introduction

This sequence of ’toot-sized digests’ (eventually), compares the accessibility of science-writing and software, using a generalised method/ process/ system (?) for thinking across distinct conceptual domains; which began as a set of principled assertions and intuitions for illuminating and reasoning around/ about anomalous phenomena (originally conceived and developed for this #project)

skip to: {#results; #analysis; #discussion}

occasion

An online conversation about the frustrating inaccessibility of the language and writing of some scientific papers; which began here (with @ngaylinn); included a variety of interpretations of the problem-space (multi-disciplinary ?); and continued through observation of curious similarities between the accessibility of science-writing and software here (see: #the initial similarities); and on to the following:-

prompt

i keep coming back to the comparison between science writing and readable code / tech writing. in some ways they have very similar requirements regarding precision, technical detail, and simultaneously targeting an audience of experts and novices (if not the general public).

yet, the outcomes seem different to me. i feel like coding projects are on average more accessible, and there are popular systems, tools, and processes to encourage that with no parallels in academia.

if that’s so, then perhaps we can learn by comparing these two domains. what makes them different? my first instinct is incentive structures, but i haven’t thought that through yet.

—@ngaylinn

response

#tbc

Yeh, same here. Some unavoidable distractions aside, i’ve been chewing over this conversation/ comparison since.

I’m confident I now better understand writing accessibility, in theory anyway… That said, theoretical understanding and ability are distinct; and of course I may be wrong entirely!

Please all critique freely, to challenge ideas and refine language.

conclusion

Software code is more accessible than science writing because software has a well-defined general-domain, to standardise special-domain composition; and science does not (but ought to)

Inaccessibility is one aspect of technical-debt. Refactoring technical-debt, by principled iterative recomposition of special-domain concepts, around standardised principles, primitives and intermediates, of a well-defined general-domain, improves accessibility, productivity and quality.

what does all this mean?

general headlines

#rewrite this section

1. Software code is more accessible than science writing because software has a well-defined general-domain, to standardise special-domain composition; and science does not (but ought to)
   1. Inaccessibility is one aspect of technical-debt. Refactoring technical-debt, by principled iterative recomposition of special-domain concepts, around standardised principles, primitives and intermediates, of a well-defined general-domain, improves accessibility, productivity and quality
2. writing (the process of) has a strong dependency on reading (the process of); writers must sufficiently understand the mind of target reader audience to optimally encode knowledge to language, such that a target reader audience sufficiently-accurately decode language to knowledge #rewrite
   1. Language is a (lossy) knowledge serialisation protocol, and inherently ambiguous
   2. Writing encodes high-dimensional concepts to low-dimensional terms and phrases, via an (effective) linguistic-conceptual encoding-space ¹
   3. In turn, reader interpretation depends upon a sufficiently-equivalent (linguistic-conceptual) encoding-space (or interpretive context) between writer and reader, such that ambiguity inherent in language is sufficiently-equivalently resolved ²
   4. When language ambiguity is insufficiently/ incorrectly resolved, the result is uninterpreted or misinterpreted language
3. Correspondingly, general writing advice is to write to a specific audience, and for technical topics, as specific (niche) an audience as possible, to reduce the complexity of interpretation, and possibility of term collision
4. Interpretability and accessibility are subjective
   1. No absolute accessible
      1. 4. Cannot serialise once for all audiences
      2. Absolute accessible
5. Stack analysis: to better understand the accessibility of science writing we must consider the phenomena in context of operational/ cognitive/ biological stack
   1. Accessibility is more than just words {accessibility; productivity; quality}
   2. Accessibility (translation from special to general domain)
   3. Interpretive generosity
      1. Additional compositional effort
         1. Recomposition from general domain increases compositional complexity/ interpretive burden
         2. Cost of communication by simpler forms may initially require reader conceptual refactor
      2. In-group/ out-group
         1. Reduction in in-group signalling, means writers must hope for change in preemptive filtering by target audience
   4. Operational cognitive and biological imperatives
      1. Language 7
6. Valid refactors are finite
   1. Inability to translate is indicative of improper form
   2. Can’t get to anywhere from anywhere
      1. Not all problem-spaces can align with all solution-spaces
   3. Valid abstractions are finite
   4. Partial translation is instructive
   5. Build out from there {standard library; design patterns}
   6. Abstractions not infinite

