No reason you couldn't do both, and just have the smartphone app interface with the same system the in-house tablets do. (And not everyone will leave a smartphone.)

Menu systems are simple, and don't need much in the way of processing power. They don't need to be full-featured web browsers, or be able to run high-end games. An e-Ink screen would be enough for most places. This approach even makes sense from a recycling point of view - just use screens from old iPads or Kindles for the in-house tablets.

The value in high-end processing is the multimedia upsell. Replacing a paper menu doesn't require much processing power, but it's worth a lot to increase the check average and/or improve margins by promoting higher margin items.

pushing out your own hardware costs more than just building software for a widespread platform...

Another fine example: the Kno

Then the tablet is just a menu. The companies that I have seen working on this kind of product are adding card swipe readers and printers to expedite payment. Otherwise you're still waiting for a server to take your card and come back with a receipt.

That, and you need better signature capture.

By the way, I don't think you need a better signature capture. You need to GET RID OF SIGNATURES, like they do already in a lot of countries... :-)

By "better" I meant "works with a stylus". Most of Europe might be paying for meals with PIN-based debit, but a lot of Americans still pay with signature-based credit. The banks need the signature.

Not just europeans, but also japanese and all countries in south america (not sure about other asian countries and africa). Also, pin works for credit cards too, in these countries, not just debt

Why do the banks need signatures in the US? Specially badly-drawn ones that don't even ressemble the actual signatures? (I can never sign properly in one of those POS machines from CVS, ever)

Signatures allegedly lower the fraud rate, or at least assist in proving fraud if a card is stolen.

Some banks now offer the option to bypass signature capture at the tradeoff of a slightly higher merchant rate. Many quick-serve restaurants (Chipotle comes to mind) will take the hit in return for processing customers faster. Time is money.

Given how badly I sign the UPS pad when picking up a delivery, I can't see these signatures as being usable.

why not pin codes, then?

I've seen companies working in this area for years too (my first company actually WORKED with this - for palm devices :-)) - my question was just about that product in specific and it's tie in to a proprietary hardware piece.

I guess ordering dishes out of your own Android/iPhone feels more like ordering a take-out instead of a sit-in? I think e la carte is intended to offer more on enhanced dining experience, rather than just an easy way to pay for your dinner.

I didn't mean the CUSTOMER's tablet/phone, I meant buying tablets (or making the restaurant buy them) instead of manufacturing your own

One possible argument for proprietary tablets is that the value to a potential thief is much less than a commodity tablet which could be repurposed. Perhaps the maker of a "standard" Android tablet could be convinced produce a version slightly modified so that their standard distro would not work without modification? While thieves clearly could hack the standard distro to make it work, would they really achieve high enough theft numbers to justify the effort?

I think buying general purpose tablet would definitely cost much more. Also, pushing restaurants to first buy an ipad/android tab, and then using your app on it is just too many hoops to jump through for any restaurant.