On 2 December 2014 at 10:29, Alec Leamas <leamas.alec@gmail.com> wrote:
On 02/12/14 16:45, Zbigniew Jędrzejewski-Szmek wrote:
On Tue, Dec 02, 2014 at 09:42:24AM -0500, Ben Cotton wrote:


So, a revised/enhanced/even worse set of questions:

[cut]

Some of those questions go too far imo and expose too much about the
individual machines and users.

While I think the privacy concerns are a core issue, wouldn't it make sense to leave this aside initially, trying to understand what kind of useful information we might need without any restraints?

We will later on face many limitations, privacy and trust is certainly some of those. But as a starter, why not try to think free?


Mainly because if you do that you sound like the NSA to most people these days.

Privacy is like security. If you don't think about it at the startup trying to implement it later is much more expensive or impossible. It also has various legal implications which have to be worked through. Skipping that and "being free" gets you a ton of "wouldn't it be great if..." that ends up being bikeshedding because it can't be done.

Everything an organization collects has to deal with these two 'facts'.
There is no way to easily anonymize data that a set of high-school students with a starter book in statistics can't undo in a week. 
There is no way to easily remove data from gathered statistics any more than it is possible to make the internet forget an email.

From that you have to figure out if collecting that information is viable for a volunteer organization to later manage.


--
Stephen J Smoogen.