well, it's certainly been a while, hasn't it? a lot has changed in the past four years - but such matter is the subject of a different post.
we wrote the vast majority of the text that follows in almost a single sitting after learning of the news that a megacorporation bought out bandcamp, which until this point had been the definitive platform to buy and sell music through. hopefully it represents a return to longer-form posts; we'll see.
a bandcamp alternative should be distributed, to balance costs and mitigate attacks [both from without, as in traditional cyberattacks - but also from within, as we saw occur with the bandcamp heist].
but does it need to be federated? cross-site recommendations can be implemented via nothing more than a simple link, and reviews should not require an account on any site - a message sent upon purchase would suffice. the real issue, then, becomes one of search - while we ourselves might only use word of mouth [an expression that, as an aside, has become increasingly anachronistic as the technology of communication has advanced - when was the last time any of us primarily used our /mouths/ to talk to one another?] as a form of discovery, several people will prefer to at least have the option to search the known body of work by means of various criteria.
the only other method by which this could be achieved without federation would require some form of centralized database; leaving aside for the moment the fact that this arrangement defeats the point of decentralization somewhat, such a database would be quite hard if not impossible to maintain in an accurate and timely manner. the depth of variety in human experience dictates this - new works will constantly be added, and for various reasons, old works will be updated or even [at a significantly lesser rate, one would hope] removed. certainly, operating such a repository would be far outside the reach of all but the most privileged individuals.
it seems, then, that a fully-fledged bandcamp replacement must eventually implement some form of federation if searching is a desired feature. given this, some light thinking on the matter immediately suggests areas where federation might elevate other features: namely, reviews. while - as stated earlier - a simple link will do perfectly well in this regard, federation would allow for an enhanced view; genre, album art, and track count might all be automatically displayed for example, or - if the user has so customized their preferences accordingly - any other number of metadata facets. a dj searching for the latest underground witch house bangers [or some such] might be chiefly interested in seeing the bpm and length of tracks, for example, and so would add this to their client's settings.
but as it stands at present, this is all mere speculation. while such a federated music platform would surely be of wonderful good to the audio world - and to our example dj above - it would not be essential. rather, the so-called "minimum viable product" for a bandcamp replacement would be comprised of the following features: first, a way to display a [lower-quality] preview of the music on offer to potential patrons; and second, a way for these patrons to purchase said music. the first of these already exists in at least one form: "blamscamp", a creation by the prolific blackle mori, has the ability to preview music very well, as we understand it [note however that we have not yet examined the code behind this tool and thus cannot confirm its efficacy]. however, this is not enough by itself - at present, blamscamp has no way to transfer money or music downloads, instead relying on the likes of noted games marketplace itch.io to perform this step.
as a stopgap it will suffice; it seems very unlikely that itch will follow bandcamp down the dark path that the latter's new overlords will undoubtedly take them down for some time. but mark our words: /it is just a matter of time/. eventually all corporations, no matter how benign they may seem now, will sooner or later be swallowed whole by the never-satisfied maw of capitalism's world-destroying influence. it is this scenario of itch's demise then that becomes the most pressing to avert - or more succinctly, the demise of the security of payment and hosting that itch provides. itch itself need not play a part in this itself. a more sustainable situation would be that of multiple hosting platforms, each interchangeable so that an artist might transfer their hosted works to any other -
but this inherently raises the issue of trust. inevitably, not every instance will be run by people of a scrupulous nature [as bandcamp has so regrettably demonstrated], and this unfortunate truth becomes several orders of magnitude worse once money becomes involved. while in theory a malevolent host is sandboxed by the browser, or other client, any such host that deals with money at once gains the ability to defraud any who would be misled into entering their credit card details. with a client this danger is moderately lessened, as an api would prevent misleading inputs, and we understand some system to verify the identity of a payment processor already exists [the "Extended Validation Certificate", we believe; this may be observed in your browser by navigating to your bank and pressing the lock icon in the address bar]. however, this does not address the problem of verifying that the account receiving money is the same as the author of a work. still, the same can be said of any simple plagiarism, so perhaps this line of thought is overcomplicating the situation.
we're not entirely sure where we were trying to go with this