   7. • Some abstractions don’t line-up with other abstraction well

   8. Cannot serialise once for all audiences
   9. Unified perspective
      1. Common general domain
      2. Common measures
      3. Where territory is continuous, maps ought to be continuable
7. A well-defined general domain eliminates the problem of knowing which minimal/ specific/ isolated audience encoding-space to target
   1. A well-defined general domain
   2. Eliminates the problem of knowing which audience to write to
   3. Better aligns with continuous territory

accessibility questions

The accessibility of science writing matters; but not all science writing is accessible.

1. why?
2. what is accessible writing?
3. why it is important; uncommon; difficult; etc?
4. what can be done to improve accessibility?
5. how accessible is accessible?
6. why care?
7. how do we reach agreement on the value of accessibility?
8. can we define targets/ measures objectively?
9. what is the problem we need to solve (or better understand)?
10. what does all this have to do with software?!

on reframing

—why reframe ’the accessibility of science writing’ in terms of ‘other seemingly-unrelated-domains’, like software?

1. for a fresh perspective
2. to ask new questions

Consider, the easiest (or only) way to resolve an entrenched problem, is from outside the trench.

For any discipline.

1. Anomalous-detail and unchecked-assumption distract and mislead
2. Both anomalous-detail and unchecked-assumption are rendered ineffective by disciplinary ignorance
3. To counter from the inside, specialists must unsentimentally de-emphasise and defamiliarise special-detail, to re-perceive phenomena in place ³, and without prejudice

thinking across domains

#tbc

Distinction is relative to commonality, necessarily.

to think across (distinct conceptual) domains is to see past that which is distinct-between arbitrary special-domains, to that which is common-across; the relative general-domain, from which all special-domain distinction, is distinct relative to ⁴

a general-domain does not replace special-domains, but relates and reframes special distinction, by relative commonality

material objective

#tbc

Consider :

1. the scientific endeavour is a collection of individuals pursuing common ideal of ’evidence based knowledge’ ⁵
2. We get system change, by aligning fundamental {effort; motivation; value} across relevant parties (which includes the ‘common ideal’ itself), to redefine a common objective
   1. Redefined objective: ‘better evidence based knowledge’
3. Explore and describe ways in which accessible scientific writing will result in ‘better evidence based knowledge’
   1. In terms which relate-to and align motivation across all relevant parties ⁶ (individuals; isolated disciplines; the common ideal)
   2. And might feasibly result in ‘better evidence based knowledge’
4. Deliverables:-
   1. Identify parties {reader; writer; specialist; relative generalist; individuals; the system} (#peer-analysis)
   2. Align {motivation; operation; material objective} (#stack-analysis)
   3. Define achievable way-points, which intersect motivation, operation, material objective

approach

#tbc

Compare the accessibility of science writing with software.

1. Illuminate phenomenal constituents (#stack-analysis; #peer-analysis)
2. Identify fundamental comparative aspects
3. Map {isolate; generalise} commonality
4. Reframe problem domain by recontextualising generalisations

#rewrite

overview

#tbc

results

domain translation

Consider accessibility as ‘domain translation’

To make writing-or-code accessible, we translate from a special-domain, to an implicit general-domain; such that special-concepts are recomposed from more general-concepts; then re-described in corresponding terms.

general-domain primitives

The basis of any general-domain are primitives: the smallest units of composition, from which all other complexity is composed.

