On: Build your own Startup Death Clock

[celoyd]

I’m saying he didn’t blur it enough. Someone could take a standard image deconvolution package, give it a good estimate of the point spread function, and probably recover at least the first and last digits of each line.

It’s probably not very important in this case, considering his general openness, but I thought I’d bring it up just to remind people.

[patrickfoley → celoyd]

I was actually wondering about that - thank you for the specificity of which approach to take instead.