Hacker News Reader

Comment by blahedo | Hacker News Reader

"Even so, the first few computer scientists and statisticians the foundation got in touch with said that it couldn’t be done."

Really? Wtf kind of incompetents were arguing that it couldn't be done?

I'm not arguing that it was or should have been easy, of course. It sounds like a fascinating challenge. And I'm also not claiming that any computer scientist, even a good one, could have solved it. But any vaguely competent computer scientist should have been able to say, "that's a cool problem, and you need to talk to a graph drawing expert."

Of course, this could have been a case of the source saying "they said it was hard" and the writer mangling that into "they said it couldn't be done". But still.

View on HN

Of course, this could have been a case of the source saying "they said it was hard" and the writer mangling that into "they said it couldn't be done". But still.

I think it is more likely they said "It isn't a well-defined problem."

It's possible that the problem was originally stated in such a way that there was no solution. Perhaps the original statement of the problem was that each name must be adjacent to every name in its cluster, and readable. But the cluster of 658 people at Cantor Fitzgerald probably makes that impossible - how do you have 658 names all readable and adjacent to each other (in two dimensions)?

Perhaps they really answered "as the problem is stated, it can't be done, without compromises."

Then again, I tend to give computer scientists and statisticians the benefit of the doubt (and not so much the writer).

If I had been asked, my reply would have been "It can't be done. If you want help from me or someone like me, you'll need to make a loose ranking of the qualities of the ideal solution, and we can try some approximations".

I might even have left the latter sentence out, since it's clearly implicit. Maybe whoever they found left it out or maybe they didn't even consider it, but the artist could have made the addition too. I suspect the first few people just weren't interested.

"fascinating challenge" != "publishable"

It's possible that any academic asked might have concluded that there wasn't a general publishable result to be had from this work and therefore that it wasn't worth doing. Just because something is interesting doesn't necessarily mean it is of any interest from a research perspective.

[NB I left academic research in CS for pretty much these reasons - I like interesting stuff].

There are plenty of people that know computer science that will do contract work.

From the sound of it, the original request was probably stated as something massive and NP hard, he asked for the 'best' placement and was told it would take more time than seconds since the big bang to locate the best placement.

Then somebody actually talked to him and took the time to realize he didn't need the optimal solution.

Actually, solving most practical instances of NP hard problems is doable and quite tractable.

Obviously it depends a whole lot on the scale. They gave some idea of number of constraints required to be satisfied, but it's not clear to me what the actual constraints were. I think it's not outrageous that on scan it might have sounded intractable.

Actually, solving most practical instances of NP hard problems is doable and quite tractable.

Isn't that a tautology? Because if it wasn't doable then it wouldn't be practical.

Practical here means: Occurs in practice. And that they are optimally solvable in practice is not a tautology. It could be that you would need human insight (like you still need for most proving interesting mathematical theorems), or that people would only solve approximations. (People do solve approximations, too.)

As someone who has had research written about in popular media, I would bet an enormous amount of money at very unfavorable odds that the words "it couldn't be done" were not uttered by the CS researchers but instead are the voice of the New Yorker author.

Reply on news.ycombinator.com