This does make me wonder about something though: is there a crashplan for a compiler? Can a very minimal gcc be written which would be capable of bootstrapping the entire gcc compiler?
It seems like it would be relevant if a new architecture was developed at least.
>Can a very minimal gcc be written which would be capable of bootstrapping the entire gcc compiler?
Yes. In fact, this is how gcc is built - in stages. First stage: build a compiler that can build the compiler. Recompile the compiler with the newly built compiler. Do it again. Repeat until completion. ;)
You can conceptually do the same thing now, but have to jump through hoops first. Download the right dev environment, work out how to start an editor, work out what code to write, work out how to run it and finally see the results. Devices are very personal now, so no one is likely to watch you do this, and trying to mimic it is far harder than turning your own device on.
Hopefully we'll end up with single function devices that can do this. The RPi is a good step in that direction, but still requires a lot to get going (look at all the cables and other bits and pieces you have to connect first.)
There are a couple of things "wrong" with the way we teach kids computers today.
Our educators are in the hands of the masters of consumers. When a new technology comes out, we all abandon the current stuff, and engage in the upgrade cycle. Repeat until generations have no clue any more about what the old ways were.
This is an absolute falsehood pushed upon society by those who wish to control the consumer base. It is consumption destroying education, plain and simple.
The point is this: Every C64 that was ever produced - heck, every 8-bit computer, ever - STILL WORKS, or can be MADE TO WORK in the area of computer education.
It is absolutely arbitrary that computers get old. Every machine that was ever made, is still just as useful as it ever was - the difference is, the user walked away (because they are consumers not users).
I have a large collection of 'antique' computers in my midst: C64, Atari, Oric, Atmos, Telestrat, MSX, heck .. even a BeBox and an SGI O2. All are still working, all are still quite capable of engaging a young mind in the exercise of exploration and discovery that makes a good developer.
And, my 3 year old and 6 year old kids LOVE THEM. They absolutely LOVE the old sprites, the old simple ways. The 6-year old takes immense joy out of typing:
.. into an old Oric Atmos thats been set up exclusively for him to be able to do that .. in fact the very first writing he was able to do was in typing in a BASIC PROGRAM!!The 3 year old absolutely loves that he can turn the machine off and on, and off and on, and off and on .. and it will still work. Can't do that with Daddy's workstation!
So the point is, parents: disconnect your kids from the consumer trap. Give them old computers to learn computing on. Everything they will ever learn, WILL STILL BE VALUABLE TODAY when they 'grow up and get a bigger computer' - the reason is, because computers still work, fundamentally, the same way.
I predict my 6 year old will be hacking in assembly by the time he is 10 - just like his Dad did. And thats what made me the developer I am today.
(BTW: yes, I also have rPi's, Beagleboards, and so on.. when they're ready, they'll be available to the kids to hack on. But if the kids can't do their own low-level programming by the time they get the rPi dusted off, I will be very surprised..)
EDIT: Another thing that is 'wrong' with computers today, imho, is the decoupling of development from use. Again, our computers have been turned into consumption platforms - the moment that Microsoft removed the developer tools from being part of the base OS image, computers started to lose a lot of value. Any OS that doesn't ship with a way of building apps for it, inherently included by design, isn't an Operating System - its a Consumer Capture System.
Do everything you can to get development tools back into the OS, people. It is more important than the desire to reduce the effect of having 'too many smart developers out there'..