?via%3Dihub https://www.sciencedirect.com/science/article/abs/pii/003801... > for over 70 years That's a short time. Unless you can show the US is somehow exceptional in its quality controls compared to Japan, then a huge disaster like Fukushima is inevitable, and life will be lost. No one complains about the dangers of the neighbors who use solar energy. No one wants a reactor in their backyard. And it has fair reasoning behind it. You'll never have to evacuate 150,000 people due to your neighbors' solar panels failing.
    - tchaffee 4y ago
      > In terms of workers killed per watt-hour generated? > No, it isn't. It's not even close. Yo. This source says you are wrong. https://ourworldindata.org/grapher/death-rates-from-energy-production-per-twh https://ourworldindata.org/grapher/death-rates-from-energy-p... What source were you using when making your claim?
      - philipkglass 4y ago
        The rate used to be higher for solar, because it used to be the case that most solar modules were installed in small rooftop systems. Now most solar power comes from large ground level solar farms. In 2020, 68% of US solar generation came from utility-scale farms [1]. Over time the solar PV deaths-per-TWh ratio has decreased in 3 ways: 1) Ground level installations have a greatly diminished risk of dangerous falls during construction/maintenance and have come to account for more wattage than rooftop systems. 2) Installing a solar panel in a large farm instead of on a roof yields more energy output per year, since it usually has mechanical sun tracking [2] and will be oriented for optimum output even if installed with a fixed tilt. 3) Newer panels generate more annual energy output per panel, per kilogram, and per square meter. That means that the same number of full time solar installation workers now produce more energy over the system's lifetime than in the past. The information from OurWorldInData that you linked is more up to date and therefore shows low deaths-per-TWh for solar. Probably Turing_Machine was remembering an older report like this one from 10 years ago: https://www.forbes.com/sites/jamesconca/2012/06/10/energys-deathprint-a-price-always-paid https://www.forbes.com/sites/jamesconca/2012/06/10/energys-d... [1] https://www.nrel.gov/docs/fy22osti/81325.pdf https://www.nrel.gov/docs/fy22osti/81325.pdf page 9 [2] https://pv-magazine-usa.com/2017/09/20/trackers-dominate-u-s-utility-scale-solar-wcharts/ https://pv-magazine-usa.com/2017/09/20/trackers-dominate-u-s...
        Turing_Machine 4y ago
        > Probably Turing_Machine was remembering an older report No, I was talking about the United States. Not Chernobyl.
        4y ago
sidewndr46 4y ago
Imagine all the dependency rot in the 16 years.
epistasis 4y ago
The delays have not come from approval reasons, but from construction mess ups, as well as design flaws (ie unconstructable designs, not designs that would be unsafe if constructed).
- sroussey 4y ago
  Actually, the rules change every couple of years, which requires change to things already constructed. It’s amazing that the team was able to deal with that and still get it done.
  - epistasis 4y ago
    What rules have changed in the past five years, when there have been the announcements of delays? I have been following closely, and never once have I heard of a delay because of regulation changes. Not once! It's all construction. And with the costs of delays, blaming this on an external party like the NRC would be an essential thing to do to explain it to the backers. If you have any knowledge of regulations that caused the big delays, I would love to know, and would be very surprised.
    - grandinj 4y ago
      It's not that the rules change, it's that the rules (as written) are open to interpretation. So if you design something, get it approved, start building, then get the NRC inspector on-site to confirm, and he disagrees with your interpretation, you get to rip out that piece and rebuild it, with all the attendant delays and costs.
      - epistasis 4y ago
        So what sort of interpretation are we talking about here? Welding quality? Concrete qualities?
        grandinj 4y ago
        No, if the rules specified that level of detail, they would probably be all right. The NRC rules appear to be higher level, and built around specifying "safety methodologies", and it's the translation of those safety methodologies to concrete (haha) level detail that gets messy. Although to fair on the NRC, there was a case where, due to lack of experience, the contractor cast an enormous slab using the wrong rating of concrete, which then had to be removed at great expense and delay.
toast0 4y ago
Imagine having good enough specifications that a 16 year old change was still valid. :P I've seen 16 year old bugs get fixed though. Usually just need to get them retriaged as a security bug, and then all things are fixable.
- kelnos 4y ago
  Funny you mention that; a couple weeks ago I fixed a 16-year-old bug that I myself wrote, in some open source software I started maintaining again after a long hiatus.
- thrdbndndn 4y ago
  Funny everyone thinks it's pull request. I thought they meant American Permanent Resident (aka green card), which has a notoriously long waiting period.
themitigating 4y ago
It's not a change to an existing system It's building the entire thing. Also the consequences are significantly worse than a software bug in most cases

> Imagine waiting 16 years for a PR :/ Imagine writing bug-free code put in continuous, mass population-critical use for the better part of a century.

