alternatives already exist
Oddly enough, rake, from ruby, comes from a language with no compilation needs, but does quite a bit many of the things you describe, I'd suggest you take a look. I've not seen "platforms" yet, but if you set a flag inside the Rakefile, call it platform, and use it inside your targets
The dependency parser came about, as a I understand, part of a desire to reimplement make in ruby, and part because of the needs of projects, like ruby on rails, where one has to upload/synchronise in order, to make automatic script generation easier, etc...
Seems compilation kept the dependency tool to itself for too long, now others are claiming it back.
Re: "The sky is falling"
> scams. Uninformed users will fall for
> these scams regardless of OS or
Most of these users aren't just uninformed, to the samples of those users I've met, requiring of them to keep informed is worst than having a non-functional Internet.
Now, conceptually, the Internet is built on the idea that participation is very much requirement-free, since the protocols were built back when it was a privilege to connect. Right now, it's considered a free speech issues in some countries. The basic protocols haven't been updated enough that we can consider excluding participants who behave obnoxiously on the technical level. On the legal/social level, those processes are slow, and the layer between physical identification/authentication is weak and permeable. Perhaps one solution would be to upgrade protocols where improper behaviour is automatically detected, and the network routes around the problem caused by the rogue node, another would be to endure the problem, since the social/political/legal implications do not allow a simple solution(think of the censorship implications of having hardware-defined "acceptable" tcp/ip behaviour. There isn't much room in the middle, either you consider it a privilege, and can revoke it, or you consider it a right, and need justifiable cause to revoke it.
It's pretty obvious to me that many of the problems due to closed source, and the fears about open-source, have a single source, accountability. Closed-source companies are so big they can hardly be considered to be accountable to anyone but their shareholders, and open-source companies are so many that tracking who is accountable for a particular bit of code is not administratively easy. Now keep in mind I'm not saying open-source is less accountable than closed source(the opposite is true, and that's my main point of agreement with the original article: open-source keeps transparency in the process), but, from the point of view of less technically-savvy, lazy administrators, it seems that way.
Unfortunately, we don't have savviness requirements for public office yet, even in technical positions. Perhaps that's one area where we can identify improvements.