Why exclude "memory and CPU"? You may have heard of something called Moore's Law, its one of the single biggest reasons why modern consumer electronics become obsolete so fast. Many of the "marketing" features you've listed are directly tied to memory and CPU performance (turn by turn navigation, Safari upgrades, multitasking, Siri). Windows 3.11 had 500,000 LOC, XP was 45,000,000 LOC, Windows 8.1 is 80,000,000 LOC. To say that "lots of it hasn't changed in 20+ year" is flying in the face of fact, or are those 79,500,000 LOC all "marketing" features that could easily be added to a Pentium III powered desktop.

I take it you've never worked in hardware manufacturing. Gross margin is typically in the ~30%-35% range (that is margin after the cost of goods sold), Apple and Samsung manage to get in the 40% range. Of that 10%-15% goes to R&D, 10%-15% to marketing, leaving ~5% for profits, jets and limos. It's just that R&D is being spent on new things as opposed to supporting legacy platforms used by a handful of people.

Firstly I asked if you expected your desktop to last 20 years, you responded "yes I expect my desktop to last 10 years", nice way to move the goal posts. A 1994 top of the line desktop would be using a brand spanking new 100Mhz Pentium Chip running Windows 3.11. What exactly would you apply this powerhouse to today? Now because you (and a small handful of people) are running 15 year old machines to perform some specific task, everyone is supposed to pay more to provide you with ongoing support and spare parts? For a 20 year desktop lifespan, Microsoft for example would need to write Windows 8 to be able to run on an original Pentium (with FDIV bug), or support every version of Windows from Windows 8 down to 3.11 (an OS that didn't even have an IP stack). Tell me who gets to pay for these armies of developers? I'd argue that the 3GS IS horrendously obsolete. While the it can up upgraded to iOS6 there are several features that do not function (VIP list, Offline Reading List, Shared Photo Stream, Siri, Maps flyover, Turn-by-turn navigation, FaceTime on 3G, Hearing aid support). It does not support iO7 (Control Center, Notification Center, Air Drop, Improved multitasking, upgrades to camera and photos, iCloud photo integration, significant upgrades to Safari, Find My Phone, Car Play, plus a bunch of stuff to support enterprise usage). Oh and iOS8 comes out this fall. The 3GS doesn't support LTE, HSUPA, nor 802.11n. The iPhone5 has 4-8 times the processing power of the 3GS, 4 times the memory, 8-12 times more powerful video processing, the camera has 3 times better resolution, does HD video, image stabilization, includes a front facing camera, plus the battery lasts longer too . This all within 5 years (technically only 4 as the iPhone5S came out in 2013), now imagine a 20 year old smart phone.

Yes, because keeping an army of developers on staff to maintain legacy applications from 20 years ago only used by 4 people doesn't cost anything. With the exception of Apple, most hardware manufacturers operate with a razor thin margin.

Why should smartphones be expected to last 20 years. Do you expect your desktop or laptop to last 20 years? Why engineer something to last an arbitrary 20 year period of time, if consumer behavior shows that 99% of them will be in the trash within 5 years? Moores law is still in play for now, meaning the phone I buy today (even if built to a level of durability needed to last 20 years) will be horrendously obsolete compared to the phone I can buy in 5 years. (Which will have 4-8 times the performance and do a wide variety of new things that my existing phone can't). Consumers respond by throwing away the old phone a buying the new. Comparing a smart phone to the old bell land line phone is like saying "My abacus was able to last 50 years, why doesn't my MacBook Air do the same"

One of the cornerstone principles of science is the concept of falsifiability, that is for something to be called science it needs to be possible to prove the hypothesis to be false through observation and experiment. String Theory for example is a work in progress, with many predictions that cannot be tested due to lack of our technical capability. That said, experiments can be devised to validate or disprove those theories, it just may be that those experiments have not yet been funded or within our current capability. This is why String theory is not held up as a generally accepted model of physics, its just one possibility that is being explored. Many variants of String Theory have already been discarded as they predicted observations that are not seen (large numbers of magnetic monopoles for example), or contrary to observed fact (e.g. atoms could not form), hence they are falsifiable. You have a lot to understand when it comes to the philosophy of science and what it means to conduct science. There are NOT "two groupings" of science, repeating the assertion does not make it so

No, the concepts of "micro evolution" and "macro evolution" are created by those who are unable to deny the vast quantities of observational data confirming evolution, but still don't want to believe that humans evolved from proto-apes. There is no scientific concept of these two separate phenomenon

People directly observe evolution every single day. Just go to your local university's undergraduate fruit fly lab. You can see it, test it, measure it, validate it. They've done fruit fly experiments where they have caused speciation (i.e. producing two branches of evolutionary fruit flies lines that cannot re-produce with one another)

Sorry, there are not "two kinds of science", no science can be absolutely proven 100%, but evolution has been tested and validated far more than the Standard Model Go into any undergraduate bio lab and you can directly observe the evolutionary process, be it with fruit flies, or antibiotic resistant bacteria. You can see evolution and natural selection taking place in real time. Of course now you're going to come back with some kind of argument that this is "micro-evolution" which is somehow different "macro-evolution" BS. Even the Roman Catholic Church has indicated that there is no intrinsic conflict between Christianity and the Theory of Evolution.

I'm sure you can then provide a citation for your conclusion. Every single thing I've read indicates pipelines are 4-10 times safer per barrel mile transported.

Returning the stage doesn't require a LOT of extra fuel, because the mass of the stage is considerably lighter once it separates (all the fuel required to launch the first stage, all of its fuel, the second stage, all of its fuel, and the payload has been consumed). This is why the re-entry burn only uses 3 of the 9 engines and the final landing 1 of 9. I believe Elon has quoted a 25% loss in payload as a result of the changes and fuel needed for reuse. Even if it costs $10M to inspect and refurb the rocket, the savings will be substantial. BTW do you get a discount on your airline ticket when you fly on a "used" aircraft? SpaceX can sell it as a qualified launch vehicle.

SpaceX has been profitable for several years now

Except it typically costs more (a lot more) to execute someone vs. incarcerating them for life.

Highly Radioactive and long-half lives are mutually exclusive. Anything that isn't gone in a few years is by definition, not highly radioactive. Again, it would be very bad, very very bad, but coming back to the topic at hand, it would not prevent future civilizations from exploiting these resources

Actually only the first 3 flights of the Falcon 1 were a total loss. All other flights have achieved every primary mission objective, including CRS-1 that suffered an engine failure that would that terminated the flight of any other rocket currently flying