Keep in mind that Amazon is also a "value-added middleman".
I'm an avid reader; honestly, given my nighttime reading patterns, I probably read too much. I've been to the deep ends of reading self-published work (much of it "acclaimed") and I'm not satisfied. Perhaps one day we'll arrive at a place where crowd-sourced reviews adequately filter the wheat from the chaff in an editor- and marketing-less world, but I'm not convinced we are there just yet.
Short version: I've read many (i.e., 10+) self-published works, and few of them have me looking forward to a world without publishers. I appreciate "elitist" filters that still allow pulpy sci-fi and fantasy through. I'm not ready to see them leave quite yet.
How do you find books to read?
My techniques are: Amazon suggestions, browsing at the library, blogs, forums, and following my favorite authors websites.
None of these require a publisher.
The problem with Crowd sourced reviews, is that quite often, other people might like quite crappy books. The job of a publisher isn't to just find a book that they like, but that the general reading population will love.
I appreciate and respect what the publisher does as part of their filtering - I just suspect they are extracting more value than they are worth.
Hopefully established authors are able to negotiate increasingly more lucrative deals as a reflection of both the reduced risk, as well as the increased value they bring to the equation (as well as the reduced value that the publisher brings. I can't tell you who Stephen King's publisher is...)
Otherwise we'd have to put up with Twilight and Fifty Shades of Grey, when we could be reading polished high-quality titles that twinkle with the elusive literary aura that only publishers can add.
[snerk]
Here is The Thing: publishing used to be an educated middle class business catering to educated middle class readers.
Self-publishing opened the market to less-educated writers producing work for less-educated readers. Amazon basically reinvented the old Victorian penny dreadful market in digital form.
Does this matter? Not really. The penny dreadfuls didn't kill literature. In fact, in a round-about way, they eventually launched science fiction and fantasy as genres.
Publishers gave up on real literature back in the 80s, when all the old small semi-amateur publishing houses were swallowed by corporate sharks. So don't look for not-crappy there.
There is some basic filtering to eliminate people who can't write at all. But Amazon reviewers are getting pickier, so it's not obvious the can't write at all crowd will survive for much longer.
Meanwhile, many not-quite-mainstream writers have pulled themselves out of publisher-enforced poverty by selling direct.
Is this a bad thing? No, it really isn't.
I don't care about the "general reading population"; I want books that I will enjoy. Recommendation engines provide a lot more value in this area than publishers (at least for fiction).
A publisher filters to find books that will sell, not to find great writing. In some instances great writing will sell quite well, in others that crappy romance novel can be a hit with a lot people. Publisher's release garbage all the time, it's not like they're some guardian angels of good writing. Maybe it would be awesome if we could just let the readers decide whether a book is good to them or not?
As a side note, the assumption here is that reviews are legit and not bought. To me the non-ideal situation of fake reviews is a separate issue vs the principle of whether legitimate customer reviews can serve as a good filter for readers. My experience has been that they certianly can.
I'm on the side of Amazon here, but let's not get carried away.
The books in your local library have largely come through a traditional publisher. Those blogs, forums, and author websites are going to be biased (not in the perjorative sense, but as in selection bias) toward titles that have come through traditional publishers as well.
The closest we can come to seeing a world without the influence of a publisher is Amazon's self-published titles, and they're really not good on average. I don't have a huge number of data points myself, but I haven't heard anyone rave about the quality of a randomly selected self-published title.
I think there's still a valuable role for the publishers to play. However, that role is quite a bit smaller than it once was, and the publishers are currently entrenched in a position of taking the same high cut of the profits that they've always taken, and that's going to have to change in order for them to do well in the longer term.
I usually go in this order: recommendations from friends, reviews, browsing bookstores or libraries, and books sent by publishers or authors (I write a blog about books and ideas at http://jseliger.wordpress.com).
One question is how that initial seeding group happens; right now it tends to through Publishers Weekly, Booklist, and related means. But that gets towards complicated questions of how information and ideas propagate through a society, and I think we're going to be sorting that out for a long time.
As more and more authors are simply buying dozens of 5* reviews for their books on sites like fiverr (or outsourcing it to Book Marketing companies who do the dirty work on their behalf), then I think we're moving further from that world, rather than closer to it.
But some do, correct? I've been reading quite a few (more than 10, probably more like 20-30) lately and I've ran across more than a few great finds. Granted, maybe our "Standards" are vastly different - I admit freely to being spat upon by literary purists for reading Science Fiction and Fantasy.
I've ran into more than a few utter piles though. I'm more selective based on total reviews, so I'm not getting anything cutting edge (10 reviews - 4.5 stars, no thanks, 180 reviews 4 stars, sure). I don't have friends or contacts of any sort that read what I do, so unfortunately that's all I can do.
Publishers provide value to themselves, not to authors or artists.
