front - face recent changes local names special software-bazaar-wiki


front - face

clock design principles

<quote> These are the principles that Danny Hillis used in the initial stages of designing a 10,000 Year Clock. We have found these are generally good principles for designing anything to last a long time.

– from the Long Now foundation </quote>

Are any of these principles relevant to the distributed wiki?

longevity: Unlike this mechanical clock, any one part of the distributed wiki can be replaced at any time, without stopping the system. So it would seem to be even easier for people to keep the distributed wiki to keep running for millenia, compared to a (not-completely-fault-tolerant) clock. But my goals are more modest – if it runs for 1 century (until 2108), that will be far more successful than I expected.

maintainability and transparency: This is even more important to the longevity of the distributed wiki than to the clock. We expect each hard drives to fail and need replacing after an average of 5 years. (So over 1 century, each hard drive is expected to be replaced 20 times). That is much more often than the expected part-replacement rate of any mechanical clock. I see no reason why the distributed wiki shouldn’t survive a transition from rotating hard drives to Flash data storage, and a transition from x86 processors to ARM processors, and a transition from IPv4 to IPv6. (While it may be possible to “translate” this system to quill-pen on paper, or stylus-on-clay, or some other bronze-age technology, that’s a different project). Even though single failures should be invisible to “regular wiki users”, we want people able to fix those failures to (a) quickly see that something failed, and (b) easy to fix.

scalability and evolvability:

Define external redirect: LionKimbro grammar check SamRose BrandonCsSanders