>>I think the paradigm we may be looking for is that there are still unused pathways where the new mind will go... so, something like "installing the same app into a different OS still needs disk space".
>
>I see where you're going with this now. You want the upload to actually forge new neural paths. Does the uploading mind have that information, or does the upload software handle it?

Now we're writing software :). Depends on what you mean by "does" - does where, in our designs, or in popular lore (as set forth by all the SF authors so far)? Most authors came from the body/mind split POV, so they treated the new body just like a new car, admitting that it did change the attitude of their character (specially if it was of a different sex). Others thought it would be more complicated, notably Egan, also Charles Stross and, why not, Pratchett. Frank Herbert had a somewhat different idea with his ghola/clone story, that the transfer is only possible genetically, i.e. the incoming mind has to live in pretty much the same genome until it can awaken. OTOH, he had the Other Memories line, also genetic, but not doing a complete takeover, just a presence.

>I would expect more that the upload simply uploads data (ie memories, etc) into the receptacle brain, but that new paths would only be forged once the uploaded mind initialises and begins to run. Of course, any new paths forged would have to be intimately tied in to the concepts and interpretations of the already existing paths.

But what is data and what is code here? Do we really know where's the difference, and is there any? Even two AIs, or two expert systems, would grow to be completely different if fed different data, no matter that they initially ran the same code.

I would expect that the upload be impossible without at least some rewriting of the pathways - minimal, sufficient, complete, total - who knows. There may be different packages, 1st class transfer, economy transfer, luxury transfer... there's a market :).

>>Or, in my case, a transfer would assume writing its own pathways (to some extent) or else we get translation loss.
>
>Interesting concept - that the uploading software would be able to forge new pathways in the receiving brain such that those pathways would exactly mirror the pathways of the uploading brain. And that those new pathways would exacly mirror the attitudes and learned biases etc of the uploading pathways. Well, I think personally, you are going to get translation loss.

You're getting translation loss when you wake up and forget your dreams, so what? It's only a matter of how much of a loss, and can we say it's the same person or not. But then, we already have ways to change a person beyond recognition - drugs, emotional shock, brainwashing (and tumble drying :) etc etc. We'd need a usable definition of "same" before we start writing software for this.

>I think it is probably inevitable. Are the already existing pathways in the receiving brain going to disappear at the same time as the new ones are being forged?

Another good question. How many unused ones are available, and if not enough, how many old ones are reusable (gladly) and how many old ones are reusable (grudgingly)?

>>In case of the adjective, right - it's the noun where I'm sticking with the philosophical meaning.
>
>When you say "metaphysical bookstore", how can 'metaphysical' be interpreted as a noun? The noun is 'metaphysics'. 'Metaphysical' is always an adjective as far as I can tell.

I started the lexicographical branch of this with a link to the noun on Wikipedia - unfortunately, its paragraph about (mal)appropriated meanings used the adjective, which is where we lost our bearings. I meant metaphysics as philosophical discipline regarding the ontology, gnoseology etc etc.

>>Never mind the score, we didn't set any rules yet :).
>
>Ok. In that case, I'm waaaaaayyy ahead. ;)

So am I. Now we only need to see whether we're going in the same direction :).