Hacker News Reader

Comment by DrStalker | Hacker News Reader

How is the data "worth billions"? That may have been the original cost, but the original cost isn't the same as the value today.

View on HN

"Cost billions" would have be correct. Their value is "priceless" as the are clearly very valuable but determining that value is impossible. What is a ride on a time machine worth?

My cynical anti-govt title: "Priceless climate data that cost taxpayers billions found unopened at government climate center" Ha!

I know you're being cynical, but these weren't the originals--they were photographic images of the originals displayed on a TV monitor (!). So though what they received were unopened boxes, they were unopened boxes of poor copies. In addition, the original data wasn't used for climate research, but rather weather monitoring and forecasting. One of the innovative things about rescuing this data was to put this data to a different use than its original purpose.

I say this just to clarify that it is not the case that we launched these satellites and then never bothered to use the data (i.e., in case readers had an image of the Ark of the Covenant being wheeled into a warehouse).

>in case readers had an image of the Ark of the Covenant being wheeled into a warehouse

Damn, you got me.

Yeah. I hate the title. Apparently the editors felt the story's importance wasn't self-evident enough, and what readers would be capable of understanding was a pricetag. In truth, the data is priceless in every sense of that word. Unearth is also strange word choice, considering it was unearthed in a warehouse.

For HN, a briefer version of the subtitle would make a better title IMO.

Linkbait title.

From the end:

"The original cost to American taxpayers to gather the images was in the billions (in today’s dollars). “For a few hundred thousand dollars I could get it all back,” he says."

I do think, though, that as lake99 pointed out there could be TREMENDOUS value in this. Extending our image record a full 17 years is has enormous scientific value. It gave examples of imagery of the Aral Sea, which allows for significant study there, and 17 years more of full polar ice records. All of that has pretty big implications for a number of fields.

It's not really extending it by 17 years though since they don't have data for the full 17 years. They have snapshots from 17 years prior to the start of actual data collection.

I haven't actually poured through the data set linked in the article, but skimming it gives the impression that it includes a variety of images from 1964-70, which seems to me like a lot more than just a snapshot. It even mentions in the article that they were able to see multiple years of polar ice.

That would still be a snapshot. There is no monthly data, or even yearly on the same date by the sounds of things, nor does it backfill up to 1979(when we started actually collecting data)

When I read the title, I did not interpret "worth billions" as referring to the cost of producing them. I took it as its potential value to current science. In fact, the blurb below the title says, "pushing back the modern satellite record of sea ice extent by 17 years". This, indeed, may be worth billions to us now. Imagine having to deduce that data by other means available to us now.

In order to price something, you have to find a buyer who is willing to put up the money. Only the US government could pay billions for it, and, in fact, they already did -- back in the 60s. The price it could fetch now is utterly irrelevant -- it's one of a kind and has no replacement value, because it's irreplaceable. Luckily the government didn't throw it away.

The way it is "worth billions" would be computing the cost to reacquire it which is impossible. It is a stretch of course, and it depends on a strained definition of value.

That said, any reliable data from earlier is always good for filling out our understanding of how the planet has changed over time.

The additional data point really makes obvious how steady the Arctic sea ice was up to somewhere between 1990 and 2000: http://www.tjradcliffe.com/?p=1634 (just a quick replotting and refitting of the data, as the graph in the paper linked from the article doesn't really give a clear sense of how far in the past this new data point is).

Reply on news.ycombinator.com