. . . within the next million years -- stick to radio.

Rose's paper is here, which I found on his website.

Discussion in the press (links at Rose's website, including an article in the NYT Science section by Dennis Overbye, one of my favorite popularizers of astronomy) makes repeated analogy to Clarke's monoliths in 2001: A Space Odyssey, yet fans of the film and book will recall that, as a communication device, the one on the Moon relied on an extradimensional energy burst to alert its constructors to its discovery (which also "bled" into the EM spectrum in our spacetime).

Thse quotes are sequential (i.e., not cherry-picked), from the final paragraph of the paper's conclusion:

Finally, for such long range channels, we have skirted the issue of what sort of messages one might want to send, how they might be detected or where they might be sent [28], [29]. The large delays associated with interstellar travel and the seeming fragility of species to cosmic insults suggests that an intelligent sender might construct messages “for posterity” as opposed to for initiating a chat.

That's an awful lot of skirting. I haven't read the entire paper yet, but I have the impression that Rose's energy efficiency calculations are a little too simplistic. Over his example of a milion years' travel time to cover a 1000 l.y. distance, the proper motions of both originating and target stars will have a large influence. This creates problems of propulsion and guidance (including mid-course corrections), and the service life of the control systems.

On the other hand -- the "Voyager Record" is a perfect example -- is the concept of sending information packages Out There, and maybe someone will find them ... rather like finding, not a needle, but a single atom of platinum in a haystack.

What seems most critical to me is what (elsewhere in the paper) Rose terms "hardening". Unfortunately for me, his reference [19], discussing theoretical limits of the information density on inscribed matter is unavailable (until I go to the library, since I don't have a subscription to Science). Nonetheless, it's fair to consider that the smaller each bit of information, the less energy required to alter it. Rose mentions RNA as a particularlt dense method of encoding -- now consider how that might be subject to 'mutation' over a million years, even to such low-probability events as neutrino interaction, which cannot be shielded against, let alone cosmic rays and the occasional gamma ray burst.

One might also think of “colonization” as a goal as well [30]. In both regards, one ostensible virtue of inscribed mass channels is that once the message arrives, it is persistent as compared to electromagnetic radiation which is transient and thus must be sent repeatedly in order to assure reception.

"One the message arrives" - implying either [a] more propulsion and autonomous controls to place it in orbit around the target star (or a planet), or [b] propulsion, controls, decision algorithms to select a planet upon which to land, and a means of making that landing mechanically survivable. The "persistence" of such encoded information now possibly includes exposure to atmospheric phenomena after a million-year journey through interstellar space -- assuming the package isn't damaged in the process of delivery.

Then there's the decoding process. The Voyager Record has pictorial instructions engraved on its casing. How might one render a similarly simple 'manual' for decoding RNA?

Of course, constructing mass packets to be hearty, easily detected and/or self replicative seems well outside our current engineering ken. Nonetheless, the notion of mass packet delivery, undertaken initially to examine assumptions about energy tradeoffs in terrestrial communications, does raise interesting questions about terrestrial biological history and perhaps SETI/xenobiological studies as well.

That's a little too far-fetched, even for me.

But, I think his class had a lot of fun with this.