The catastrophic collapse scenario is of course the most interesting in many ways. Vinge pointed out that we depend heavily of embedded microprocessors and those devices are all very phyiscally similar. One obvious failure is the electromagnetic pulse of a high-altitude nuclear blast. This would cause effects such as the immediate failure of most modern automobiles due to the destruction of controlling microprocessors.
His second scenario was called, "legacy software forever" and was the plateau situation. In this case, future civilizations could end up with literally thousands of years of basically the same software as we have now. Vernor has written about this in some of his work and raises an interesting question: when will it be better to dig through all that accumulated software versus reinventing the wheel?
Another good point was that if hardware reaches this plateau, certain kinds of regularly-constructed devices could escape that fate for a while and continue to expand. The result in the future could be a laptop that is essentially the same as what we have now - but with a memory card containing everything that has ever been written. That is both intriguing and depressing at the same time.
The third possible future related to Moore's law is that computing power advancements continue forever. This veers completely into science fiction because Vinge believes there will be some sort of 'singularity' where there is a complete break with the past. For example, the internet itself could become self-aware. While this is a fun speculation, it's also so fantastical that I'm not sure you can draw any useful conclusions from it.
Vinge's final scenario is "Ubiquitous Law Enforcement". In this future, every microprocessor contains a government-owned section that controls all I/O and any other function the government sees fit to control. It would be illegal to use a system that did not have this feature. Vince noted that while this future is in many ways terrifying, everyone finds something to like about it. It could enable completely seamless tax collection as all transactions are monitored. It would allow authors to reliably collect micropayments for their work. The big caveat is that someone would have to design this hardware, and it wouldn't be the policy makers. It would be the chip designers, and they could end up as the most powerful people in the world. Vinge feels someone is guaranteed to attempt to implement this future (and it is occurring now for things like Digital Right Management), but it is the least likely to succeed.
While there are more USENIX activities this afternoon, I don't have time to write about them because of scheduling conflicts. Thus, I will wrap up my report now. First of all, USENIX is an amazing experience. One key reason for this is it is self-selecting. Nobody is forced to go to this conference. People come because they want to. That leads to a very high level of technical presentations and discussions.
At the same time, USENIX has some problems that have to be addressed. The big one is there aren't enough young computer professionals here. The organization has to do a better job of recruiting people. This should probably start at the college level. If someone goes to USENIX in college they are much more likely to ask their employer to send them after they graduate.
The bottom line is that if you love Unix and computer science, you should be going to USENIX. I plan on attending next year and I hope to see some of you there as well. I hope you enjoyed my report and please feel free to contact me if you have any questions.
About the Author:
Phil Hollenback is a System Administrator at a financial company in Manhattan. When he's not taking care of computers, he
spends his time skateboarding and working on his website.
If you would like to see your thoughts or experiences with technology published, please consider writing an article for OSNews.
- "Usenix Day Three, Page 1/2"
- "Usenix Day Three, Page 2/2"
