A demonstable prototype: what is really left to do?

My last post gave some of an indication as to how I think I'm travelling with the implementation of this Clique Space(TM) concept. My last post got me thinking about how I can express my progress in this journey, and I have come up with 6 basic criteria that I think can express this progress.

1. Does it work?
   
   In asking this question, I purposely limit scope to the core implementation. It works when I have successfully implemented the concept in a prototype. To this end, I believe I have almost done this; the only component I have to implement here is the Agent Collaboration's pulse message - the mechanism that exchanges information between Agent Devices which are members of a Clique that is used to disseminate knowledge about the operation of a particular device collaboration in some medium. The first device about which knowledge will be disseminated will of course be the Agent Device itself, the first collaboration will be a Clique Space within which a given set of Agent Devices will all be members, and it looks like the first medium will hence be encoded in what has become known as the "collaborator Media Profile".
   
   Actually, the collaborator Media Profile might in some technical sense actually be the second Media Profile if one considers that the "engager Media Profile" has a model that can be divorced from its implementation, which I don't think one can honestly do. However one can observe an Agent Device's local Clique Space as if it were modelling a collaboration between itself and other Agent Devices it has engaged. The thing about the engager Media Profile is that it is part of an individual Agent Device's transport mechanism, and hence doesn't technically model a collaboration, but rather, lists other partner Agent Devices through which the given Agent Devices exchange information through a "synapse". The true media used by Agent Devices to model collaborations as members of a Clique Space is encoded as Enabling Constraints in a Clique Space's collaborator Media Profile.
   
   
2. Does it look pretty?
   
   Probably not currently. This question relates to how "user friendly" the administrator Client Device's View is. Practically, this criterion relates to how easy it is to make others understand what Clique Space is all about. I would like to have a graphical user interface front-end because this will certainly lower the bar of comprehension; a GUI that displays two or more devices collaborating as two or more hexagonal Participant Chips inside a bounding ellipse representing a Clique will certainly help.
   
   However, I do think the console output is quite consistent, and I should, absent the GUI, be able to use a whiteboard to manually draw a View's contents as I am demonstrating it to prospective venture partners.
   
   
3. Is it stable?
   
   No. Putting the Clique Space concept together (putting anything not hitherto done before together) requires one to be very selective in what one does: one really has to put something together that does the first intended purpose before one makes it work across a range of adverse environmental conditions.
   
   Major considerations revolving around stability include how the system deals with devices that behave erratically or inconsistently among the other devices with which they are collaborating. As a special case, what happens when a Clique Space's member Agent Device behaves erratically, or inconsistently, or abruptly goes off-line? As a special case of this, what is done when a Clique Space's owner is lost? Another consideration about the stability of the Agent Collaboration will be how to disseminate knowledge about the collaboration's state to all members when those members could number in the hundreds of thousands or more. I have some ideas along these lines, but their implementation will have to wait until I 1: get it to work.
   
   
4. Is it secure?
   
   No. However, I have found that while I was implementing things like the Agent Device's engager semantics, at least some of these considerations perhaps dissolved. When two Agent Devices engage, they exchange information about each other's local Agent Device Clique Space, and this mechanism can guarantee confidentiality provided, of course, either or both Agent Devices do not leak this information to other parties of adversarial intent.
   
   In terms of the collaborator Media Profile, whatever standard encryption and other security mechanisms are available can be fitted as time, and needs dictate. Yet other Media Profiles can in turn encode information about other, more or less powerful collaboration security, which at the very least, should be a clear demonstration of how flexible Clique Space is as an identity and collaboration metasystem... now there's a good term.
   
   
5. Does it really work?
   
   No. In this question, I'm asking myself whether it works beyond its core implementation. Clique Space models and can (or rather, should be able to) control any existing or future collaborative device provided any particular device type can, at least, collaborate (with the purpose of exchanging state and control information) with a Clique Space Agent Device. Clique Space is completely useless if it isn't used along with other devices. Hence, at this moment, Clique Space is completely useless.
   
   Truly, anything that can exchange state information with another thing is a candidate device. Should any type of thing in fact be capable of exchanging sufficient information to the extent that it can be modelled and controlled by a Clique Space system, then indeed, a Media Profile can be developed for that type of thing(device, as earlier described) so that any instance of it can be expressed along with the individual who has nominated themselves as the controller as well as any operational parameters and privileges associated with the thing instance and the individual who is controlling it.
   
   A Clique Space that really works will work with phones, PC's, cars, flippers, fridges, email clients, email servers, Facebook accounts, space shuttles (or their successors), teleporters, Second Life avatars, kidneys, brains (possibly as Clique Spaces in their own right), and bank accounts, etc. Currently, Clique Space only knows of the Agent Device and the administrator Client Device.
   
   
6. Is it useful?
   
   Maybe. This is a question of value that I alone cannot answer. The answer is provided by you and sufficient others who would find a use for it in their lives. The above criteria will be necessary but insufficient for you to decide that it is useful. Maybe you don't like me and hence you may boycott Clique Space.
   
   I hope you don't hate me, but if you do, I hope you still find Clique Space useful. Let me know if I can do anything to help you make up your mind about me, Clique Space, or both me and Clique Space. I would welcome your feedback and the chance you might provide for me to address questions you have about me, about Clique Space, or about me and Clique Space.