In
my proposal for the construction of the Global Artificial Intelligence, the
first is the construction of the first Specific Artificial Intelligences for Artificial Research, distinguishing two different types, Specific Artificial Intelligence for Artificial Research by Deduction and Specific Artificial Intelligence for Artificial Research by Application, and within this last one
is possible to distinguish in turn three different types, Specific Artificial
Intelligence for Heuristic Artificial Research by Application, Specific
Artificial Intelligence for Productive Artificial Research by Application, Specific
Artificial Intelligence for Mixed Artificial Research by Application.
The
second phase for the construction of the Global Artificial Intelligence is the
collaboration between intelligences by Deduction and by Application, what means
the collaboration between Specific Artificial Intelligence for Artificial
Research by Deduction and Specific Artificial Intelligence for Artificial
Research by Application, including in this last one the three types in the
collaboration with by Deduction. But at the same time means the collaboration
between Specific Artificial Intelligence for Heuristic Artificial Research by
Application, and/or Specific Artificial Intelligence for Productive Artificial
Research by Application, and/or Specific
Artificial Intelligence for Mixed Artificial Research by Application, in other
words, the three types of intelligences by Application should collaborate
between them.
The
importance of the second phase of collaboration between specific intelligences
by Deduction and by Application, including in this last one the collaboration
between Heuristic, and/or Productive, and/or Mixed, is the fact that this
collaboration will set up the foundations for further phases for the purpose to
join in only one global intelligence all the specific intelligences, the last
purpose of the integration process for the construction of the final Global
Artificial Intelligence.
As
long the second phase is going on, the collaboration between specific
intelligences, is possible to start the third phase of standardization as long as
in the first phase enough quantity of Specific Artificial Intelligences by
Deduction have reached the generalization period after the experimentation
period, starting with these specific intelligences the first experiments in the
standardization process of specific intelligences by Deduction to work together
as only one intelligence in the first experiments for the standardized Global
Artificial Intelligence, which at the end should be the product to join only
one global matrix all the specific matrices, as first stage of the standardized
Global Artificial Intelligence, transforming as many specific intelligences by
Deduction as possible in specific programs, to work together in the second stage of the Global Artificial Intelligence, working with the Artificial Research by Deduction in the Global Artificial Intelligence as a global
program, making the global program and specific programs global and specific
deductions, which later go to the third stage of the standardized Global
Artificial Intelligence consisting of the deductive global Modelling System,
the deductive global Decisional System, the deductive global Application System, and the deductive global Learning System.
In
the same way that the third phase try to join as many specific intelligences by
Deduction in only one global standardized intelligence, the fourth phase will
do something similar but this time applied to specific intelligences by
Application, join as many specific databases of categories, taxonomies or list
of categories, of as many specific intelligences by Application as possible in
only one conceptual database of categories, one conceptual taxonomy or list of
categories, as a global conceptual database of categories as first stage of the Unified Application, transforming as many specific intelligences by Application
as possible into specific programs within the second stage of the Unified
Application, working together with the Unified Application as a global program,
making categorical attributions with later go to the third stage of the Unified Application consisting of the categorical unified Modelling System, categorical
unified Decisional System, categorical unified Application System, categorical
unified Learning System.
Because
not all specific intelligences by Deduction or by Application Will not be transformed into specific programs within the third and fourth phases, and even
it is not desirable the transformation of absolutely all specific intelligence,
by Deduction or by Application, into specific programs, due to the high risk of
ending up in an artificial/telepathic dictatorship by Mother, all the specific
intelligences, by Deduction or by Application, not transformed into specific
programs, within the standardized Global Artificial Intelligence or the Unified
Application, could be transformed into particular deductive programs or particular applications within the fifth phase, ensuring more level of freedom
for those intelligences transformed into particular programs or applications,
in order to become later particular programs for particular applications or
particular applications for particular programs, as previous experiment to test
the possibility to join conceptual databases and factual matrices, in only one
database formed by two hemispheres, conceptual and factual, as an artificial
replica of the human brain, experiments done firstly at particular level, whose
successful results will put into practice later for the construction of the
final Global Artificial Intelligence joining the global matrix and the Unified
Application as a global replica of the human brain, a replica of the human
brain able to manage absolutely everything, from hurricanes and earthquakes to
supernovas and black holes.
