Still, I'm inclined to think it makes more sense to listen to someone who's done something that worked than someone who isn't interested in trying it "because it won't work"...

...until the power bill spoils your fun.

Especially 'older' x86 gear is easily in the 130-150 watts range idle, compared to ~10 watts for a typical home router. Another issue is the antenna situation, you don't want long cables to 2.4 GHz or 5 GHz antennas, but at the same time keeping the close to a big steel PC case affects your reception as well. The same goes with the price, while you can get a decent 2.4 GHz wlan card for around 20 EUR, 5 GHz capable ones start around 40 EUR - so the radios alone easily reach the price ranged asked for pretty good mass-produced plastic router (which have no interference/ shielding issues).

In most cases, unless we're counting the number of concurrent users in the medium 2-figure range, a cheap plastic router is a much better choice, which pays off within a few months just through saved electricity. With only a bit of searching you can even find pretty hackable devices as well.

OP said "an old netbook".
I don't know which one he has, but my 2009 Atom N270-based Aspire One netbook ran a little under 20 watts, per Powertop. That's hardly worthy of mention.

If it's a netbook, there's no steel case.

If it supports master mode in all the operating systems named, my guess is he has an Atheros card.
Those can be pretty good, depending on the card; a number of the commercial routers use them, though DD-WRT targets Broadcom cards.

That would be much better.

"So what do you think would BG's return signal a reemergence of MS?"

Either omit the "what" or add a dash after "signal". As it stands it's nonsense.

This has 8 MB flash.
(FYI, with Busybox 1.20.2, I've not found scripts that make it barf to be common.)

Can't say how much of the funtionality is onboard, but they claim that Quark is a SOC.
There's a note about compilers to the effect that it's 586.

I think part of the point is to provide a design that uses the Quark.

My point in that comparison was "just linux" vs. "full android stack".

If it's NT kernel vs Linux kernel, I can boot Linux in 4 MB--with 3 login shells.
(I'm serious: I linked busybox statically against musl, configured a pure busybox /, set the login shell to ash, and booted with mem=4096. About 2 MB free once boot was over, IIRC. No swap.)
Android, on the other hand, has a display manager and a VM to fit in there. "free" on a Gingerbread phone just after boot claims ~ 200M used. That probably includes a bit of bloat, but I don't imagine it booting in much under 128 MB.

So when the servers for the exchange are this far from working on October 4, you think we should go ahead and fine anyone who hasn't gotten an approved insurance policy by Jan. 1?

If you're speaking of Bowman v. Monsanto, that is incorrect.
Bowman was about a second crop.
In case of a legitimate sale, the purchaser has the right to use the product sold for the purpose indicated: growing a crop for market sale. Any licenses from the seller which may be necessary to use it for the purpose indicated must be included with the sale.
Bowman had the right to plant the seeds he bought. That was not what he got sued over.

Bowman also had a second generation; what he planted the second time were seeds that, per the license, he had the right to sell--but not to plant.

Monsanto does sell licenses to grow seed for planting, but that comes with a royalty. If you use traits from Monsanto, you end up paying Monsanto for each generation you plant. It works the same way for seed companies.

(Note: I'm not arguing that the current laws are ideal. But they aren't nearly as bad as some people make them out to be.)

That's unlikely. So much so that I suspect sarcasm.

Corn is grown by individual farmers, who purchase seed from several large seed companies and smaller companies that sometimes license traits.
So, big picture:
1) Picture the difficulty in manipulationg over a thousand grain elevators, or in suing half a million farmers at the same time...
I doubt that many companies could pull that off.
2) Patent exhaustion incontestably applies to the first crop when a company sells its own seed.
3) Monsanto's far from a monopoly; even if they managed to shut down all the corn that was grown from saved seeds and from their seeds, that would probably be less than half what has been planted.
4) It's not like them. They have been hard at work improving public image, though not everyone is convinced.
(I say this as someone who interned at Pioneer.)
5) Two words: corporate suicide.
It's not like Pioneer (DuPont), BASF, Bayer, and Dow are irrelevant in corn or soybean production. And two of those are larger companies than Monsanto.
Meanwhile, Monsanto specializes in seeds and chemicals.
I don't see any way that scenario could happen without their seed marketshare going to 0.

But how many of the 200+ above Monsanto have a real impact on food production worldwide? How many more or less decide the steady march towards agricultural monoculture that has been predicted by many to be the first step in a crop collapse?

Let's see who's involved in ag-related industries and above Monsanto:
Food processors:
Archer-Daniels-Midland, ConAgra, Tyson Foods, Smithfield, and a few more. I'm excluding bottling companies like Pepsi and Coca-Cola.
Manufacurers producing ag equipment among other products:
Ford, Caterpillar, Deere & Co.
Chemical/drug companies with major ag lines and a larger total size:
Dow (ag chemicals, seeds), Merck (veterinary), DuPont (ag chemicals, seeds)

Monsanto is in the same vicinity as Waste Management and DISH Network. I named ten companies that are larger.