A short story about why I love @waybackmachine.

I'll start with two pictures. Spot the difference.

The first is the current correction on Richard Epstein's Covid-19 piece. The second is a Wayback Machine capture of the correction on the piece as of 2.5 weeks ago.

The current link to Richard Epstein's Covid-19 piece (originally published on March 16th) states:

"My original erroneous estimate of 5,000 dead in the US is a number ten times smaller than I intended to state..." https://www.hoover.org/research/coronavirus-pandemic

I saw this earlier tonight and thought to myself "I could've sworn I heard his original estimate was 500. If he said 5,000 and meant 50,000 US deaths, I still think he'll end up being too low, but that's not dramatically out of line with current projections for the first wave."

So I dug a bit more.

I found numerous contemporaneous sources citing the 500 figure, so my next thought was "Ok, I'll check the Wayback Machine. Maybe he just updated the piece and made a typo in the correction." But there were no captures of the link above from before April 7!

Wayback Machine link:
https://web.archive.org/web/*/https://www.hoover.org/research/coronavirus-pandemic

At this point I was puzzled. Eventually I found a piece linking to Epstein's essay and noticed that while the link went to the current version of the document, the actual URL was different.

The old link ended "/coronavirus-isnt-pandemic" whereas the new link ends "/coronavirus-pandemic". The old page has been taken down and the link set to redirect to the new page.

Ok, maybe a little weird, but at least the new description is more accurate. But then I looked at the captures from the old link and saw the difference highlighted above.

While the main text was identical other than some minor copyediting...

The original correction had the different wording seen in the second screenshot above:

"That estimate is ten times greater than the 500 number I erroneously put in the initial draft of the essay."

So to summarize:

(1) There was an original correction added March 24 that correctly stated the 500-death estimate.

(2) Around the time he added a follow-up piece on April 6, Epstein changed the link to the original essay.

(3) At the new link, he left the main text of the essay essentially unchanged from the March 24 update, but moved the location of the Correction & Addendum and changed the Correction to state that his original erroneous estimate was 5,000 instead of 500.

I'll let people draw their own conclusions, but I am very grateful to the folks @waybackmachine for preserving these documents. (end)