Primitives: for writing are {language; basic concepts}; for software are {language; standard library}

Software’s general-domain is ‘well-defined’ because it includes standards for recomposition, including standardised intermediate-generalisations, known as ‘common design patterns’

general-domain intermediate-generalisations

Common design patterns, or intermediate generalisations :

1. Represent structural and behavioural characteristics, common across arbitrarily-plural special-domains (special-domain invariant)
2. Increase the volume and sophistication of defined conceptual vocabulary available for recomposition

The more well-defined the general-domain, the easier the task of recomposing special-domain-concepts, to standardised form.

general perspective

Consider :

1. The result of a successful push to make science writing accessible: arbitrarily plural special-concepts, from distinct special-domains, are (to varying degree) recomposed from simpler common constituents – standardised concepts from a single common general-domain – a new general-perspective
2. A general-perspective does not replace special-perspectives, but relates and reframes special distinction, by relative commonality

intermediate-generalisation summary

1. To use intermediate-generalisations (common design patterns), is to translate by pre-synchronised well-defined concepts
2. The more well-defined the general-domain, the easier the task of recomposing special-domain-concepts, to standardised form
3. Once familiar with translating own special-knowledge to the general-domain, interpretive generosity (self-recontextualisation of others ambiguous communication) improves

–how might common patterns might apply to scientific knowledge?

general-domain summary

#tbc

accessibility summary

#tbc see: language

Accessibility is more than presentation to ease reading.

• Accessibility depends upon conceptual formation (structural composition)
• and refactoring; to separate out distinct aspects of conceptual concern into representational fragments which comply with signature of fundamental forms, which correspond to general terms. supporting forms must also be synchronised to ensure sufficient resultant/ corresponding understanding #rewrite

To describe concepts in simpler or general terms, we must refactor concepts into compositions whose constituents map to those terms: we must first restructure the knowledge and understanding behind words.

#todo

• All cognition and representation is necessarily abstract

writing summary

#tbc see: language

cost

effort and time

Translation from special to general-domain, is arbitrarily challenging and costly in effort and time; And so is translation back, from general to special – the reader is asked to reinterpret and recompose special-concepts from arbitrarily-plural fractions of the whole.

Accessibility increases compositional-complexity of well-formed special-domains.

Between two specialists of the same domain, this challenge and cost might appear redundant and unnecessary.

–why bother?

motivation

–how does costly accessibility result in better evidence based knowledge?

Accessibility is more than words and readability.

For software, accessibility, by well-defined general-domain, also increases productivity, and quality.

In practice, framing costly pursuits in terms of improved productivity and quality, matters.

productivity

To make accessible, we translate special-domain concepts to a pre-synchronised well-defined general-domain.

The more well-defined the general-domain, the easier the task of recomposing special-domain-concepts, to standardised form.

When familiar with translating own speciality intrinsic special-concepts to the general-domain, the interpretation of speciality-extrinsic accessible writing becomes easier. Good compositional abstractions simplify and reduce cognitive-load.

quality

Consider inaccessibility as lack of visibility of explicit conceptual form.

1. For software, accessibility by well-defined general-domain, increases quality
2. To make accessible, is to decompose special-concepts into well-defined constituents
3. Implicit relational detail, is isolated and encapsulated; made architecturally explicit

Irreducible composition commonly hides unseen, unchecked assumptions, and inexplicit behaviour; which evaporates when form is decomposed and recomposed to explicit well-formed general-domain.

like refining ore; smithing metal to remove impurities; refactoring code to leave fewer places for bugs to hide

special-concepts

#tbc

special-domains

#tbc

circumstantial (shallow/ lateral) grounding

#rewrite describe as constraint compliance

• Circumstantial (shallow) grounding
• Minimal viability; prototype; first-stage
• Input and side-effect/ artefact
• Integration test: arbitrary implementation

compositional (deep/ vertical) grounding

#rewrite describe as constraint compliance

• Architectural constraint (interfaces); single responsibility principle/ unique behavioural profile (interface); unit test: specific implementation
• To recompose special-domain concepts from a well-formed general-domain, is to compositionally ground understanding, by fundamental {prior; ancestor; constituent}
• To compositionally ground, is to correctly account for inherited phenomenal characteristics
• When we fail to recompose understanding by more fundamental forms, we contrive complexity, by double-counting the phenomenal origin and nature/ characteristics of derived special forms

natural understanding

#tbc on territory

—are we saying that maps which are compositionally restructured around well-formed relative general-domains are better aligned with territory?

