Frequently asked questions about parasitic computing
- Will you stop using my computer?
We are not using your computer.
Most external testing was completed months ago, on consenting
hosts.
All current tests are conducted within our domain.
- Is your code available?
No. It is not.
- Can I have a copy of it?
No. We do not plan to distribute the code.
- Is someone using my computer for parasitic
computing?
It is doubtful that anyone is using the TCP checksum technique
described in a
Nature
article.
However, it is certain that if your space is public, someone is
using it without your knowledge and in a way you did not intend.
- How do I reliably stop parasitic computing
from occuring on my web server?
Unplug it from the net.
- What assurance can you give that the TCP
checksum technique is not a threat to my machine?
All messages are ordinary TCP segments going to an HTTP server.
Such messages are simply invalid HTTP requests or TCP segments.
Your web site receives such messages from browsers on a regular
basis.
Therefore, the TCP checksum technique is no more of a threat to
the integrity of your web site than a web browser.
- Do you think computing by the TCP checksum
will become widespread?
No.
Consider fusion.
Nobody uses a fusion reactor. Why?
It is not cost effective. It is endothermic. It fails
the cost-benefits analysis.
So does the TCP checksum technique.
Nobody doing real work will use this technique.
- What about a hacker? Could this technique
be used against my site?
Any request loads a web site.
If one's goal is malicious, there are far more effective
techniques for loading a machine than the TCP checksum
technique. For example, ping is
better at exploiting your system
than TCP checksum computing.
Our technique uses less resources per message on a server
than ping, and it sends messages at a much slower rate than ping.
- Is parasitic computing
hacking?
No, it is not. It does not compromise your computer security, as
it simply sends standard packages to it, which are either
dropped by your computer, or returned to the sender.
- Why publish this?
This is a discovery, not an invention.
Therefore, parasitic computing has always existed. If we
didn't expose it, someone else would have.
If it is a threat, then it needs publicized, where it can be
discussed and addressed.
It is better to be made public now in this way, then to be
discovered after someone has made it efficient, and widespread.
- What if I'm still worried about
this?
Read the paper.
vin@nd.edu
Last modified: Mon Sep 3 14:54:56 EST 2001