Problem. Consider the opening paragraphs on http://gregegan.customer.netspace.net.au/DIASPORA/01/Orphanogenesis.html and particularly this part when you're done:

> Elsewhere, the map recorded a spread of possibilities: a broad
> landmass, or a scattered archipelago. Trait fields offered a
> selection of codes, each with a known effect on the mind's detailed
> structure, with variations ranging from polar extremes of innate
> temperament or aesthetics down to minute differences in neural
> architecture less significant than the creases on a flesher's palm.
> They appeared in shades of green as wildly contrasting or as flatly
> indistinguishable as the traits themselves.

Tell me: if there are known effects, doesn't this render the whole damn map pointless? The map is not the territory, but if you "know" the territory, then why the hell have the map? I can replicate this map's functionality if this issue is resolved. Greg took one too many creative shortcuts here.

In other words: What does the "gene" have to do with the "mind"? I guess what I need is either (1) ways to map the information back into the map, i.e. once you build the brain, how are you going to recover information? For instance: the brain happens to be interesting in some way, great, now how do you record this 'way' and place it on the map? (2) Some way to avoid the necessity of that feedback/mapping. Biology does this by uniting the two packages together (genome/brain; causal relationship of course). Another interesting aspect to note is that the majority of genetic algorithms have top-down enforced meaning of the genes, as opposed to bottom-up meaning ... Melanie Mitchell hasn't heard of this, and neither Nicholas Chaimov, so that might be another interesting avenue to explore here. (I suspect it may be synonymous with 'philanthropy'.)

One idea: you have the code and then the chosen/selected combinations or permutations and variations, and the only thing that can truly represent the functionality is the whole brain itself that was using those variations. And so perhaps it would be interesting to play with a rule saying that once a path is chosen, never again may it be chosen. That brain lives and dies and makes its mark, and that 'mark' upon reality and whatever processes it influenced, is the preserved functionality information. Trying to map the results onto the map itself, in any other way, may turn out to be an impossible task. "Every orphan was an explorer, sent to map uncharted territory. And every orphan was the uncharted territory itself." (isn't it charted if the orphan is truly there, living?)

Here's why: the Google Mouse or Google Earth/Brain interface can be used to somewhat recover this map from our current civilization. The difficulty is in figuring out how to extract information from these people in terms of the 'functionality'. We can most certainly get genomic information in on there, but what the hell else can we get there? I was figuring that it could be made to allow user's "tagging" the information with something more specific than a few words -- perhaps RDF/YAML of some sort. Then the entire territory/map concept can turn into a massive SKDB validation app, or a giant real-time strategy game with interesting cybernetic implications ;-). Probably with the online graphviz interface from the site with 'blood' in the domain name (meh, I forget). What sort of information would be useful for users to put back into the system? The real problem is that anything that a normal user would write down will be something pretty simple, and the interface can't be confusing ... and it's just using natural language anyway, which is somewhat related to grammar and so on, and this grammar is being executed/implemented/dealt-with-by the user's brain itself, which is the whole thing that we're supposedly mapping, tracking, tagging in the first place. And it's not like single traits are going to be immediately isolated ... this is a very massive statistical/correlational-study nightmare. Calling all mentostaticians, I guess.



> The conceptory was non-sentient software, as ancient as Konishi polis
> itself. Its main purpose was to enable the citizens of the polis to
> create offspring: a child of one parent, or two, or twenty — formed
> partly in their own image, partly according to their wishes, and
> partly by chance. Sporadically, though, every teratau or so, the
> conceptory created a citizen with no parents at all.
>
> In Konishi, every home-born citizen was grown from a mind seed, a
> string of instruction codes like a digital genome. The first mind
> seeds had been translated from DNA nine centuries before, when the
> polis founders had invented the Shaper programming language to
> recreate the essential processes of neuroembryology in software. But
> any such translation was necessarily imperfect, glossing over the
> biochemical details in favour of broad, functional equivalence, and
> the full diversity of the flesher genome could not be brought through
> intact. Starting from a diminished trait pool, with the old DNA-based
> maps rendered obsolete, it was crucial for the conceptory to chart
> the consequences of new variations to the mind seed. To eschew all
> change would be to risk stagnation; to embrace it recklessly would be
> to endanger the sanity of every child.
>
> The Konishi mind seed was divided into a billion fields: short
> segments, six bits long, each containing a simple instruction code.
> Sequences of a few dozen instructions comprised shapers: the basic
> subprograms employed during psychogenesis. The effects of untried
> mutations on fifteen million interacting shapers could rarely be
> predicted in advance; in most cases, the only reliable method would
> have been to perform every computation that the altered seed itself
> would have performed … which was no different from going ahead and
> growing the seed, creating the mind, predicting nothing.
>
> The conceptory's accumulated knowledge of its craft took the form of
> a collection of annotated maps of the Konishi mind seed. The
> highest-level maps were elaborate, multi-dimensional structures,
> dwarfing the seed itself by orders of magnitude. But there was one
> simple map which the citizens of Konishi had used to gauge the
> conceptory's progress over the centuries; it showed the billion
> fields as lines of latitude, and the sixty-four possible instruction
> codes as meridians. Any individual seed could be thought of as a path
> which zig-zagged down the map from top to bottom, singling out an
> instruction code for every field along the way.
>
> Where it was known that only one code could lead to successful
> psychogenesis, every route on the map converged on a lone island or a
> narrow isthmus, ochre against ocean blue. These infrastructure fields
> built the basic mental architecture every citizen had in common,
> shaping both the mind's overarching design and the fine details of
> vital subsystems.
>
> Elsewhere, the map recorded a spread of possibilities: a broad
> landmass, or a scattered archipelago. Trait fields offered a
> selection of codes, each with a known effect on the mind's detailed
> structure, with variations ranging from polar extremes of innate
> temperament or aesthetics down to minute differences in neural
> architecture less significant than the creases on a flesher's palm.
> They appeared in shades of green as wildly contrasting or as flatly
> indistinguishable as the traits themselves.
>
> The remaining fields — where no changes to the seed had yet been
> tested, and no predictions could be made — were classified as
> indeterminate. Here, the one tried code, the known landmark, was
> shown as grey against white: a mountain peak protruding through a
> band of clouds which concealed everything to the east or west of it.
> No more detail could be resolved from afar; whatever lay beneath the
> clouds could only be discovered firsthand.
>
> Whenever the conceptory created an orphan, it set all the benignly
> mutable trait fields to valid codes chosen at random, since there
> were no parents to mimic or please. Then it selected a thousand
> indeterminate fields, and treated them in much the same fashion:
> throwing a thousand quantum dice to choose a random path through
> terra incognita. Every orphan was an explorer, sent to map uncharted
> territory.
>
> And every orphan was the uncharted territory itself.