Malaysia is not supposed to offer religious equality

MALAYSIA IS NOT SUPPOSED TO OFFER RELIGIOUS EQUALITY 

/ a review of legal and emotional structures /

For education's sake, let's look at the Malaysian constitutionality of the issue.  

1.

Article 11.4. delegates to EACH STATE the power to restrict any non-Muslim religion NM, from proselytisng to Muslims M. 

1.1. EACH STATE has in the 1970s, subscribed to a CIVIL LAW framework called the EUBI or Enakmen Ugama Bukan Islam. This variously includes the arrest without warrant, and detention without trial, of NM who do things such as talk to M about other religions, or who use certain words legislated as exclusive to M.

1.2. You can Google the exact text of these laws. Because we have a poor legal rigour about precision of all law writing and enforcement in general, some of this contradicts with the practice of allowing, for example, other religions to use the word Allah. It's a broad and messy landscape, which politicians are incentivised to "avoid tidying up".

2.

Article 160.2. defines what a Malay person is. This includes professing Islam. Famously pivotal. I call this, with respect, "kedukaan sebangsa Malaysia" , because its adjacent articles have led to 70 years of violent and inefficient manipulation by politicians. 

2.1. The result of articles 11.4 and 160.2 combined : MANY citizens of Malaysia view inequality of religious rights as a core element of Malaysia's federal identity. To them, any discourse about making such inequality into an equality of rights, is TREASON against the implicit values expressed by the explicit text of the constitution. 

2.2. So people like Ronnie are, with some awareness of the environment, taking a risk by raising such issues. This is a political decision, within present environmental conditions. As environments change, such discussions are more or less politically risky. Right now, they remain quite risky.

3.

The risk is less one of breaking the law, and more one of scaring and angering the sorts of citizens who feel that their values are being threatened. The constitution itself does not forbid proselytism to M by NM, leaving it to the states. The state laws may of course be challenged for constitutionality in the future. 

3.1. The risk of raising the issues comes with opportunity for the following sort of strategic nuance. Using the current concern as an example, 

/ WHAT TO DO /

3.2.a. NM must succinctly express their concerns, citing constitutional articles, about where NM rights are being infringed. 

3.2.b. NM must avoid placing blame on rights constitutionally granted to M. 

4.

In order to do this, it would seem that Malaysian politicians should consult with both lawyers and public relations specialists before making public statements. But perhaps this is expecting too much from politicians.

transformer architecture's non-determinism problem in normative applications

( Admittedly, my verbalisation of this concern is an incomplete study. )

Broadly, T-arch is popular because it transposes time series data to spatial embeddings, allowing timeless parallel compute, for next-token prediction.

1. The parallelism itself in some cases introduces non-deterministic results depending on the hardware and kernel implementations. Folks take shortcuts when imprecision can be afforded.

2. More broadly, presuming a deterministic implementation : a common inference pattern is [ first input + first output -> next input, loop ] in scenarios where end-users want to know how past training data : probabilistically generates series data, under the assumption "present series behave like past series". This is to use T-arch as a DESCRIPTIVE tool.

However : T-arch is often used in tools expected to be NORMATIVE, enforcing specific governance (control). Because T-arch is fundamentally descriptive, governance internal to the looping inference process cannot be deductively guaranteed for novel inputs. Instead the entire T-arch loop must be wrapped in something like old-fashioned logic programming, as in proof assistants, and perhaps including supervised learning architecture.

Buyer beware.

Credit for discussions 

- https://lnkd.in/gdAYcMaY

Kant (1781) to Hassabis (2007)

Of late, Md Ismail Sojal has been one of the most prolific, coherent, Facebook content curators, on topics related to artificial intelligence. Today he posted a brief summary of Demis Hassabis' career, which caught my attention. I took particular interest in learning that Hassabis published important papers on the hippocampus, from 2007 onwards, in an area of my research : delineating the software topology of human intelligence. Given how old the paper is, I've probably been influenced by it in ways immemorial. Its findings align very well with my own practical study of how to quantify my conscious experience on moment to moment basis, something I've been studying since around 2003.

In short, the hippocampus ( whose shape reminds me of the clit ), is a part of the brain, which appears to enable what is broadly called "apperception" in Kant's Critique of Pure Reason, Edition A, published in 1781. The hippocampus is structurally significant in data pipelines, because it mediates both the sensory-spatio modelling of ...

• live data, before that data can be sent to long term memory, and
• recalled data, from long term memory

... and in both cases, before it can be consciously contemplated ( "reasoned about" ). 

In the AI development ecosystem of 2026, of course we can then association such research with the current trend into "latent world models", such as JEPA, MuZero (also associated with Hassabis), and their kin.

A couple of brief notes :

• Live sensory information is passed from the sensory nervous system (SNS) to the hippocampus, from regions of the brain which evolved earlier, e.g. thalamus > neocortex > hippocampus.
• The types of long term memory affected by the hippocampus include "explicit memory" which refers to abstract verbal concepts, and non-motor sensory data bound into gestalten by such concepts ( such as faces ), but it excludes "implicit memory" which refers to the limbic association cortex which learns processes of motor neural system (MNS), such as riding a bike.

Further reading : 

• In another vein, the evolutionary history of the hippocampus deserves some reading also, if you're into developmental cognitive science. It turns out that the hippocampus plays a role in the biological phyla of chordates ( including animals, birds, reptiles, amphibians, and fish, among others ), which is played by differently evolved anatomical structures in other phylla of the kingdom animalia.
• In humans, while the hippocampus integrates short-term sensory data into multi-modal spatio-temporal models, longer-term memories are stored in unimodal cortices of the neocortex. These unimodal cortices are also linked to association cortices, which semantically relate various sensory information together. At this junction I must admit it's still not entirely clear to me how much of what we call "conscious experience" can be said to be associated with the hippocampus versus the association cortices. I'll have to continue this study later, though it is not a priority for me this year.
• The thalamocortical and basal ganglia circuits form three parallel, segregated loops, that are worthy of study : 
• Thalamostriatal Pipeline ( input )
• Basal Ganglia Processing ( filter )
• Pallidothalamocortical Pipeline ( output )
• The thalamus and superior colliculus are involved in the salient network ( see "triple network model" ), and their key pathways are 
• Tectopulvinar Pathway ( feedforward )
• Thalamotectal Pathway ( feedback )
• Thalamic Reticular Network ( modulatory )
• The hippocampus and cerebellum participate in a bi-directional non-motor network, which should also be read deeply.