> Personal Digital Habitat is a metaphor that c...
# linking-together
m
Personal Digital Habitat is a metaphor that can help us imagine how to unify all our computer-based personal devices and simplify our digital lives.
http://www.wirfs-brock.com/allen/posts/1033
❤️ 2
s
While I agree with the aspiration, I can't agree with the conception. I don't think the metaphor of living in a "digital habitat" accurately describes the relationship people have with personal technology. Especially with phones, smartwatches, a closer metaphor is (expensive and highly defective) prosthetic organs. This difference might seem academic, and it kind of is—from a reductionist standpoint there is no essential difference between a biological entity and it's environment aside from scale. But I find profoundly impacts what things we think are acceptable, especially with regards to things like ({bodily, digital}) autonomy. It's hard to conceive of it this way at first, because we are used to thinking in standard terms. i.e what is my "digital stuff", or "data". I think a good argument could be made for them really being a new memory organ, but the metaphor seems to fall apart because we normally think of our memory as somethlng coherent—because it normally is. A person without technology is a distributed system with one node, and that is what we are used to thinking of—all our conceptions of what it means be are in terms of that special case. However, people are fundamentally distributed systems. Memory is not something coherent, even if we consider basic technology like pen and paper. Our technological stack is so dysfunctional that we are a distributed system that is perpetually stuck in a netsplit, because we don't have the persistent data transfer and such that the author of the article is advocating for. Furthermore, only one node (the carbon-based one) is the single source of truth of the reliability of the system. No wonder we can't see it for what it is—no netsplitted node can a priori know that it is netsplitted!
🤔 1
If we take this metaphor a little bit further, it's natural to ask what kind of system we are talking about. In the case of distributed systems, there are classic criteria—CAP. It's clear that we only have the A as it is, at best. The vision for the future outlined in the article can be rephrased in these terms to mean that we want the P as well. Which is good, as the joke in distributed systems goes, you can make tradeoffs between A and C, but you always have to pick P. The indsutty has more or less converged on AP and eventual consistency, and that is probably what we want in this case too.
m
How would that look like concretely? how would you name it?
s
I find this vision very compelling. We have a poor, siloed version of this now, where various things I create, use and share live across my devices in various modes of sync and a large number of disjoint namespaces. Eg.I have some documents in Google drive (its own namespace, own sync model etc.) There's some code in git, synced manually via github and mapped onto my local filesystem namespace as well as github URLs. If we think of all such artifacts as "digital objects" it would be nice to have them in one virtual space, where different devices I own are just portals into the same space. There would be a unified log of changes, if one exists. Further even "apps" and tools or mini-apps would be other objects that I import into this space. This is how I perceive the concept. Technically, doing syncs is an interesting issues. Eg. if one device lost connectivity, it will only have information about the locally cached objects. Changes will eventually have to be merged. This might require automated and manual merging mechanisms and have to introduce ideas like "pinning" - forcing an object to always be present in one portal.
🤔 1