Siemens doesn't have to be a bunch of idiots to not pursue devices that test on smaller amounts of blood; they might simply not have an incentive to do so. If their business model is to sell devices to labs for $$$$, reducing the amount of blood is probably not a big selling point. It only becomes so if you're trying to bypass the lab and go more directly to the customer (as in Theranos with Walgreens). But why would Siemens do that, when it's a less profitable product that directly competes with the other?
For all we know, Siemens already has a smaller, cheaper device that uses only a drop of blood, yet no reason to sell it until they get some competition.
The biggest risk from a medical standpoint of using a machine like Theranos wanted was using capillary blood. You really would need a lot of data to interpret the results like we have for venous blood.
Do you have a source for this? As far as I remember past discussions on HN about Theranos, testing on small samples is notoriously difficult. As an experimental physicist, this seems plausible to me: Say 5% error margin on your result is required. This leaves a lot more leeway for dimensional errors (of the sampling chamber) compared to a few µl. Not to speak of introducing external impurities (or those induced by the sampling as mentioned by the article). I am pretty sure Siemens has lots of incentive to err on the safe side.
If we really like to disrupt this industry, it would be nicer to be able to make these tests in-vivo, in my opinion.
In case you're not a native English speaker: "for all we know" means something like "hypothetically". icebraining's point is that Siemens' behavior is not inconsistent with Siemens having a secret device that can reliably test on small samples, even if they don't sell it. icebraining did not mean to imply that Siemens actually has such a device.
The car industry has essentially always been very competitive, with a number of innovative companies. Consider the rise of hybrid cars with Honda and Toyota leading the way. GM famously did a lot of the pioneering of modern EVs.
Also, the point grandparent was making is that you see a similar pattern of deception from Tesla. Sure, maybe sometimes it works out. But not always.
Another interesting question is whether deception is actually bad for human advancement in the long run. In fact, maybe the answer is no. I bet a lot of innovations came about because of founders essentially lying to investors to get a check and then ultimately finding a way.
Its really interesting, that SV VC seem to have forgotten that very basic question: why have the existing players not come up with something like that? Very often the answer is, they have and they tried and they figured out a long time ago, that it probably doesnt work well enough in the real world.