Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Mirroring US climate data

climatemirror.org

Post reply on HN
Print view View original

51–60 of 166 posts

Re: Mirroring US climate data

#51

Post by andygates »

danbruc wrote: post #26
crispinb wrote: post #18

Earlier quoted context omitted.

If you only create redundancy when you are sure it's necessary, you're too late.

Of course and I think having those mirrors is a good thing, stuff disappears from the internet all the time. But that is not the question I am asking. We could have started mirroring climate data and all kinds of other data last year or five years ago, there is always a risk of losing data. But this has obviously been triggered by the Trump presidency and is focused on climate data which implies that people are now m…

The specific problem is difficult to quantify: he's clearly anti-science and a loose cannon, and he's in control. Where a normal handover may include a brief news blackout during transition, this is all a bit Arbitrary Diktats Of Dear Leader and people are jumpy as heck.

Data resilience in the case of a rogue executive is an unusual thing to come across, but it can't be unique. Executives change, and organization data needs to persist beyond the current board.

Re: Mirroring US climate data

#53

Post by tracker1 »

Tempest1981 wrote: post #45

I heard that Trump has given NOAA 100 days to justify their existence, or face heavy cuts. Does anyone know the details? Here is the closest article I could find: http://www.salon.com/2016/11/23/politicizing-climate-change-...

It seems like Trump has been doing similar to quite a few federal agencies. I'm concerned with some of the early moves, but it's going to be interesting, I do hope that some common sense prevails. For example, bowing out of TPP isn't so bad.

Re: Mirroring US climate data

#54

Post by rmc »

jl6 wrote: post #41
fsttx wrote: post #38

Another useful tactic is to archive news and commentary (left and right) about current political events, both for accountability and perhaps, to prevent the rewriting of history in the era of "alternative facts".

And not just archive it, but sign it, time stamp it, verify it, etc.. I see little to prevent someone using ML to auto-generate archives worth of fake history. Imagine a purported DB dump of Reddit with billions of subtlely fraudulent comments scattered throughout.

Is there any good way to timestamp things like this?

Re: Mirroring US climate data

#55

Post by tracker1 »

esaym wrote: post #25
trothamel wrote: post #2

Can I be the first to claim that the people running the mirror are changing the data for their own nefarious purposes? Not that I believe this or anything, but come on - you know that accusation is going to go around. It might make sense to use a remote server to timestamp files as they're uploaded, or something like that.

I would say the same thing. Do you think you will really convince a climate denier with this data that you copied off of the Internet that, was copied from a gov server, on to another server and then torrented and copied to another server and then downloaded onto you PC... doubt it.

It's worth noting that not everyone who rebuffs some of the proposed actions is a climate denier. There's still some question as to the extent of human involvement. That said, smog is enough for me to be in favor of working towards renewable, less polluting energy sources. In my lifetime I've gone from having allergy-like flare ups once every few years, to now 2-3 times a year, mostly seems to follow the pollution levels not just the pollen count.

By the same token, not every "green" program is a great thing... and even in treaties, exempting the worst actors (China, India) because they are often considered third world doesn't do a lot to help the global issues. Improved working requirements as a requisite for international trade deals would go a long way towards "cleaning up."

I bring this up, as I tend to see a lot of people who question the approach, intensity, and cost getting lumped in with climate deniers.

Re: Mirroring US climate data

#56

Post by tracker1 »

pdonis wrote: post #7
nirav72 wrote: post #3

Earlier quoted context omitted.

When there is a broad consensus across the scientific community about climate change, does it even matter at this point if the data has been changed? The U.S isn't the only country that collects this data. So even if changed, it doesn't matter.

> When there is a broad consensus across the scientific community about climate change, does it even matter at this point if the data has been changed? It does if the data that was changed is the data that was used to support the consensus. More generally, all data collected by research that is funded by taxpayers should be collected, cryptographically signed and verified, and archived where any taxpayer can see it.…

Usually around IP incentives as part of the grant, in order to further entice applicants. This is taken further by pressure from educational institutions and their researchers to file for patents against said research.

I'm not saying it's right, and agree with you. But often there's a lot of entropy with the status quo, and to the seated incumbents in a space.

Re: Mirroring US climate data

#57

Post by tracker1 »

gns24 wrote: post #47
ted_dunning wrote: post #5

Earlier quoted context omitted.

Actually it really does matter. The US collects far more of this data than anybody else and archives far more of it. Losing that collection or archive would be a disaster for science and a loss for all of humanity. Destroying that data would be reminiscent of ISIS destroying relics in Palmyra except that destroying that data jeopardizes our future.

Personally I don't see that it's that important. We know already that CO2 and other gases cause climate change and we understand the mechanism. I think that most of us would probably also agree that the earth is a sufficiently complex system that predicting accurately what effects we will see when is impossible. We expect to see more droughts, bigger storms and rising sea levels - but we can't really predict very acc…

But "we need to act now" can lead to a lot of ineffective policies and practices that won't lead to the most impacting outcomes long term. I think reciprocal clean working environments as part of trade negotiations could go farther than a lot of the attempts so far. I also feel that no participating country should be exempt, beyond a reasonable grace period (2-3 years).

It's also worth considering the impact of some movements... as an example, there's significant impact in creating/shipping and replacing batteries on a large scale. It may not make as much sense compared to more efficient engines, and even additional looks at other engine techniques.

I find pollution relatively abhorrent on its' own though... recently driving through west Louisianan/east Texas and my initial thoughts as to the painful to breath air there was how can this be allowed to persist. I have mixed feelings all around, I just don't want to spend a lot of taxpayer money on ill-advised and lesser impacting processes. We do too much of that already.

Re: Mirroring US climate data

#58

Post by laurent123456 »

trothamel wrote: post #2

Can I be the first to claim that the people running the mirror are changing the data for their own nefarious purposes? Not that I believe this or anything, but come on - you know that accusation is going to go around. It might make sense to use a remote server to timestamp files as they're uploaded, or something like that.

I think that's why they suggest to use Webrecorder, to "create verifiable web archive files".

Re: Mirroring US climate data

#59

Post by jl6 »

rmc wrote: post #54
jl6 wrote: post #41

Earlier quoted context omitted.

And not just archive it, but sign it, time stamp it, verify it, etc.. I see little to prevent someone using ML to auto-generate archives worth of fake history. Imagine a purported DB dump of Reddit with billions of subtlely fraudulent comments scattered throughout.

Is there any good way to timestamp things like this?

Hash the content and store the hash in the Bitcoin blockchain.

Re: Mirroring US climate data

#60

Post by tukelully »

esaym wrote: post #25
trothamel wrote: post #2

Can I be the first to claim that the people running the mirror are changing the data for their own nefarious purposes? Not that I believe this or anything, but come on - you know that accusation is going to go around. It might make sense to use a remote server to timestamp files as they're uploaded, or something like that.

I would say the same thing. Do you think you will really convince a climate denier with this data that you copied off of the Internet that, was copied from a gov server, on to another server and then torrented and copied to another server and then downloaded onto you PC... doubt it.

How successful is the effort to convince clinate deniers as of right now? I see this as an archival effort. If the source is destroyed, this data, among everything else being archived, will still be around and publicly accessible by anyone who wants to use for any purpose. That is inherently important. That said, once the first part is well taken care of I'm a big advocate for building more refined systems for storing and verifying the information.
Post reply on HN