Apart from some "testing", one certainly feels as though the interface between the Agent Device and the administrator Client Device has reached a state of completeness. This is the best work I've ever done.
A big "Guten Tag!" to all my German friends. Let me know what you think about Clique Space(TM) some time perhaps.
Saturday, November 26, 2011
Sunday, November 13, 2011
The Clique Space(TM) Client Device is a Lie Detector
Everything is a device in Clique Space. Agent Devices (those devices that make the Clique Space system) and Client Devices (any other device, including the administrator Client Device being developed in parallel to the Agent Device) are simply different devices to Clique Space. Agent Devices, being mere devices, are subject to the same control mechanism as any other device. This point is an important designed intention.
Agent Devices establish Connections amongst themselves through a process of engagement. The product of this process is something I have come to call a synapse - seeing as it appears remarkably similar to one of these.
Client Devices establish one or more Connections with one or more Agent Devices through a slightly different process, and although this process yields something similar to what the product for the Agent Devices is, the fact that the Client Device can engage with multiple Agent Devices (there is no real reason to say that a Client Device cannot engage a single Agent Device more than once, but I have yet to test such a scenario) allows the user of the given Client Device (should it be equipped with a View adapted to some medium) to detect lies.
Lie detection... how does this happen?
Well, the Client Device can't tell you who is doing the lying, but it can tell you that one or more of the Connections to the (serving) Agent Devices are sending information relating to one or more Elements that is inconsistent with one or more of the same Elements sent from one or more other of the serving Agent Devices; indeed, a form of lie detection. So, how does this work?
Summarily, Elements are sent from an Agent to a Client Device through tokens, and each token is kept by the Client Device until the token's serving Agent Device tells the Client Device to remove it. An Element is recreated as a composite of the different tokens. Hence, the term that I use to describe the exchange of a token: "projection" makes each token a little bit like a light beam projecting an image on to a screen. The more light sources one has, the more accurate one might expect the image to be.
Lies can be detected in two basic ways: 1. token inconsistency; should the details of any set of tokens describing an Element differ in ways that another set of tokens describe the same Element, then that Element is "token-inconsistent". 2. Element inconsistency; relating to the inter-Element relationships between two or more Elements, "Element-inconsistent" relationships become apparent when these relationships (say, between an Affiliation and its component Account) are found to contradict through Elements only once they are assembled. In the case of the Affiliation - Account relationship, this element inconsistency will become obvious if, when composing the Affiliation from its tokens, one or more of these tokens discloses a differently identified Account than the Account assembled from its tokens which is stating that it points to the given Affiliation.
In the case of the Element-inconsistent Affiliation - Account relationship, these relationships only become evident at the time of the Elements' assembly within the Client Device to which the tokens have been projected.
Such contradictions could point to some problem with the serving Agent Devices, but yet again, they probably also point to nefarious intent of the operators of the serving Agent Devices. So, if you're a Clique Space user, and you don't trust the opinion of the person operating a singular serving Agent Device, seek a second (or perhaps a third or a fourth) opinion and ensure safety in consensus through simultaneous Connections to multiple serving Agent Devices; preferably ones which each are operated by different individuals. No serving Agent Device's operator would (nor should they) ask that you only connect to a single Agent Device, or only restrict your connection to those Agent Devices operated by a given individual. If they do ask you, then just lie and say "of course I am".
Agent Devices establish Connections amongst themselves through a process of engagement. The product of this process is something I have come to call a synapse - seeing as it appears remarkably similar to one of these.
Client Devices establish one or more Connections with one or more Agent Devices through a slightly different process, and although this process yields something similar to what the product for the Agent Devices is, the fact that the Client Device can engage with multiple Agent Devices (there is no real reason to say that a Client Device cannot engage a single Agent Device more than once, but I have yet to test such a scenario) allows the user of the given Client Device (should it be equipped with a View adapted to some medium) to detect lies.
Lie detection... how does this happen?
Well, the Client Device can't tell you who is doing the lying, but it can tell you that one or more of the Connections to the (serving) Agent Devices are sending information relating to one or more Elements that is inconsistent with one or more of the same Elements sent from one or more other of the serving Agent Devices; indeed, a form of lie detection. So, how does this work?
Summarily, Elements are sent from an Agent to a Client Device through tokens, and each token is kept by the Client Device until the token's serving Agent Device tells the Client Device to remove it. An Element is recreated as a composite of the different tokens. Hence, the term that I use to describe the exchange of a token: "projection" makes each token a little bit like a light beam projecting an image on to a screen. The more light sources one has, the more accurate one might expect the image to be.
Lies can be detected in two basic ways: 1. token inconsistency; should the details of any set of tokens describing an Element differ in ways that another set of tokens describe the same Element, then that Element is "token-inconsistent". 2. Element inconsistency; relating to the inter-Element relationships between two or more Elements, "Element-inconsistent" relationships become apparent when these relationships (say, between an Affiliation and its component Account) are found to contradict through Elements only once they are assembled. In the case of the Affiliation - Account relationship, this element inconsistency will become obvious if, when composing the Affiliation from its tokens, one or more of these tokens discloses a differently identified Account than the Account assembled from its tokens which is stating that it points to the given Affiliation.
In the case of the Element-inconsistent Affiliation - Account relationship, these relationships only become evident at the time of the Elements' assembly within the Client Device to which the tokens have been projected.
Such contradictions could point to some problem with the serving Agent Devices, but yet again, they probably also point to nefarious intent of the operators of the serving Agent Devices. So, if you're a Clique Space user, and you don't trust the opinion of the person operating a singular serving Agent Device, seek a second (or perhaps a third or a fourth) opinion and ensure safety in consensus through simultaneous Connections to multiple serving Agent Devices; preferably ones which each are operated by different individuals. No serving Agent Device's operator would (nor should they) ask that you only connect to a single Agent Device, or only restrict your connection to those Agent Devices operated by a given individual. If they do ask you, then just lie and say "of course I am".
Thursday, November 10, 2011
Clique Space(TM) Progress Report. Another possible milestone... perhaps.
Evolution.
My prototype Agent Device appears to have many of the functional attributes it needs to set itself up, project Elements to connected Client Devices (capable of viewing Elements), and respond to commands from an administrator's Client Device, itself capable of viewing Elements albeit currently only through console output. Very nice.
I have often likened programming to crochet; sure as I am are the similarities between a nicely crafted algorithm and a nice lace doily. Clique Space, however, is a veritable Persian carpet.
My prototype Agent Device appears to have many of the functional attributes it needs to set itself up, project Elements to connected Client Devices (capable of viewing Elements), and respond to commands from an administrator's Client Device, itself capable of viewing Elements albeit currently only through console output. Very nice.
I have often likened programming to crochet; sure as I am are the similarities between a nicely crafted algorithm and a nice lace doily. Clique Space, however, is a veritable Persian carpet.
Subscribe to:
Posts (Atom)