With the flare-producing Iridium communication satellites [link] just about all deorbited now, you can no longer just check a viewing schedule on your phone (Heavens- Above! [link]) and head out on just about any given night to see with your naked eyes a bright streak of sunlight reflected off a satellite flying above you. Iridium's new satellites don't have the big flat antenna panels that reflected the light so well.
Therefore, I propose launching a few satellites to deliberately produce flares, on demand. Members of the public would be able to submit requests via a website or phone app to receive a flare. The satellites would also produce flares on a predetermined schedule, to allow spontaneous viewing like with the Iridium ones.
The satellites would have aimable mirrors, as opposed to the fixed antennas on the Iridium satellites. This would mean that fewer satellites could produce more flares with ground tracks where viewers want to view them (mostly cities), while avoiding flares where they're diswanted (i.e. astronomical observatories, normally sited away from cities). The Iridium satellites couldn't control their flares' ground tracks at all—they just occurred at effectively random locations on Earth, most of which passed over no viewers.
The satellite design could be pretty simple: Just a great big flat mirror (as big as will fit in the launch vehicle fairing), and, at one end, a small box with the computer, thrusters, reaction wheels, solar cells, and radio. The flare brightness could be modulated, if desired, with a liquid crystal light valve over the mirror (expensive, and reduces max brightness by ~50%, but allows fast modulation) or by tilting the mirror away from the desired ground track and back onto it (cheaper, but slower, and produces the inverse signal on an adjacent ground track).
The ability to go outside at night, make a request by tapping a few buttons on your phone, and, a few minutes later, see a point of light zooming across the sky exactly where and when the app said it would, optionally flashing a Morse coded message*, would probably help remind people that we really do have thousands of things up there doing work for us.
*paid extra for regular users, free for humanitarian communication or when the message is "EARTH IS ROUND"
(Urban astronomers, just be glad I didn't put this in the category Public: Street Lighting!)
N/A [2019-04-08]-- notexactly, Apr 08 2019 Wikipedia: Satellite flare § Iridium flares https://en.wikipedi...lare#Iridium_flaresMentioned in idea body [notexactly, Sep 21 2019] Heavens-Above: Iridium flare prediction page https://www.heavens.../IridiumFlares.aspxMentioned in idea body [notexactly, Sep 21 2019] Sorry—clicked the wrong button again and deleted [MaxwellBuchanan] anno saying the ISS is good for flare viewing. True. But you can't request a flare at a specific time (from a list of possibilities) or with a specific Morse code message from it. (But you can talk to real live humans aboard it by ham radio, which is pretty neat!)-- notexactly, Apr 09 2019 I really need to do something (uBlock Origin cosmetic filter, userstyle, userscript, …) to hide those [delete] buttons. Or I could just post an idea about the site having something to solve that problem.-- notexactly, Apr 09 2019 I thought flares went out of date in the 1970's. Presumably the factories could be de- mothballed?-- not_morrison_rm, Apr 09 2019 Heaven forfend ...
The Russians proposed lofting a mirror to provide artificial moonlight in Siberia.
Maybe something using Cubesats could be done. We like Cubes ...-- 8th of 7, Apr 09 2019 I was envisioning the body of the satellite to which the mirror is attached as being in the shape of a cube.-- notexactly, Apr 09 2019 <Turns up Cube Intensifier Ray another notch/>-- 8th of 7, Apr 09 2019 Today I was thinking about this again, specifically about the possibility of making the first one a CubeSat (because those are a lot less expensive than regular satellites to build and launch) to prove the technology and demand before scaling up.
Also, upon rereading the idea, I had the thought that the mirror could be made in two parts that slide over each other. Then the flare brightness could be reduced from 100% to 50% as the modulation, without the expense and optical inefficiency of a liquid crystal light valve, and without the inverse signal on an adjacent ground track that you'd get by just tilting the mirror, though it would still be slow—unless you did it in two fully-overlapping arrays of parallel stripes of mirror, so that the moving part wouldn't have to move as far to go from no mirror overlap to full mirror overlap. Only modulating down to 50% is probably okay; it would enable viewers to visually track the satellite between the flashes, like a bike light with a 'pulse' mode.
Another thought I just had is that, IIRC, the Iridium flares had pretty narrow beams*, meaning that you have to be in just about exactly the right place on Earth (across-track beamwidth) and you only see the flare for a few seconds even though the satellite takes a minute or two to cross most of the sky (along-track beamwidth). That's fine if you're sending a "will you marry me?" message, but we'd probably want a wider beam for greeting whole cities (which would be a regularly-done thing, not needing to be ordered). To achieve that, the mirror could be slightly flexible, with a cord to pull inward on the far corners of it to make it curve a bit, widening the beam in a controllable manner. But that can be a feature of the later satellites, not necessarily included in the CubeSat pathfinder.
Also, commercial advertising should probably not be allowed on these satellites, even though it could bring in a lot of money (from advertisers whose target audiences know Morse code, anyway).
*Later, upon looking it up, I see Wikipedia says 10 km in diameter.-- notexactly, Sep 16 2019 I had another thought. A seemingly good CubeSat configuration for this (if a smaller mirror is sufficient) would be 3U, with a fixed mirror on one long side and hinged mirrors on the two adjacent long sides, deployed after separation from the launcher to fold out and make a .3 m × .3 m mirror. A very slightly rippled surface on the mirror—as would probably naturally result from manufacturing—would be enough to widen the beam so that it could be practically seen. (If the mirror was perfectly flat, and the satellite orbited at around 200 km, the spot size would be about 1.9 km due to the angular size of the Sun as viewed through the mirror.)
Wikipedia describes the Iridium antennas as "door-sized" and capable of generating flares up to -9.5 mag, even though they don't look anywhere near as reflective as a proper mirror, so this specific CubeSat configuration might be visible without binoculars and without having to view it from a very dark place.
If the two folding mirror segments were connected to voice coil actuators (such as are used for hard drive head arms), they could be used for pretty fast modulation from 100% down to 33%, while only producing an inverse signal of 0–33% amplitude on an adjacent ground track, which would be less visible than a 100% inverse signal on the adjacent ground track.
To further reduce interference on adjacent ground tracks, it also occurs to me that tilting the mirrors away and back according to a triangular profile (rather than a square profile, where they tilt to a second position, hold that position, and tilt back) would spread the interference over a wider area, reducing the maximum amount received at any one point on the ground.
To enable modulation down to near 0%, all that's needed is to tilt the two hinged mirrors by different amounts. That will cause an imbalanced torque on the satellite body, resulting in the fixed mirror tilting to the side somewhat as well. However, its amplitude will be smaller, which, combined with the diffuse shape of the beam, will mean some light still gets reflected toward the target, but it'll be pretty close to 0%.
(Aiming of flares could be achieved with reaction wheels, as with any active flare satellite configuration, but care would be needed to keep the roll-axis RW from fighting the above- described reaction roll that enables modulation down to near 0%.)-- notexactly, Sep 16 2019 [Links] added because I showed this to a friend and he didn't know what was referred to, so didn't have much context to understand the purpose of the idea.-- notexactly, Sep 21 2019 random, halfbakery