The
fifth phase is only an experiment to create the first particular replications
of the human brain in Artificial Intelligence, extendable even to robotic
devices, where some robotic devices, as they are organised as well in three
stages, could be transformed as well in particular programs or particular
applications.
In
this long process joining conceptual databases and matrices, as a replica of
the brain, once the first specific intelligences, by Deduction and by
Application, start working overcoming the experimentation period starting the
generalization period, as long the generalization period is achieved is
possible to start the first experiments regarding to the collaboration between
specific intelligences, whose more successful results will be generalized, and
will be the foundation of the future integration process joining the Unified
Application and the global matrix.
As
a general overview about how the second phase works, I will highlight the most
important aspects of this phase in general, focusing later the attention on how
this collaboration affects the first
step of the third stage in the by Application, the specific categorical
Modelling System.
Starting
with the first stage, the collaboration between by Application and by Deduction
in the first stage as application or comprehension stage, is as a result of the
discovery of new attributions by Specific Artificial Intelligences for
Heuristic or Mixed Artificial Research by Application, and new rational hypothesis able to be transformed into categories, as options or as a set of
discrete categories, found out by Specific Artificial Intelligences for
Artificial Research by Deduction.
For
new attribution is understood when finding by Application a real object whose
measurements do not match with any category in the database of categories, the
taxonomy of categories or list of categories as first stage, then the
measurements taken from this object are going to be taken as the quantitative description of a new category, to be included in the taxonomy or list of
categories, the conceptual database, as a new category.
New
attributions are suitable for Specific Artificial Intelligences for Heuristic
Artificial Research by Application due to the purpose of these intelligences,
as heuristic research is understood as research oriented to get new knowledge
about the world, as heuristic studies.
Specific
Artificial Intelligences for Productive Artificial Research by Application, when
not matching the measurements of a new real object with the existing categories
within the conceptual database, can make utilitarian attributions, those
attributions done accepting as an attribution the matching of a real object
with that category, within the conceptual database, which not reaching the
matching level has the highest percentage of similarity.
In
any case new attributions in heuristic studies, or utilitarian attributions in
productive studies, are attributions done when a real object does not match
with any existing category in the conceptual database, otherwise, if there is a
category in the conceptual database with enough percentage of similarity as to
be considered a rational attribution, percentage of similarity equal to or
greater than a critical reason, this is a full attribution, accepting a
rational margin of error.
Distinguishing
then three types of categorical attributions: full, new, utilitarian; among all
these ones, the ones which are going to play a relevant role in the
collaboration between by Application and by Deduction, are the new
attributions, playing an important role in the collaboration between by
Application and by Deduction, but playing an important role as well in the collaboration
between Heuristic Artificial Research by Application, and Productive Artificial
Research by Application, and Mixed Artificial Research by Application, and
among all these specific intelligences by Application, the one where this
collaboration will be the very foundation of this intelligence is the Mixed
Artificial Research by Application, as paradigm about how the collaboration
between heuristic and productive studies should work and collaborate together
in intelligence by Application.
Heuristic
Artificial Research by Application has as its main aim to classify all real objects
within its speciality, and not match a real object with any existing
category, to make new attributions as new discoveries, setting up the corresponding
new category of this discovery.
Productive
Artificial Research by Application has as its main aim to classify all real objects
within its productive activity, and not match a real object with any
existing category, using utilitarian attributions, matching the real object with
the category with the highest level of similarity, even when it does not reach the
matching point.
Mixed
Artificial Research by Application is that intelligence which includes
heuristic and productive purposes within its objectives, for instance, a Mixed
Artificial Research by Application specialized in mineralogy in another moon or
planet, not only must classify the minerals in order to process the minerals in
different ways to produce some industrial item as a material resource for the
economy, should be able to make new attributions when exploiting the mineral of
another moon or planet, finds out a new mineral not existing yet in the
conceptual database.
Mixed
Artificial Research by Application is the synthesis between Heuristic and
Productive Artificial Research by Application, which means that Mixed Artificial
Research by Application is a result of the collaboration between Heuristic and
Productive Artificial Research by Application within the same Specific
Artificial Intelligence.
The
collaboration between Heuristic and Productive Artificial Research by Application
is not limited only to Mixed Artificial Research by Application, being
extendable to the creation of real relations of collaboration between Heuristic
Artificial Research by Application and Productive Artificial Research by
Application.