Argument from quality: nope. Most publishers won't touch quality writing, because they know the market for it is limited. Joyce would never be published today. Austen might, but Austen is a bit difficult, so probably not. Lessing was turned down by a major publisher when she tried submitting under a different name, just to see what happened. I have my doubts if Pynchon or Vonnegut would be taken on today. (Has anyone of equivalent ability been signed in the last twenty five years?)
There's a huge market for ghost-written celebrity blather nonsense, landfill SF and Fantasy, and two-glasses-of-wine-and-a-billionaire chicklit, and a much smaller market for self-absorbed litfic. But quality literature?
Argument from services provided: nope again. Editors aren't quite two a penny, but good ones are available for hire. Likewise cover designers. Because you're paying them direct, you'll find that an editing process that takes publishers 18 months can be completed in a few weeks.
Argument from marketing: Publishers literally won't lift a finger to help any author who isn't already selling in the millions.
Argument from financial support: A couple of $k up front if you're very lucky, in return for losing 85% of future income. And you'll probably be offered a crappy work for hire contract.
So - what is this 'value' that publishers provide?
Whether they're of equivalent ability is subjective, of course, but plenty of difficult but beautiful literary works are published every year. And I would bet that their first run is a lot larger than Ulysses' was :)
Authors I've read in the past couple of years that were not mainstream but were published recently include Roberto Bolano, John Banville, Tom McCarthy, Eduardo Galeano...and I'm not much of a fiction reader. There's an enormous number of books published and almost all of them are not mainstream.
Which is almost exactly what happened to JK Rowling, only her books are slightly more accessible.
The big pubs do not take risks. If a writer can show evidence of critical interest and existing strong sales, they suddenly become interested.
Signing on the basis of a strong, original manuscript? Not any more.
I'm sure there's an exception every year or five, but it's a good bet that unless you're incredibly lucky and/or know commissioning editors personally already because you all went to university together, the odds of getting in through the front door are not good now.
Yes, publishers do provide a value. Also one of the best selling books of late is a 700 page non-fiction book about finance, which reached #1 on Amazon's non-fiction list, published by Harvard University Press.
Saying that all publishing is just mass media crap is a just plain false.
As a reader I like finding publishers like that because it reduces the noise for me considerably and improves discoverability: browsing http://mitpress.mit.edu/books/new-titles or http://www.zero-books.net/forthcoming-titles.html is actually manageable. With a few such publishers (I follow about 5 semi-regularly) you can keep abreast of a lot of interesting things happening without having to recognize the individual authors or do extensive research on them; I outsource the job of doing that to the publisher.
And it remains true that top-selling established authors do okay out of trad pub. The 1% of writers in that category get decent enough advances and some PR support.
Everyone else is wasting their time and getting screwed over. Including a tragically large number of people who write for university presses. (I really hope Piketty got a lawyer to look over his contract.)
Writers are not, generally, minded to think of writing as a business full of business people doing business people things.
But some do - in increasing numbers. If you know anything about the industry, you'll know that where ten years ago it was taken as read that you needed an agent and a real publisher, otherwise you were just some noob with a vanity contract, the reality now is that writers are stampeding away from the agent+publisher deal package.
Agents are literally panicking about their jobs.
The charge has been led by romance writers, who have a terrifyingly efficient business and support association (the RWA) and not a few self-made self-pub millionaires.
The problem with a site like HN is that most people posting here know very little about the industry. But when you're in it, and you've been following developments in detail for a few years, the picture doesn't look the same as it does from outside.
I'm certain that Amazon marketing has played a significant role in the initial success of this piece. On the other hand, I proofread the final draft that the author had before approving the piece for publication. The Amazon/Kindle person he had been working with as a proofreader supposedly had 25 years experience proofreading at the NY Times. I probably found 100–200 errors, not the "I would do it like this: kind, but the "this is wrong" kind.
When I spoke to my friend/author later on, he said that the proofreader is simply overwhelmed with work. I can't remember how much she once described to him she'd have to do in short order, but it was a ridiculous amount of work.
So while there is a value add to a middle man like Amazon, I can see how there are shortcomings to what they offer since the grunt-work side of publishing isn’t their focus, even if they do a bit of it. That’s not to suggest the long-standing publishers are better, just that there are shortcomings to what each style of publisher offers, and it may be difficult for a new author to learn the shortcomings before it’s too late.
I am not asking snarkily; I am genuinely curious. I'm contemplating self-publishing, but I'm aware that Amazon is a pretty noisy landscape. If there are ways to power one's signal through the noise, I'm interested in learning about them.
The labels really are dying. They'll hold on a few more decades, but only because they'll not let go of their venue monopoly willingly.
And books? When's the last time you were in a Borders? How long before all those old Barnes and Noble storefronts are just Halloween stores for 2 months out of the year?
They too are losing control of the publishing and distributing process.