Re: a "long period" for Stability

Hi Ron,

On 9 Feb 2012, at 19:16, Ronald P. Reck wrote:
> The problem I have with this is that "long periods" is a very ambiguous concept on the web. I know we touched on this at the F2F but I wish I knew how to tighten it up a bit.
> 
> - long periods to data at my house is through 2 hard drive standards
>   (MFM/IDE/EIDE/SATA..)
> - Long periods in the scope of the Internet is a couple decades...?
> - Long periods to a person might mean a generation...
> - Long periods of weather data could mean since the last ice age?
> 
> Any formative comments about how I can rephrase "long periods" to scope it better would be appreciated.

As I said at the F2F, one way of approach this is to think about events that threaten stability, and talk about steps that can be taken to mitigate the risk of breakage for each type of event. For example, perhaps in order of decreasing probability:

1. The person who published the data changes jobs
2. Departments are reorganized
3. IT infrastructure overhaul (e.g., change from one CMS to another)
4. Merger/acquisition
5. Primary stakeholder loses interest in the data
6. The DNS becomes obsolete
7. China buys your country
8. Downfall of human civilization

Some of these are probably out of scope. The document should make this clear in the beginning.

All the best,
Richard

Received on Thursday, 9 February 2012 20:02:13 UTC