Radius of a Silicon atom is about 0.11 nanometers; Intel produces CPUs that have 22 nanometers technology. Not exactly "atom-by-atom", but pretty close (only two orders of magnitude).
Then there's the smaller stuff that adds up. In the "golden era", you had to tune up your car every 3,000 miles or so. My 2004 and 2005 cars needed tune-ups after 100,000 miles. In the "golden era", we took an un-air-conditioned car to visit my grandparents in Arizona. We did it in the summer, because that's when my brother and I were off of school. Yeah, that was fun... not. Now, you would have a very hard time buying a car without air conditioning.
Its easy to list whole categories of innovation, and say "Anything else?" Aren't those enough?
What qualitatively new capability Bitcoin gives me? What's Twitter? Online movies? Well, yes! It does not improve our lives radically but still.
> Privatized space industry.
You get the point here. Nice!
> Gene sequencing for pennies.
We need faster/wider adoption of it. Article mentions it. Right now I cannot benefit from it in everyday hospital.
> Working from home.
Was possible before. Only quantitative improvement. Although awesome.
> 3D printing
Is a good counter-example. It's been around for years. And still I didn't hear about someone who uses it on a daily basis to improve their lives. Amazing applications in medicine is the only exclusion so far. Expensive industrial 3D-printers for prototyping etc. are well expensive and not wide-spread.
I work in a lab with people who are working on developing diagnostics for transplant rejection based on sequencing the RNA from a small sample of your blood and using a trained machine learning classifier to diagnose rejection (I'm tangentially involved in this project). It's still a ways away from clinical application, but if successful, it could replace the current gold standard for diagnosis, which is an invasive biopsy of the graft that is then manually inspected under a microscope by a trained histologist.
That's a pretty big step up, from a biopsy and a histologist to a few drops of blood and a computer classifier to diagnose transplant rejection with the same accuracy. It would mean that you could monitor for signs of rejection much more often because drawing blood is much less invasive. This, in turn, would allow you to detect rejection events much earlier and take action to control the inflammation and prevent loss of the graft before it's too late. There's huge potential for quality-of-life improvements here. And I can assure you many other labs are applying the same methods to all sorts of diagnostic problems, not just transplant rejection. I could imagine a future where we have hundreds of trained classifiers for all sorts of diseases based on blood gene expression, and as a routine part of a doctor's visit, your blood will be drawn, sequenced in a matter of minutes, and run through all of these classifiers to identify any diseases you might possibly have, which can then be tested for by more accurate conventional tests now that the doctor knows what to test for.
Here's the paper I was involved in: http://www.ncbi.nlm.nih.gov/pubmed/24725967
Here's a similar earlier paper that will also give you the idea of what we're up to and that I think is public access: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2041877/