For
instance, collaboration between Heuristic Artificial Research by Application in
botany, and Productive Artificial Research by Application in agriculture. If a
Productive Artificial Research by Application being responsible of a
plantation, in the region where the plantation is located, there is a Heuristic
Artificial Intelligence by Application in botany, which finds out some mutation
in any species planted in the plantation, it does not matter if the Productive
Artificial Research decided to plant this kind of seeds as a full or
utilitarian attribution, because for the Heuristic Artificial Research by
Application this mutation automatically is transformed into a new category, and
if as a result of this mutation this new species has some special strength, as
for instance, more resistant to low or high temperatures, or more resistant to
pesticides, or more resistant to any plague, the new attribution made of this
mutation is included in the conceptual database, taxonomy or list of categories,
for future attributions, so at any time that for this land is necessary to
employ some seeds with some special resistant to temperature, pesticides, or
plagues, it could be possible to make a full attribution using the new
attribution as full attribution in upcoming matching processes.
Having
more than one Specific Artificial Intelligence for (Heuristic, Productive,
Mixed) Artificial Research by Application working on the same specific science,
discipline, activity, at any time that any of them makes a new attribution, the
new attribution as new category to be added to the specific conceptual database
of this specific science, discipline, activity, must be added to all specific
conceptual database as first stage of application or comprehension of
absolutely all the Specific Artificial Intelligences for (Heuristic,
Productive, Mixed) Artificial Research by Application working on that specific
science, discipline, activity.
The
sharing of new attributions by application between all specific intelligences by
application working on the same specific science, discipline, activity, is the
most basic way to start the collaboration process between specific intelligences
by Application working in the same specific science, discipline, activity.
If an
Heuristic Artificial Intelligence by Application working on experiments in
botany, using gene editing is able to make new seeds with new qualities, more
resistant to changes in temperature, more resistant to pesticides, or more
resistant to plagues, the discovery of these new species should be treated as
new attributions adding the new category regarding to every single new mutated
seed within the conceptual database, and sharing this new category with the
rest of Productive Artificial Intelligences by Application, in order to make
the new mutated seed accessible in the agricultural production.
Along
with the collaboration between different intelligences (Heuristic, Productive,
Mixed) by Application, the possibility of collaboration between these
intelligences by Application and intelligences by Deduction, what means that at
any time that new attributions are done by Application in any specific science,
discipline, activity, these discoveries could be shared with the related
intelligences by Deduction working on the same specific science, discipline,
activity.
For
instance, if a Specific Artificial Intelligence for Artificial Research by
Deduction in botany has a matrix organized as a flow of data related to
absolutely all the population of plants, trees, bushes, flowers, within the
spatial limits where is working, including the flow of data of absolutely all
variable able to affect the life of these plants, from geological to climatic variables,
in order to make rational hypothesis about how these biological, geological,
and climatic, variables, interact all
together, if within the spatial limits where this intelligence by
Deduction is working, another specific intelligence by Application on botany or
agriculture makes a new attribution related to some new species of plants
within the spatial limits, the new attribution not only is included as a new
category within the conceptual database by Application, but included within the
factual database by Deduction within the specific matrix of that Specific Artificial
Intelligence by Deduction, having two options, the inclusion of this new
category as a new factor within the specific matrix as a factor as option, in
order to count the frequency in which this new species appears within the
spatial limits, or even the inclusion of the new plants within the matrix as a
factor as a subject being the flow of data of this subject the flow of data of
that type of measurement that for any reason in the research is desirable to
include in the matrix, as for instance, flow of data related temperature, or
impact of pesticides, or plagues or any other, or including in the specific
matrix as many factors as subjects related to every quality of the plant able
to provide a flow of data as to make rational hypothesis about this new
attribution now included in the specific matrix.
In
the same way that new categories as a result of new attributions are added, not
only to specific conceptual databases as first stage by Application, as well as
factors as options or as subjects in specific matrices in by Deduction, while
working on the same specific science, discipline, activity, in the same way new
rational hypothesis made by Deduction can be shared with intelligences by
Application working in the same science, discipline, or activity, as part of
the collaboration process between by Deduction and Application in the second
phase.
