Jump to content

Talk:Separation of protection and security

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Deletion of this Article?

[edit]

I suggest we delete this article. It appears no one can explain what this means. John (talk) 17:56, 23 June 2010 (UTC)[reply]

Yes, please do. — Preceding unsigned comment added by 79.168.138.50 (talk) 21:22, 14 October 2016 (UTC)[reply]
"I don't understand it" is not a reason for deletion. - SummerPhDv2.0 01:25, 9 January 2017 (UTC)[reply]
No, but "nobody can understand it" might be. — Preceding unsigned comment added by 79.168.138.50 (talk) 21:01, 12 January 2017 (UTC)[reply]
This article ought to be deleted. When this page was first created in 2007, the primary source of this claimed distinction was the paper by Wuft et al., which is about HYDRA, an early microkernel. That paper states on page 450, "it is important at the outset to distinguish between protection and security," following with: "In our view, protection is a mechanism" while "security is a policy." This page's other references (e.g., Landwehr, 1981), while meaningful as early computer security publications, do not affirm this proposed security–protection distinction. The other references similarly lack relevance to the primary purpose of this page.
Even were this distinction used in early computer security publications, it is clearly unused today. To the extent this distinction has been used past the 1980s (if at all), it seems to only be used in the lecture notes of college professors who found this Wikipedia page and copied parts of it in (as can be seen by searching around Google). The only pages that link to this one either do so without explanation or do so as a misunderstanding of this page as addressing the mechanism–policy distinction).
I personally found this page from the Bell–LaPadula model page, which contains a sentence in its intro section stating that the Bell–LaPadula model is an example of a model "where there is no clear distinction between protection and security." Verifying the cited reference for that (Landwehr, 1981) brought me here. Reviewing all of what is said and the associated references for this article, I agree it is ripe for deletion. 72.198.108.97 (talk) 01:02, 22 December 2024 (UTC)[reply]

Limitations

[edit]

This article needs to acknowledge the limitations of expecting trusted application-level enforcement of certain policies (e.g. those modeled by partially-ordered lattices) with mechanisms that ensure only separation. John 21:09, 28 August 2007 (UTC)[reply]

Clarity

[edit]

Many sentences are vague, but starting with the first sentence:

It is not clear how or if architectures adopt distinctions.

"Usually means" is very ad-hoc qualifier, can we articulate what this means with precision?

(Some or all) protection (from what?) is provided by fault tolerance... Does this refer to protection from faults?

Both fault tolerance mechanisms and security policies can be implemented or enforced with OS or HW.

Is the intended distinction clear enough?

Some of the things cited subsequently as policies are not obviously policies.

If we can't make this article say something, we need to consider deleting it.

John (talk) 00:50, 21 September 2008 (UTC)[reply]

Can this stuff be illustrated with examples, like with cows?

[edit]

If cows can be used to explain communism and capitalism, why not separation of protection and security? Anyway, I think that some clarifying examples are in order. 76.24.104.52 (talk) 15:30, 7 June 2010 (UTC)[reply]

Do you have a reliable source explaining the concept with cows? - SummerPhDv2.0 01:24, 9 January 2017 (UTC)[reply]
More importanally, can cows be used to explain cows, preferably to the cows themselves? Now there's a mindbender. — Preceding unsigned comment added by 79.168.78.24 (talk) 09:04, 10 April 2018 (UTC)[reply]