So, how's that new drone initiative going over at Amazon? The one where octocopters the size of pizza boxes are going to wing lightweight packages directly to your front steps? My guess is meh. At least if today's Wall Street Journal is to be believed.

The paper claims that Amazon is developing a mobile app that will enable you and I, and pretty much anyone else, to deliver Amazon packages while we're on our way to other destinations. The idea is that Amazon would pay other retailers, mostly in urban areas, to store the packages until the crowdsourced couriers could pick them up. The Journal says that the timing for the initiative was unknown, and the whole thing might not even happen.

There are also other big hurdles, including the fact that brick-and-mortar stores aren't exactly dying to lend Amazon a helping hand. And the whole people-as-a-service aspect of the renting economy (sorry, it's not a "sharing" economy if you're getting paid to share) seems to be losing some of its luster as it becomes clear just how little many of these contingent workers, such as Uber drivers, are earning.

But the general idea--that Amazon is looking for a way to provide delivery, not just stuff--makes sense, especially in light of the company's drone initiative. Given that Amazon delivers 3.5 million packages a day, according to the Journal story, if it could deliver a good share of those itself, it only makes sense that it would start offering delivery to other companies, too.

The problem is that it doesn't just make sense for Amazon, it makes sense for plenty of other companies with deep pockets as well. The expectation that Uber will soon deliver a lot more besides people is just one of the uncertainties supporting the company's $40 billion valuation. A crowdsourced delivery network for all kinds of things--not just groceries--makes sense for Instacart, another unicorn company, and one that's raised $274 million so far. It makes sense for Deliv, which has raised $12.4 million. The boostrapped SlingShop (although it has its own drivers) is also working on same-day delivery, starting in Atlanta and leveraging its success with Zifty. The list goes on.

The big questions remain: How much stuff do we really need delivered near-instantly, and how much will we pay for that convenience? More to the point: How many fully funded companies can be supported in the market for instant gratification?

Most likely: Less than these big players think. Obviously, if people are hailing a taxi of any kind, they need to get somewhere pretty fast. It's hard to argue with Uber's original business model (aside from the fact that it was illegal in many cities).

Instacart makes sense, too. If your fridge is empty, and you don't want to do takeout yet again, you need your groceries to show up before your hunger pangs do. But beyond food--which is obviously a big market--you have to wonder how much room there is to expand this app-driven, deliver-it-right-away model.

When you buy a new rug for the bathroom or a gift for a new baby, does it have to show up that afternoon? More to the point, how much would you pay for it to show up in a few hours? And is the U.S. Postal Service's second-day delivery--one of the best bargains out there, if you ask me--really so bad? FedEx, UPS, and DHL aren't exactly pushovers, either.

Don't get me wrong: Someone's going to succeed at this, and make a mint. But lots of folks are going to fail, too.

With the success of Prime, Amazon itself has proved that for many, many purchases, two-day delivery is just fine. You can wait two days for almost anything. When I spoke with Forrester vice president and analyst Sucharita Mulpuru for a retailing story last year, she said the dirty little secret of Amazon Prime is that it often doesn't even deliver in two days. Some 20 percent of packages ordered through Prime, she says, take longer than two days to show up. But mostly, people don't care that much. "In most situations, for most customers, three to five business days is good enough," she said. "As long as you do that, people are happy."