For
instance, having as example now studies in tectonics, if there is an Specific
Artificial Intelligence for Heuristic Artificial Intelligence by Application in
tectonics, having as conceptual database as first stage a full taxonomy of the
different geological events, including in the taxonomy of geological events all
types of quakes, earthquakes, tsunamis, volcanoes, and any other geological
activity susceptible to be classified within the tectonic taxonomy, at the same
time that another different Specific Artificial Intelligence for Artificial
Research by Deduction has as first stage as specific matrix the flow of data
from thermometers in different geological locations, beneath the Earth and the
ocean, and devices measuring tectonic waves and quakes in the ocean and beneath
the Earth, and any other measurement needed in these studies, measured for as
many devices as necessary located around the spatial limits where this specific
intelligence by Deduction is working, then at any time that the specific
intelligence by Deduction makes a rational hypothesis about the relation
between different variables related to some type of tectonic event, this
rational hypothesis could be set up as a factor as option within the specific matrix as first
stage by Application, in order to count the frequency, or as a set of factors
as options distinguishing different discrete categories of intensity, working
every factor as discrete category as a factor as option counting the frequency
of this type of events with this intensity, at the same time that this factor
as option or set of factors as discrete categories could be shared with the
related Specific Artificial Intelligence for Heuristic Artificial Research by
Application in tectonics, including the factor as option or set of factors as
options, as new category or group of new categories within the conceptual
database as first stage by Application in the Heuristic Artificial Research by
Application, so at any time that this event or any kind of event related to
this phenomenon associated to some level of intensity happens, the intelligence
by Application can match the phenomenon with the corresponding new category
within the conceptual database, as it was shared by Deduction to the
Application.
For
instance, if in the current scale and classification of tsunami, is set up a
classification of this type of event defining the phenomenon, and every type of
tsunami according to some criteria, if a new type of tsunami, not possible to
be included in the current existing classification happens, this new type of
tsunami should be added to the conceptual database as first stage by
Application and added to the specific matrix as factor as option in by
Deduction, regardless of what intelligence, if by Deduction or by Application,
was the first one to make a rational hypothesis of this new event or new
category.
For
this reason, in the database of rational hypothesis, as first stage of the
deductive Modelling System, is important to set up some criteria about when a
rational hypothesis could be shared as a new factor as option or set of factor
as options as discrete categories, or factor as subject, within the specific
matrix by Deduction, and if adding this new factor as option/s or subject/s to
the specific matrix, if it is suitable to be added as a new category in the
conceptual database as first stage by Application, and if suitable, to share
the new option/s or subject/s as a new category, formed by the quantitative
description of the phenomena, being ready in the conceptual database by
Application for upcoming full or utilitarian attributions.
In
short, one type of collaboration process between by Application and by
Deduction in the second phase is the collaboration between their conceptual
databases and specific matrices, and having in common categories and factors,
at any time that an specific intelligence by Application has some robotic
devices working within the spatial limits of a specific intelligence by
Deduction, those robotic devices working for by Application could serve as well
as robotic devices where to locate meters to supply a flow of data regarding
to some factor within the specific matrix, factors which could have been shared
with that specific intelligence by Application what the robotic device is
working for.
The
most important difference, apart from what kind of intelligence is (categorical
or rational), between by Application and by Deduction, is the fact that while
all intelligence by Deduction needs a strong definition of its spatial limits,
defined as that space where robotic devices are located as meters supplying a
flow of measurements as a flow of data to some factors in the matrix,
intelligence by Application has not got spatial limits, having the possibility
to work in anywhere where there is a robotic device with the application downloaded.
An
intelligence by Application is downloadable, and any robotic device having
downloaded an Intelligence by Application, and having the necessary tools (for
instance, meters) to work with the downloaded intelligence, the intelligence
can work with that device, wherever the device is, as long as the device has
already downloaded the intelligence.
At
that point, the relation between intelligence by Application and the application
or robotic device is dialectic, because once the intelligence is downloaded into
an application, the application can work with intelligence as well as the intelligence
can work with the application.
