The first thing I found that was closest to a real experiment with "large" objects going through double slit setup is an experiment done with buckyballs discussed in the paper Quantum interference experiments with large molecules by Olaf Nairz, Markus Arndt, and Anton Zeilinger (a full copy of the paper is available for free online, and is rather lightweight compared to most papers on quantum mechanics).
While creating a coherent beam of cats in the same sense of some of the atom lasers being done now sounds by far harder than finding that many kittens, I don't think it is necessary to get interference patterns as discussed in the buckyball paper. That experiment was done with 900 K (1100+ Fahrenheit) buckyballs, and the internal state of the molecule could be ignored as it was only doing single particle interference (i.e. each was effectively going through the slits one at a time, interfering with itself, and acting as a single, large nondescript particle). In that sense, it could be possible to have interference from kittens that are not identical on the atomic scale.
What really matters is that they all have as close to the same mass and velocity, so they all have the same wavelength. The referenced paper talks of a longitudinal coherence, which adapted for allowing varying mass would give Lc~y/(dm/m+dv/v) (using y for lambda/wavelength), and as long as (dm/m+dv/v) is smaller than about 0.5, you could start to see the first set of fringes. That would actually be pretty easy to achieve, even with your everyday random kittens (assuming they were sealed in something, as in a vacuum kittens, like most organic matter, would outgas a lot, changing their mass). And it wouldn't even need the velocity chopped the paper discusses, as that would be much messier adapted to kittens, although it would need some form of collimation (i.e. kitten optics) which might be difficult or messy.
I'm not entirely sure how they derived the number of kittens needed, but I have a hunch it might be possible with much few kittens, as in on the order of thousands to million, not 10^36 kittens, which would have 50 billion times the mass of the Earth. I would need to think about that more.
The main concern would be the distance needed (although you could probably shave a zero or two off the end of it by careful experiment design), and especially the time it would take for the kittens to transit that distance. That would be on the order of 10^23 years. The good news is this is below the lower bound estimate to the proton decay, and well below the timescale that quantum tunnelling would cause all objects to macroscopically rearrange, so there would still potentially be a kitten at the end. But for comparison it is longer than the estimated stability times of star systems and galaxies.
In summary, I think the kitten double slit experiment might be a lot easier than then the blog suggests, but there is still a problem or two preventing it from being tested with in (almost farthest possible) foreseeable future.
(Disclaimer, while I've had quite a few courses on quantum mechanics, they spend a lot of emphasis on things other than thinking about experiments like this... and my field of research is different so I haven't really used quantum much since early grad school other than answering student questions and things like this... and I haven't had coffee yet today.)
The first thing I found that was closest to a real experiment with "large" objects going through double slit setup is an experiment done with buckyballs discussed in the paper Quantum interference experiments with large molecules by Olaf Nairz, Markus Arndt, and Anton Zeilinger (a full copy of the paper is available for free online, and is rather lightweight compared to most papers on quantum mechanics).
While creating a coherent beam of cats in the same sense of some of the atom lasers being done now sounds by far harder than finding that many kittens, I don't think it is necessary to get interference patterns as discussed in the buckyball paper. That experiment was done with 900 K (1100+ Fahrenheit) buckyballs, and the internal state of the molecule could be ignored as it was only doing single particle interference (i.e. each was effectively going through the slits one at a time, interfering with itself, and acting as a single, large nondescript particle). In that sense, it could be possible to have interference from kittens that are not identical on the atomic scale.
What really matters is that they all have as close to the same mass and velocity, so they all have the same wavelength. The referenced paper talks of a longitudinal coherence, which adapted for allowing varying mass would give Lc~y/(dm/m+dv/v) (using y for lambda/wavelength), and as long as (dm/m+dv/v) is smaller than about 0.5, you could start to see the first set of fringes. That would actually be pretty easy to achieve, even with your everyday random kittens (assuming they were sealed in something, as in a vacuum kittens, like most organic matter, would outgas a lot, changing their mass). And it wouldn't even need the velocity chopped the paper discusses, as that would be much messier adapted to kittens, although it would need some form of collimation (i.e. kitten optics) which might be difficult or messy.
I'm not entirely sure how they derived the number of kittens needed, but I have a hunch it might be possible with much few kittens, as in on the order of thousands to million, not 10^36 kittens, which would have 50 billion times the mass of the Earth. I would need to think about that more.
The main concern would be the distance needed (although you could probably shave a zero or two off the end of it by careful experiment design), and especially the time it would take for the kittens to transit that distance. That would be on the order of 10^23 years. The good news is this is below the lower bound estimate to the proton decay, and well below the timescale that quantum tunnelling would cause all objects to macroscopically rearrange, so there would still potentially be a kitten at the end. But for comparison it is longer than the estimated stability times of star systems and galaxies.
In summary, I think the kitten double slit experiment might be a lot easier than then the blog suggests, but there is still a problem or two preventing it from being tested with in (almost farthest possible) foreseeable future.
(Disclaimer, while I've had quite a few courses on quantum mechanics, they spend a lot of emphasis on things other than thinking about experiments like this... and my field of research is different so I haven't really used quantum much since early grad school other than answering student questions and things like this... and I haven't had coffee yet today.)