3 ms·
Let me use your apple example, apple found out the hard way that paying its competitors us against its interest so they make their own M1/2 SoC now, their samsu
by _8j50 3y ago
Let me use your apple example, apple found out the hard way that paying its competitors us against its interest so they make their own M1/2 SoC now, their samsung dependency on memory chips is because there is no option. They do everything in house where possible, they line to control as much if the supply chain as possible.
Amazon is launching a new product, so they can calculate short term losses and plan for long term success. Initial committments like spacex would be difficult to move away from. Total cost of anything is worse if you have to migrate from it anyways. If they know they will find it cheaper to use blue origin in the long run then the only decision is what to use that is cheapest to move away from down the road.
The bottom line is, the board and company of amazon has no fiduciary responsibility to pursue short term profits. They can only do wrong if it can be proven that they made a decision that is against the proftiability of the company intentionally or out of gross negligence.
- philwelch 3y ago> [Apple’s] samsung dependency on memory chips is because there is no option Which is also the state of the launch market at this point. OneWeb uses SpaceX launches as well. Apple also buys screens and batteries from Samsung, by the way. > They do everything in house where possible, they line to control as much if the supply chain as possible. Apple is more vertically integrated than a lot of companies, but this is not true. They don’t even do their own manufacturing—Foxconn does it. The only parts they do in house are the parts that they have a competitive advantage on, which at this point includes chip and systems design and software development. > Amazon is launching a new product This is actually more incriminating, not less. Amazon may argue (not convincingly, in my opinion) that partnering with SpaceX would be a bad business decision. But even if that’s true, it doesn’t make partnering with Blue Origin a good business decision when they have the alternative of not launching this product in the first place. > Initial committments like spacex would be difficult to move away from. How? There’s no inherent lock-in with satellite launch. If you contract with SpaceX for n launches, you’re not obligated to continue using them afterwards. Many if not most SpaceX customers have also used other launch providers. > If they know they will find it cheaper to use blue origin in the long run They know no such thing. Blue Origin has faced years of delays and has never made an orbital launch. > They can only do wrong if it can be proven that they made a decision that is against the proftiability of the company intentionally or out of gross negligence. And in this case, there is an obvious ulterior motive at play along with virtually no sensible business case for the decision they ended up making.