—what does this say about territory?

paradigmatic unification

#tbc

• scientific disciplines are isolated and fundamentally anomalous, because science lacks a well-defined general-domain
• isolated perspectives divide populations: making scientific writing accessible will unify perspectives, and consequently, populations of scientists

—but might it do more?

puzzle-space

#tbc

A well-defined general-domain renders all special-concerns, puzzle-space.

alignment

#tbc

representational compliance

Not every special-domain concept is equally valid.

For any occasion, the space-of-all :

1. Legal, appropriate representation, is finite
2. Illegal, inappropriate representation, is approximately infinite

Legal, appropriate representation requires.

1. Sufficient, necessary, detail
2. Compositional form compliant with the inherited constraints of respective representational substrate

Standardised intermediate-generalisations model sufficient, necessary detail; in a way which complies with the inherited constraints of the general-domain.

dogfooding

Dogfooding refers to using your own software products, as a demonstration of quality.

–what does that have to do with accessibility?

The method of analysis used here, to compare the accessibility of science writing and software, is a direct application of the analysis developed for this project, to isolate, reconcile and account, for phenomenal constituents of direct-experience and cognition, across distinct conceptual domains, laterally, and vertically.

consilience

Consilience: (also ‘convergence of evidence’ or ‘concordance of evidence’) is the principle that evidence from independent, unrelated sources can “converge” on strong conclusions. That is, when multiple sources of evidence are in agreement, the conclusion can be very strong even when none of the individual sources of evidence is significantly so on its own.

Standardised phenomena-invariant scientific intermediate-generalisations, objectively satisfy high-dimensional convergence/ concordance/ consilience.

simplicity

“a theory is the more impressive the greater the simplicity of its premises is, the more different kinds of things it relates, and the more extended is its area of applicability” —albert

review

#tbc

analysis

science software mapping

Think of software code-languages more as a document-format than a distinct discipline.

For both software and science, consider the pursuit to be: ‘representational world-modelling, within representational constraint’

Not all scientists/ software-engineers do the same thing, but trends relate to :

1. Scope {breadth; depth} of phenomenal concerns {groups; individuals}
2. Representation evaluation; evaluation lifecycle
3. Consequence
4. Self-governance/ evolution; evolutionary lifecycle
5. Scope of standardisation

standardised representation

Standardised representation is the result of cross-domain generalisation.

To generalise is to simultaneously consider arbitrarily-plural special-domain forms, and dismiss uncommon (typically contextual) detail from each.

To generalise is to decontextualise form/ characteristics.

{constraint; relation (structure); mutation (behaviour)}

standards

#tbc

Common standards include :

• Principles {single responsibility; separation of concerns; dry; kiss}
  • Measures
  • Validation
• Representational composition
  • Primitives
  • Intermediates

discretionary standards compliance

#tbc

Software projects :

• Are live documents; iteratively maintained
• Discretionary/ staged standards compliance {minimal-viability; prototype; beta; pre-release; release; versioned}

Software :

1. Conditionally accepts non-standards compliant expression
2. Allows intuition and reckoning to
   1. Initially map and reason a space
   2. Demonstrate viability (fundamental alignment)
   3. Before formal reconciliation

software notes

#tbc

domain reconciliation

#tbc

stack-analysis

vertical

1. Academia; enterprise
2. Science writing; software code
3. General writing
4. Writing (process); text (state)
5. Reading (process)
6. Language (state: serialised knowledge) {encoding; expression; etc}
7. Communication (process: knowledge synchronisation) {language serialisation; interpretation}
8. Cognition (process) {phenomenal interpretation; cognitive imperatives}
9. Knowledge (state) {representation; context}
10. Biology {autonomy; biological imperatives; etc}
11. Other physical concerns

conditionality stack

#tbc

1. Collective conditions
2. Condition domain
3. Encapsulated conditions
4. Persisted conditions
5. Interpreting conditions
6. Serialised conditions
7. Synchronising conditions
8. Operational conditions
9. Integrated/ configured conditions
10. Material/ physical conditions
11. Foundational condition substrate/ primitives

peer-analysis

lateral

1. Individual/ collective/ demographic
   1. Writers; readers
   2. Specialists; relative generalists (specialists of any other discipline)
2. System (the collective pursuit: the scientific endeavour; or writing)

–why?

