hyrarx operating system
computer internal security
notes on an hierarchy of confidences
The current mode of internet computer-internal security operates with
file and folder distinctions among the owner, a group, and the public.
But intrinsic security exists widely among the programmatic entities:
architecture levels:
- The lone user;
- An uncertified application;
- The authorized user at various levels;
- A certified service application program operated by millions,
and surviving some level of hacking by expert thousands (scrutiny);
- The certified browser built by a security concerned company,
and regularly used by tens to hundreds of millions,
and its software code and specifications scrutinized by millions,
and surviving some degree of hack-attempts by expert thousands (severe scrutiny);
- The operating system built by the computer-concerned manufacturer,
accessible predominantly by the browser and applications;
- The certified operating kernel;
- The certified computation engine, machine (hardware, firmware);
- And various others programmatic entities of various intents on security,
various intents of usage, various intents of purpose,
various intents of scrutiny (and hacking, intense scrutiny),
and various prerogatives on actions.
The hyrarx (operating system hierarchy)
is concerned with the responsibility and reliability of requests:
Something akin to a law of auto-managing computics:
The well-constructed system shall not deprive the authorized user
of its computational responsibility, reliability, and resourcibility:
the user-computer interface general attributes.
Common technology follows two regimes:
the organizational and the interactional:
The former defines which pieces are sent where, to perform;
The latter defines pieces as performing entities with capabilities, of performance.
This is like the two methods describing 8-complexor octonions:
The outer, complexor description, puts and sends numbers in array,
while the inner, the octonion description, localizes rules for each type of element,
the 7 vector [virtual, including imaginary]
components are discovered to be fairly equivalent,
and distinguished by affixed tag-names rather than by array-placement.
[The 8th component is scalar, real, and not so complexified]
But the results are not directly compatible [not used both together]
in format and process, though essentially equivalent - such anomaly
being classified meta-political and not in the scope of this article.
[In mathematics and sciences we generally learn both methods]
MEMORY-STORAGE TAGS:
Program memory-storage is usually given a parity (one) bit check, which can catch
spurious single values, with the expectation that it will do so before a double-bit
can pass uncaught (a parity bit being simple binary-modulus count of memory).
But a simple furtherance to 2-bit or more, can check both the occasions of spurious
double-bit values, as are quite possible in very large memories, and other types of
spurious values, such as tied bits, and especially program zoning: catching a
mislocated program jump in very large memory storage:
Akin to the four-color-map theorem-construction: by setting each program it's own
parity-scheme, a jump out-of-range would land to execute in inconsistent parity,-
and in short order, find itself:
Each program or segment would need its own parity rule (masked ex'or).
[under construction]
.
.
.
A premise discovery under the title,
© 2001 GrandAdmiralPetry@Lanthus.net