Great! Now, when I have something I want to remember for a whole two seconds, I know where to turn.

store information in them for nearly two seconds, an increase of nearly six orders of magnitude over the life span of earlier systems.

But

"The end result is that we're able to push the coherence time from a millisecond to nearly two seconds."

...or *three* orders of magnitude.
(Unless the first reference was to very early qubit systems with coherence times measureable in microseconds?)

Great! Now, when I have something I want to remember for a whole two seconds, I know where to turn.

You're aware that DRAMs (the stuff in your computer) need to be refreshed every 30-60 milliseconds or it will lose its values?

The article provides a good explanation, but why the need to report on the story for (at least) the third time already?

"Using massive amounts of laser light..." and "...is bombarded with a specific set of radio frequency pulses...". Would that massive amount and that bombarding be expensive? Prohibitively expensive to build a useful quantum computer?

Quantum phenomenon is mysterious for me. I think i understand the article more than 50%. Enjoy reading the article. "We can encode data in it, and we can store it for a relatively long time" and "we can encode a bit of information into that spin, and use this system to read it out", I guess it could maintain the data longer using this low-tech method : read quantum data from first qubit within the 2 seconds then encode it in second qubit followed by "copying" the 2nd qubit to the 3rd qubit. After several copy and paste in every ~2 seconds it might form a loop. If the technique is workable for quantum data, would the loop oscillate is another question. If it is possible to fork and store a quantum data in two "loops" I wonder could it compare whether 2 qubits are identical.
Also, "that nitrogen-vacancy (NV) centers, atomic-scale impurities in lab-grown diamonds, behave in the same way as single atoms", is this useful for quantum entanglement ?