1. system change follows individual change
2. align motivations by reframing the topic of writing accessibility in terms common to all parties
3. necessary system change follows sufficient motivational alignment between parties #rewrite

discussion

granular alignment constraint

#tbc reckon, reason, reconcile

• Shut up and walk
• We banished reckoning; at a time before we understood fuzzy?
• Reckon, reason, reconcile

kiss

KISS: keep it simple, stupid!

Direct recognition of our collective predisposition to contrive redundant complexity.

cause

#tbc

universal intermediate generalisations

#tbc

• On machines
  • Metaphor
  • Generalisation
  • As conceptual design pattern
    • Standard library

academia

#tbc

conclusions

#tbc

questions

more

—why? —how do we get to coordinated promotion of the values of accessible writing? align motivation multi-aspectual —how might an concerted push to ‘make scientific writing accessible’ result in ‘better evidence based knowledge’ (for {individuals; disciplines, and; the scientific endeavour} specifically)? —what might be said of academic measures:- of the health of individuals? of individual productivity? and the quality of output? —what more can we say about software accessibility? —what can we say about the origin of software accessibility? —what do common software design patterns do? —how do common software design patterns relate to accessible scientific writing?! —why care about writing accessibly?

notes

incoming

the initial similarities

observed similarities between scientific writing, and software code

On scientific writing (independentpen post link) :

1. Precise writing does not need to be illegible
2. Most of the illegibility in my field comes from extreme abstraction, which is imo the opposite of precision
3. To write clearly and precisely requires thinking clearly and precisely, and that is hard work
4. It’s easier for many not to write clearly, but to imitate a style steeped in status whose inaccessibility is an implicit gesture of hierarchy

On software code (themanual4am post link) :

1. Precise code does not need to be illegible
2. Excessive abstraction affects legibility; which is very different to precision
3. To code clearly/ precisely is to abstract well (is circumstantially appropriate)
4. Use of terse/ short variable-name style: implicit gesture of hierarchy/ status

subject object

#tbc

• Isolated disciplines are subjective
  • Doesn’t matter how many people have agreed
  • Higher-population subjective
    • Synchronised frame of reference
• The general-domain is is the most fundamental common reference frame for relative distinction

general intelligence

#tbc

A well-formed scientific general-domain: is the general, behind general intelligence.

agi

#tbc

A well-formed scientific general-domain: is the general, behind artificial general intelligence.

street-light effect

#tbc

A well-formed scientific general-domain: is to illuminate the street which frames and relates all street-lit areas.

the conversation

#tbc

to nate, independentpen and others, the problem of inaccessible scientific writing matters – because researchers time and efforts are unnecessarily {consumed; hampered; wasted}, by various combinations of uninterpretable ⁷ language, and writing style. and consequently, opportunities to learn are being missed; specifically, and generally

the initial similarities (between the accessibility of science writing and software code) are especially interesting to me, because they hint at an opportunity to apply the approach and methods of this project to a domain i had not yet considered (the accessibility of scientific writing/ and writing accessibility more generally)

domains

#tbc

• Thinking across domains
  • We tend to think lateral domains
  • We navigate lateral domains by depth
  • Control flow
• Stack analysis
• Constraint dependency flow
  • General special
  • Ancestor descendant
  • Constituent composition
• Abstractions
• Approach
  • Thinking across conceptual domains
    • New perspective, a way to ask new questions
  • Lateral and vertical alignment
    • Commonality {state; process}; constraint {propagation; inheritance}

premise

#tbc

stack analysis notes

#tbc

1. Scientific writing/ academia
2. Writing generally; the written word (state)
3. The act of writing (process)
4. The act of reading (process)
5. Language (state) {encoding; expression; conceptual mapping; etc} serialised knowledge
6. The act of communication (process) knowledge synchronisation; language serialisation/ interpretation
7. Cognition (process) {interpretation; distinction; ambiguity; viability; coherence; translation} cognitive imperatives
8. Knowledge (state) {representation; context} representational constraints
9. Biology {autonomy; etc} biological imperatives
10. Fundamentals physical constraints

language

see: language #tbc

It’s easy to forget that language isn’t knowledge.

