Unfortunately that's not that simple. Your observation may well be the case, but that has nothing to do with whether it's going to interfere with someone else. Here's why:
(tl;dr: The output of these "transmitters" contains spurs you're not equipped to detect but which can indeed be problematic for others. Transmitters are regulated for sound physical reasons, some which are enumerated here.)
The range at which your FM receiver can pick up an FM signal may indeed be ten feet, but that's a product of the transmitter power, the transmit antenna gain, the path loss, the receive antenna gain, and the receiver sensitivity. To infer anything about one of those parameters (for this discussion, the transmit power), you have to know the other four, and you don't. (And neither do I. Even the measurements I'm equipped to make require a lot of assumptions and have significant uncertainty.)
Secondly, FM (radio folks would call this wide-FM) is particularly power-hungry because it occupies a relatively wide swath of bandwidth. Typically limited to 75kHz and in practice somewhere around 50 for analog stereo or just shy of 60 with RDS (which pifmrds can do). For example, a 104.3MHz station is actually transmitting anywhere between 104.225 and 104.375 at any given instant. This means that the receiver has to demodulate the carrier which may be varying all over that range, and it requires a fairly strong signal to do that. If the FM receiver loses lock, it doesn't mean the signal doesn't exist, merely that it's below the receiver's sensitivity.
Other spectrum users, like radar for example, may be dramatically more sensitive, and pick up a signal of the same strength many miles away. A mode that occupies less spectrum generally requires less power. They may also have better antennas (radar dishes tend to be tightly focused, that's the whole point), better receivers (local self-noise can be minimized through careful-but-costly engineering), et cetera.
They may also be above you, so while you see the signal disappearing as soon as you get on the other side of some hedges, there's no foliage between you and an airplane.
So while _for you_, the effective range may be ten feet, but for someone else, the signal may be a problem dozens of miles away.
Furthermore, and the whole point here, is that you're not just interfering with other users in the 88-108MHz FM broadcast band. (If you were, frankly, I wouldn't care one bit; there's little actual harm to be done.) It's that as a digital pin being bashed up and down at those frequencies, the resulting signal is profoundly not sinusoidal. It has much faster rising and falling edges, which mean the signal has plenty of energy in harmonics, not just the fundamental frequency of the carrier. My spectrum analyzer only goes to 2.7GHz and I was able to see (admittedly diminishing) energy up beyond 1GHz, which is both impressive in that the BCM chip's GPIOs can slew that fast, and terrible in that you're interfering with _everyone_ when you do it.
So, say you've told pifm to transmit at 88MHz, right at the bottom of the broadcast band where off-the-shelf receivers can pick it up. Your little Pi and random-wire antenna are also producing quite a bit of energy at 176Mhz (maritime VHF ship-to-shore and stuff), 264MHz (military air-band, adjacent to a satellite-earth-station band), 352MHz (more military), 440MHz (amateur UHF), etc. All the way up.
This may be counterintuitive, but that's why radio is governed by regulations rather than YOLO, and why even unlicensed toy transmitters still have to comply with harmonic-content rules.
Filter your Pi to knock out all those harmonics, and the odds of the Coast Guard or Air Force knocking on your door will plummet.
(Also, I've measured the power coming out of one of these things, and if you can only pick it up ten feet away, either your antenna is total crap, your receiver is deaf as a post, or both. Even after applying proper filtering which eats some power, I was able to receive mine on a pocket radio out to about 200 feet, which surprised the heck out of me and I added lots more attenuation until my most sensitive receiver could no longer see it at the property line. It's a fantastic toy for exercising RDS/RBDS receiver and parser code, but it requires a little care and knowledge to use it responsibly.)
Anyone can learn this stuff, and everyone should, especially if they're going to put it on the air.