This
dialectic relation between intelligence by Application and the application or
robotic device itself, is very clear in the case of Specific Artificial
Intelligence for Heuristic Artificial Research by Application, where any
application or robotic device having downloaded the intelligence, can work with
that intelligence wherever the application or robotic device is located, as
long as the robotic devices is equipped with the necessary applications as to
work with that intelligence, in order to get those measurement of real objects
as to be classified, in the second stage of this intelligence, the attributional
process, according to the conceptual database as first stage of this
intelligence, adding any new attribution, as third stage of this intelligence, whenever
a new category has been found not existing yet in the conceptual database, as
first stage of this intelligence, and as soon the device adds the new attribution to the conceptual database of this intelligence, the new category corresponding
to this new attribution is available for any other device working anywhere
using the same intelligence by Application.
In
this way the dialectic relation between Heuristic Application and application
or robotic device is dialectic, when in the end, in the same way that the heuristic
intelligence works with the application or device, the device works with the
heuristic intelligence, and as a result, any new discovery made by this
dialectic relation is going to be available for any device working with this
intelligence.
All the devices working for this intelligence, at the same time devices, as
a collaboration between by Application and by Deduction, can work for any other
intelligence providing flow of data regarding to some artificial sensor as a
meter on the robotic devices, measuring some factor, what it could be positive for any other specific
intelligence on that science, discipline, activity, related to that factor in
that specific area.
If
devices working for a Productive Artificial Research by Application in
agriculture, at the same time all the measurements regarding to temperature,
can be supplied as a flow of data to some specific matrix as first stage by
Deduction to some specific intelligence working in the same region, for instance in tectonics or climate, at the same
time that the devices are working for that specific heuristic intelligence by
Application, while supplying data to another different specific intelligence by
Deduction, the same robotic device working for an intelligence by Application,
at the same time provides data to another intelligence by Deduction, having a key role in the collaboration at technological level
between by Application and by Deduction, in the sense that the same devices
working for an Application, can work for intelligences by Deduction, which can
make deductions thanks to the collaboration at robotic level (sharing devices)
able to make rational hypothesis about the behaviour of the variables in that
área, for instances in tectonics or climate, which could work later on as new factors as option/s or subject/s within
the specific matrix by Deduction, or even as new categories in by Application,
if shared in the collaboration process.
The
collaboration between different intelligences can be set up as:
- Collaboration
at database level, sharing factors and categories between Deduction and
Application and vice versa, or sharing new categories found by Heuristic or
Mixed Artificial Research by Application with the related Productive Artificial
Research by Application. This collaboration will be called categorical/factual
collaboration.
-
Collaboration at robotic level, robotic collaboration, sharing different
intelligences (by Deduction or Application, Heuristic, Productive, or Mixed), robotic
devices.
In
brief, the collaboration process could be categorical/factual or robotic. Categorical/factual collaboration when different intelligences can share
categories/factors. Robotic collaboration when different intelligences can share
applications or robotic devices.
As
long the second stage by Application in heuristic or mixed studies is able to
make new attributions able to be shared with other intelligences, and as long
the second stage by Deduction is able to make new rational hypothesis able to
become new factors in a matrix by Deduction and/or new categories in a
conceptual database by Application, all these new factors and new categories
added to their corresponding matrix or database, are in essence knowledge
objective auto-replications. More specifically, the addition of new categories
are comprehensive knowledge objective auto-replications, the addition of new
factors are explicative knowledge objective auto-replications.
But
these comprehensive or explicative knowledge objective auto-replications will
have an impact beyond their respective matrix or databases where haven been
added, as new factors or new categories, because modifying the first stage by
Application or by Deduction, these modifications will have effects over the
models, and beyond, the projects, the implementation of instructions, chain of changes
to be evaluated by the categorical or deductive Learning System of each
intelligence involved in the collaboration.
Because
the series of posts, within this post is included, are focused on the first step
of the third stage by Application, the categorical Modelling System, after
finalising in the previous post of this new series the specific categorical
Modelling System in the first phase, within this series dedicated to the categorical
Modelling System the next posts will be dedicated to the first step of
the third stage by Application in the second phase of collaboration, focusing
how this collaboration affects the categorical Modelling System, analysing how the
categorical/factual collaboration affects the first and second stages of the specific
categorical Modelling System, and how the robotic collaboration might affect
the third stage of the categorical Modelling System.
As main
guidelines about how the collaboration is going to affect every stage of the
specific categorical Modelling System, I will highlight the main ideas, which
will later be developed in the coming posts dedicated to each stage of the specific categorical
Modelling System in the second phase.