robertlagrant 4y ago
That would be the analog of a fully autonomous power station. Plenty of people in current ones to watch and patch things!
tchaffee 4y ago
> Imagine writing bug-free code Can't possibly go tits up. These new reactors are surely infallible, right?
- Turing_Machine 4y ago
  Nuclear power has been in use in the United States for over 70 years without a single fatality to a member of the general public. Better than any other energy source. A few plant and supply chain workers have been killed, but again, fewer than with any other source (mining coal and climbing on roofs and towers to install wind and solar is dangerous, yo). Since I've been muzzled again: > Climbing on the roof of a two story house to install solar panels is considerably less dangerous than building a nuclear power plant. Yo. In terms of workers killed per watt-hour generated? No, it isn't. It's not even close. Yo.
  - tchaffee 4y ago
    The claim was not "relatively safe compared to other forms of energy". The claim was bug free. Which -- if you want to use past history -- is false. > climbing on roofs and towers to install wind and solar is dangerous, yo Climbing on the roof of a two story house to install solar panels is considerably less dangerous than building a nuclear power plant. Yo. > without a single fatality to a member of the general public. No immediate fatality is the only accurate claim you can make. Over 150,000 people were evacuated from the Three-Mile Island disaster. Due to the stress, there were some increases in consumption of alcohol, cigarettes, and tranquilizers immediately following the accident. The eventual long term effects of those behaviors surely resulted in shorter life spans for at least some of the people. https://www.sciencedirect.com/science/article/abs/pii/0038012184900181?via%3Dihub https://www.sciencedirect.com/science/article/abs/pii/003801... > for over 70 years That's a short time. Unless you can show the US is somehow exceptional in its quality controls compared to Japan, then a huge disaster like Fukushima is inevitable, and life will be lost. No one complains about the dangers of the neighbors who use solar energy. No one wants a reactor in their backyard. And it has fair reasoning behind it. You'll never have to evacuate 150,000 people due to your neighbors' solar panels failing.
  - tchaffee 4y ago
    > In terms of workers killed per watt-hour generated? > No, it isn't. It's not even close. Yo. This source says you are wrong. https://ourworldindata.org/grapher/death-rates-from-energy-production-per-twh https://ourworldindata.org/grapher/death-rates-from-energy-p... What source were you using when making your claim?
    - philipkglass 4y ago
      The rate used to be higher for solar, because it used to be the case that most solar modules were installed in small rooftop systems. Now most solar power comes from large ground level solar farms. In 2020, 68% of US solar generation came from utility-scale farms [1]. Over time the solar PV deaths-per-TWh ratio has decreased in 3 ways: 1) Ground level installations have a greatly diminished risk of dangerous falls during construction/maintenance and have come to account for more wattage than rooftop systems. 2) Installing a solar panel in a large farm instead of on a roof yields more energy output per year, since it usually has mechanical sun tracking [2] and will be oriented for optimum output even if installed with a fixed tilt. 3) Newer panels generate more annual energy output per panel, per kilogram, and per square meter. That means that the same number of full time solar installation workers now produce more energy over the system's lifetime than in the past. The information from OurWorldInData that you linked is more up to date and therefore shows low deaths-per-TWh for solar. Probably Turing_Machine was remembering an older report like this one from 10 years ago: https://www.forbes.com/sites/jamesconca/2012/06/10/energys-deathprint-a-price-always-paid https://www.forbes.com/sites/jamesconca/2012/06/10/energys-d... [1] https://www.nrel.gov/docs/fy22osti/81325.pdf https://www.nrel.gov/docs/fy22osti/81325.pdf page 9 [2] https://pv-magazine-usa.com/2017/09/20/trackers-dominate-u-s-utility-scale-solar-wcharts/ https://pv-magazine-usa.com/2017/09/20/trackers-dominate-u-s...
      - Turing_Machine 4y ago
        > Probably Turing_Machine was remembering an older report No, I was talking about the United States. Not Chernobyl.
        4y ago

sidewndr46 4y ago

Imagine all the dependency rot in the 16 years.

epistasis 4y ago

The delays have not come from approval reasons, but from construction mess ups, as well as design flaws (ie unconstructable designs, not designs that would be unsafe if constructed).

sroussey 4y ago
Actually, the rules change every couple of years, which requires change to things already constructed. It’s amazing that the team was able to deal with that and still get it done.
- epistasis 4y ago
  What rules have changed in the past five years, when there have been the announcements of delays? I have been following closely, and never once have I heard of a delay because of regulation changes. Not once! It's all construction. And with the costs of delays, blaming this on an external party like the NRC would be an essential thing to do to explain it to the backers. If you have any knowledge of regulations that caused the big delays, I would love to know, and would be very surprised.
  - grandinj 4y ago
    It's not that the rules change, it's that the rules (as written) are open to interpretation. So if you design something, get it approved, start building, then get the NRC inspector on-site to confirm, and he disagrees with your interpretation, you get to rip out that piece and rebuild it, with all the attendant delays and costs.
    - epistasis 4y ago
      So what sort of interpretation are we talking about here? Welding quality? Concrete qualities?
      - grandinj 4y ago
        No, if the rules specified that level of detail, they would probably be all right. The NRC rules appear to be higher level, and built around specifying "safety methodologies", and it's the translation of those safety methodologies to concrete (haha) level detail that gets messy. Although to fair on the NRC, there was a case where, due to lack of experience, the contractor cast an enormous slab using the wrong rating of concrete, which then had to be removed at great expense and delay.

toast0 4y ago

Imagine having good enough specifications that a 16 year old change was still valid. :P I've seen 16 year old bugs get fixed though. Usually just need to get them retriaged as a security bug, and then all things are fixable.

kelnos 4y ago
Funny you mention that; a couple weeks ago I fixed a 16-year-old bug that I myself wrote, in some open source software I started maintaining again after a long hiatus.
thrdbndndn 4y ago
Funny everyone thinks it's pull request. I thought they meant American Permanent Resident (aka green card), which has a notoriously long waiting period.

themitigating 4y ago

It's not a change to an existing system It's building the entire thing. Also the consequences are significantly worse than a software bug in most cases