Language is a (lossy) knowledge serialisation protocol, which must be interpreted by sufficiently correct pre-synchronised linguistic-conceptual {encoding; embedding} space.

writing audience

#tbc

writing accessibility

#tbc

national language

—is someone writing in a foreign language failing to make texts accessible? —does the accessibility process also apply/ map/ translate across languages? —might a side-effect of the formalisation of process aide future translation between national languages —do language characteristics/ tolerances tell us anything about the outer scope of —communication? —including dialects? —slang?

Consider traditional perception of unfamiliar {national language; dialects; slang} on significance of domain speciality.

complexity

#tbc

• Difference between contrived and innate complexity
  • Entropy
  • Generous interpretation

perspective

1. Irreconcilable perspectives divide populations
2. A unifying perspective does not replace other perspectives, by reframes and aligns those parts (of other perspectives) which are mutually consistent

Continuous territory, requires continuable maps.

staging

scoped accessibility #tbc

project

#tbc

This project, the manual (for all minds), is the result of a {method; system; process} for ’thinking across traditionally isolated domains of concern’, applied to ‘our collective understanding of the mind’ (the quartet)

The mind is an unhelpfully controversial topic at the best of times, delicate, and contentious ⁸.

Before ‘going loud’ with results of the primary analysis (of the quartet), I needed a way to bypass topical contentions, to continue refinement and validation of the {method; system; process}, by application to other less-contentious problem-spaces/ domains of concern. Helpfully the analysis generalised ⁹; and for several months, I have been exploring arbitrarily random domains/ problem-spaces ¹⁰.

It turns out this generalised analysis also applies directly to both ’the accessibility of science writing’, and software refactoring: which is particularly interesting because both include concrete examples of different domain-specifics implementations; but also the former presents an accessible example of the analysis on our fundamental understanding of universal phenomena, and by extension, the nature of phenomena directly.

system change

#tbc

• System change follows
  • Individual change
    • Who are?
      • Readers
      • Writers
      • Relative specialists (same domain)
      • Relative generalists (other domain)
    • Aligned motivations
      • Reframing the topic of writing accessibility in terms common to all parties
  • Necessary system change follows sufficient motivational alignment between parties
    • Writers; readers
    • Specialists; relative generalists (specialists of any other discipline)
    • System (the collective pursuit: the scientific endeavour; or writing)

sagan

1. One way to build intuition for this idea, it to refer to carl sagan’s apple pie, albeit described somewhat differently :
2. An apple-pie does not exist in isolation
3. The physical characteristics of an apple pie are the result of the physical characteristics of ingredients (and the physical mutations applied by preparing and cooking), and as follows, the physical characteristics of each constituent ingredient do not exist in isolation, and are the result of biological, and chemical, and physical constituents, all the way down
4. The important observation, is that however varied any individual instance of apple-pie may be, each incremental phenomenal composition, from fundamental physical priors through chemical and biological forms, is the result-of, and constrained-by, simpler prior forms
5. The phenomenal structural and behavioural characteristics of an apple pie do not exist in isolation. The space-of-all-possible manifestations of apple pie, is finite (relatively); constrained and defined by priors, constituents and causal peers

musical remixes

Musical remixes as recontextualisation (new special-domain) of isolated generalised intermediates.

plagiarism

#tbc

In the event avocation for accessible texts is successful, the broader system (of intersecting domains of concern) will be forced to change: because there are finite ways to describe the same phenomena, and by decreasing vocabulary, we further decrease the space-of-all-possible descriptive variations; such that, in short order, present means of considering and verifying plagiarism will cease to be valid/ feasible, at all (this is not an argument against!)

conclusion

#tbc