4 ms·

I have to play devil's advocate here: I finally have enough bandwidth for my server, so now that it's working I should reduce it. This makes sense? You've sai

by 51Cards 13y ago

I have to play devil's advocate here: I finally have enough bandwidth for my server, so now that it's working I should reduce it. This makes sense? You've said that now it's working we should start to take it away? I for one kind of like the fact they have all failed. I would like to keep this track record up since I spend a lot of time on planes. Just my 2c.

sanderjd 13y ago
To devil's advocate your devil's advocate with a similar analogy: Ok so I've deployed an app to heroku. I'm using 2 dynos, but suddenly my app becomes fairly popular and a bunch of people start getting timeouts. Oh no, better up the dynos to 20. Now no more timeouts but pretty expensive. Seems like now is the time to start benchmarking and figuring out what the right balance is. I think we're starting to see some of that sort of benchmarking taking place with the TSA, so that's good, but the incentives don't seem like they're lined up properly for the bureaucracy to shrink significantly, regardless of the true level of risk.
- burke 13y ago
  This is essentially the problem with all government agencies. They're never motivated by efficiency, but rather by justifying as large a budget as possible. If your hypothetical heroku scenario were a government project, they'd be proposing a "conservative" increase to 23 dynos for next year.
anonymoushn 13y ago
This is a reasonable argument, but we should pay attention to both the good and bad effects when tallying the value produced by the TSA. After adding up all of the terrorist attacks prevented by enhanced screening (zero?), we should be sure to subtract the wasted hours of people waiting at airports and the deaths caused by increased use of cars for moderate to long distance travel.[0] In the unlikely event that our analysis finds that the TSA saves lives, we should then compare it to other methods of saving lives and be sure to implement only the most cost-effective ones, potentially shutting down programs that are not as cost-effective. [0]: https://www.schneier.com/blog/archives/2013/09/excess_automobi.html https://www.schneier.com/blog/archives/2013/09/excess_automo...
chrismcb 13y ago
These failed attempt's failures had nothing to do with TSA and security.

by 51Cards 13y ago

sanderjd 13y ago

To devil's advocate your devil's advocate with a similar analogy: Ok so I've deployed an app to heroku. I'm using 2 dynos, but suddenly my app becomes fairly popular and a bunch of people start getting timeouts. Oh no, better up the dynos to 20. Now no more timeouts but pretty expensive. Seems like now is the time to start benchmarking and figuring out what the right balance is. I think we're starting to see some of that sort of benchmarking taking place with the TSA, so that's good, but the incentives don't seem like they're lined up properly for the bureaucracy to shrink significantly, regardless of the true level of risk.

burke 13y ago
This is essentially the problem with all government agencies. They're never motivated by efficiency, but rather by justifying as large a budget as possible. If your hypothetical heroku scenario were a government project, they'd be proposing a "conservative" increase to 23 dynos for next year.

anonymoushn 13y ago

This is a reasonable argument, but we should pay attention to both the good and bad effects when tallying the value produced by the TSA. After adding up all of the terrorist attacks prevented by enhanced screening (zero?), we should be sure to subtract the wasted hours of people waiting at airports and the deaths caused by increased use of cars for moderate to long distance travel.[0] In the unlikely event that our analysis finds that the TSA saves lives, we should then compare it to other methods of saving lives and be sure to implement only the most cost-effective ones, potentially shutting down programs that are not as cost-effective. [0]: https://www.schneier.com/blog/archives/2013/09/excess_automobi.html https://www.schneier.com/blog/archives/2013/09/excess_automo...

chrismcb 13y ago

These failed attempt's failures had nothing to do with TSA and security.