[skreutzer]

The answer to your question depends on what you're trying to achieve. Sure, the easiest way to deal with it is to upload an .epub to your website and give people the link to it, so they can download it and sideload it by USB cable or whatever onto their reading device. Depending on the reading device, probably one could even browse your website download link with the device and open it directly, if the e-readers software is capable of doing so. In terms of the “uncontrollable” nature of this distribution method, it might occur that people share the download link within their group of friends or make it even publicly known, so that everybody can download the file from your server. But please keep in mind that in digital technology, which is built to copy and transmit data, there's in general no way to prevent that, because if you e-mail your .epub to somebody, he might as well just upload the file onto a server and making the link to it publicly known. Additionally, there's no way to prevent “uncontrolled” sideloading over multiple devices. If you want, you can collect download statistics with the webserver logs (a webserver is able to log access to files, which can be turned on/off and can be displayed in a human readable form with some tools), but as said, those download statistics are limited to distribution done by your webserver, while you can't monitor distribution done by others after somebody somewhere in the chain downloaded the file initially from your webserver.

Still, you might want to create some kind of artificial scarcity on your side by preventing unlimited downloads from your server (which doesn't make much sense technically, but most people are more interested in convenience than doing stuff with your files). One way to do this is to let the webserver create individual download links for every download request you receive, which stops to work after the first download, so at least nobody can download with this link again after it was used once. You may use this to monitor who actually downloaded the file. This doesn't help at all against “uncontrolled” distribution after this initial download was made. Usually there's software for doing so off the shelf, but it is too easy for a web software developer of your choice to implement such a tool for you, so in general that's a method that can be used. In order to track “uncontrolled” distribution to some extend, you might place links in your .epub which go to a website, while those links include individualized identifiers, which you can match to the initial download provided by you. If a reader clicks on it, not knowing that the link is containing an identifier of the initial download, the linked webserver may track to which extend the file is distributed, but please note that such links can easily be removed, exclude all offline readers, and links can be clicked multiple times by the same person.

I'm in favor of providing e-books to a cheaper price or even without cost to people who can't afford it, I'm against artificial limitation of downloading, distribution and copying, I'm in favor of authors who don't try to sell files after they've made them artificially scarce (which is a pretty bad concept) and sell their ability to create those files (= texts) instead, maybe by offering incentives to willingly support the author or maybe by providing a combined online/offline experience. The only working way to protect your text from getting copied and distributed without your prior consent is to never write it down at all.