Nine requirements 8 of 9
Security
A larger question than a product, and one Jakob has started answering somewhere else. This page states the question and points there.
Compiled 4 September 2026.
Physical safety and the rule of law. Policing, defence and diplomacy that protect people rather than the arrangement that currently holds power.
This page deliberately does not describe a security product, because there should not be one. Security is the single requirement on the matrix where all three judging modes fail: nothing here can be lent, there is no shelf to grade, and nothing material is reused. A requirement with no consumer surface is one where inventing a consumer surface is the error, not the contribution.
It is also the requirement most easily sold back to people as a thing to buy. Alarms, cameras, private guards and subscriptions all address the feeling while leaving the arrangement untouched. The discipline this page keeps is to state the question and leave it stated.
Jakob's position, in his own words
For security it is a larger question, that I have started answering in Europe of the Regions.
How this page reads that
Two claims, and they are different. The first is that security is not a product question. The second is that an answer is already under way, and that it is a political one, worked at the level of how Europe is arranged rather than at the level of what a household buys.
So the job of this page is to point, accurately, at where that work sits, and to be precise about how far it currently goes. Overstating it would be the same failure as inventing a product.
The nearest note in the 9EPhR material
There is no node named security under the 9EPhR root. The closest is one named defence, and it is two lines long. It is quoted because its brevity is consistent with the position above.
- It is political.
- Again, choice of location.
Tana 2uX9xvsT8IHk, read 4 September 2026.
Which of the four modes apply
Read from the matrix on the nine requirements page, not restated. The wording below is the wording in the grid.
Sharing somebody else made it
does not apply
Security is not an object that can be lent. What looks like shared security is either a public institution or a private contract, and neither is sharing in the sense every other row uses the word.
Ideal consumption somebody else made it
does not apply
There is no shelf. Policing, courts, defence and diplomacy are not bought by an individual. The lever is political rather than consumer, and pretending otherwise is how security gets sold back to people as a product.
Reuse somebody else made it
does not apply
Nothing material is reused. The nearest analogue, keeping institutions rather than founding new ones, is an argument about governance and not about consumption.
Cosmolocal production you make it
The only mode that applies here, which is the whole argument in one row. A community can build its own trust infrastructure: standing vouched for by people who know you, over protocols whose specification is public and whose instances run locally.
Element, Sifted [B]Signal [A]
The question, stated
The requirement as the matrix names it is physical safety and the rule of law: policing, defence and diplomacy that protect people rather than the arrangement that currently holds power. The last clause is the whole difficulty. Every security institution is also a power arrangement, and the question of whether it protects people or protects itself is settled politically, at a particular scale, by whoever gets a say.
Which makes the real question one about scale and legitimacy. At what level should the capacity for legitimate force sit, who authorises it, and how do the people it acts upon hold it to account. That is a constitutional question long before it is a technology question.
Where the started answer sits
Europe of the Regions is Jakob's work on exactly that layer. Published today it is an interactive map and dataset of 191 practical European regions, built on Eurostat NUTS 2024 to 2027, with comparisons of representation models, a self-sufficiency view across several dimensions, overlays for congruence and contested zones, and a methodology page carrying its own tiered sources.
Its stated purpose is to make regions rather than nation states the load-bearing unit: to defuse Catalonia and Madrid style tensions, to even out how much prominence different members get, and to describe an organic path as nations weaken relative to Europe on one side and to regions and municipalities on the other.
- Europe of the Regions live The project itself, and the landing page that states what it is arguing for.
- The regions map live 191 practical regions, with the congruence toggle, contested zone overlays and the self-sufficiency view.
- The table and the representation models live All 191 rows, sortable, next to the comparison of how seats would be allocated under different models.
- Methodology and sources live Where the pushback on the region definitions is argued out, with tiered sources.
How far that actually goes, stated precisely
As published today, Europe of the Regions is about political geography and representation. It argues which units should exist and how they should be represented. It does not yet address policing, courts, defence or diplomacy directly, and no material stating a position on those was found in this pass.
So the accurate summary is this. Security is a question about political order and scale. Jakob has started answering the question of political order and scale. He has not yet published an answer about security itself, and the link between the two is a reasonable expectation rather than a completed argument. Confidence in that reading is about 0.75, and it is a reading rather than a finding.
The one piece that does exist, and its limits
The matrix puts the Web of Trust in the production cell for this requirement, because a community can build its own trust infrastructure over protocols whose specification is public and whose instances run locally. That is real, and it is narrow.
It addresses trust between people. It is not physical safety and it is not the rule of law, which is what this requirement actually names. The matrix already says so and this page repeats it rather than letting the link imply more.
- Web of Trust live Trust between people, represented digitally. A narrow slice of this requirement and not a security system.
Where this leaves it
No security product here, and none should be built. The page exists so that the requirement is not silently empty on the matrix and so that anyone following the question lands on the work that is actually under way.
The open item is the honest one: connecting a regional constitutional argument to an answer about policing, courts and defence has not been done yet, and until it is, this requirement has a direction rather than an answer.
Recorded as missing on the matrix Nothing here is about physical safety or the rule of law, which is what the requirement actually names. The web of trust addresses a narrow slice, trust between people, and should not be presented as more.
Sources
Tier A is a legal, governmental or scientific primary source, B is journalism or trade press, C is a community, project or company describing itself. Confidence below 0.7 is flagged and carries a stated reason. Where a figure could not be established it is named as unestablished rather than replaced with a plausible one.
| Claim | Publisher | Date | Tier | Confidence |
|---|---|---|---|---|
| Europe of the Regions: an interactive map and dataset of 191 practical European regions built on Eurostat NUTS 2024 to 2027, with representation model comparisons and a methodology page | meta-democracy.org, Jakob Possert-Bienzle | v0.2.2, accessed 4 September 2026 | [C] | 0.85 This project describing its own work. Cited here as the location of a position, not as independent evidence for it. |
| Germany runs the Matrix-based BwMessenger for the Bundeswehr, live since November 2020, and France runs Tchap on the same protocol | Element, Sifted | accessed 4 September 2026 | [B] | 0.75 Vendor case study and tech press. No government primary source was reached. |
| Signal core protocol library libsignal is licensed AGPL-3.0-only | Signal | accessed 4 September 2026 | [A] | 0.9 |