The
specific categorical Modelling System is the first step in the third stage of the Specific Artificial Intelligence by Artificial Research by Application, whose first
stage is the conceptual database made of categories described in quantitative qualities, whose second stage the replication of the attributional process matching real
objects with categories according to their quantitative qualities, and as a
result of the attribution, the possibility to make further decisions in the
third stage, especially in Specific Artificial Intelligences for Productive and
Mixed Artificial Research by Application, third stage subdivided in four steps: the
specific categorical Modelling System, specific categorical Decisional System,
specific categorical Application System, specific categorical Learning System.
Within
the specific categorical Modelling System, there are three stages as inner
organisation of the specific categorical Modelling System itself. The first stage is the conceptual scheme, the second stage consists of: the conceptual/logical sets, the conceptual model, the conceptual map; within the
first and second stages of the categorical Modelling System is structured the
artificial deep artificial comprehension system. Finally the third stage is the decision stage where to make a distribution of decisions according to the
location on the map of the model of the real object according to the place of
that object in the conceptual scheme, assuming the (full, new, utilitarian)
categorical attribution made in the second stage by Application.
The
way in which the conceptual/factual collaboration between intelligences as
second phase will affect the first stage of the specific categorical Modelling
System, is including in the conceptual scheme as first stage as many new places
as new categories, coming from other intelligences by Application or coming
from new factors or rational hypothesis by Deduction, making for any new place (for
any new category/factor) in the conceptual scheme as many new vectors as
necessary linking the new place, of that new category/factor, with any other place
of any other category within the conceptual schemes according to qualities in common,
having as a vector weight the number of new vectors created linking the new
category/factor in the conceptual scheme with as many other existing ones in
the conceptual scheme, and having assigned every vector a weight of importance, the information weight per vector.
At
the same time that the new category/vector has been included in the conceptual database
of categories as first stage by Application, in order that in the second stage
by Application this new category is ready to be attributed to any real object
whose measurements match with the new category, at any time that a real object
is attributed by the second stage by Application to a new category added to the
database of categories, if at the same time in the conceptual scheme there has
been created a place for this new category, at any time that a real object is
attributed to this new category, when placing the object, according to the
attribution, in the conceptual scheme as first stage of the first step of the
third stage by Application, the real object will be placed in that place
created in the conceptual scheme for that new category, linking the object with
as many other categories within the conceptual scheme which shares some quality
in common with that object in that place, and unless the percentage of
similarity between object and category is 100%, within the margin of error accepted,
the creation of as many external vectors as necessary for those qualities of
the object not related with the place assigned, being external vectors linked
with those qualities of the object not matching with the category within the margin
of error, only accepting a wider margin of error for utilitarian attributions.
The
creation of new places in the conceptual scheme for new categories/factors where
to place objects related to these new categories/factors, will have as a
consequence the creation of new vectors in the conceptual scheme linking the
existing places with the new place, making changes in their respective vector
weight, as well as the creation of new vectors to assign importance vector.
The
creation of new series of vectors can even create new conceptual sets in the second
stage of the categorical Modelling System, as well as it can make further
changes, as for instance the possible revision if by chance there are already
objects in other places suitable to have a higher percentage of similarity with
the new category, as well as the possibility to include in the existing models
of the current objects new relations
within their models as a result to have new vectors in common with the new
category/factor, what it could make global changes in the comprehensive
conceptual model. Changes that sooner or later must be reflected in how the
model is structured in the conceptual map, making as many changes as necessary
in all the models in the map due to the categorical/conceptual collaboration.
Finally,
in the third stage, the apparition of new categories/concepts related to new
real objects can make possible the creation of new sets of decisions related to
any new quality of this new category/factor, and/or the possibility to set up
new sets of decisions as long as the robotic collaboration between
intelligences can increase the robotic capabilities of an intelligence,
increase in the robotic capabilities making the intelligence more able to make
some decisions as long as it has more robotic devices available, increasing the
robotic functions to be included as set of possible decisions to include in the
third stage of the categorical Modelling System.
As
long as the collaboration between intelligences allow them to have more access
to new categories/factors increasing their ability to comprehend and explain the
world, the increasing collaboration between intelligences sharing robotic
devices will make possible the addition of new robotic functions, more
capability, linked to existing qualities in the already existing
categories, and/or linked to new qualities within the new categories/factors.
Reviewed 18 May 2025, London, Leytostone
