roirpt

(pronounced /ʁwaʁp/)

Stuff for stoiners (pronounced /stwæ.ɲe/)

I basically just git initted in my plover dir and started committing WIPs

This README is better viewed in full width (> 1100 pixels)

Contents

A hundred dictionary files? Why so many?

File formats for storing stenography dictionaries are extremely limited: JSON is the default (among other interchangeable rudimentary key–value formats), while a few plugins allow for Python scripts (dynamic dictionaries) or Markdown documents (entries embedded in code blocks interweave with verbose comments in Literate Programming style) to behave as dictionaries.

There is no systematic capability available for rich, well-structured dictionary data with annotations, tags, or metadata holding explanatory comments, part of speech, multiple categories (such as domain or context of use), interlinking cross-references, creation/modification dates, usage or misstroke statistics, review or conflict status, preferred/prescribed outlines, or any other elaborating information or methodology. There is no preprocessor or advanced abstraction to factor common entries in terms of other entries, like affixes, constituents, compounds (e.g. “always capitalize any combination of united followed by kingdom”). This is not conducive to learning or communicating, since dictionaries often function as first references in the absence of adequate pedagogic material.

The only option is to maintain many separate dictionaries.

There are many advantages to splitting: Each dictionary’s filename labels its purpose. It is easier to manually inspect and manipulate entries, since similar entries tend to be closer to each other. Separate dictionaries can be reordered, enabled, or disabled at will. Entries demonstrating one concept are found in a single file easily shared with others (“here are all of, and only, the entries that follow the AEBGT → ject pattern”) that can be borrowed without worrying about massive conflicts.

Indeed, it is unwieldy to maintain a large number of dictionaries in Plover’s interface, and not all entries cleanly fit within a single dictionary (but choosing one is not a science).

caret-*.json

Introduces ^ to represent a pre-initial schwa.* Requires the plover-stenotype-extended plugin.

A few groups of rhymes so you can see why this might be useful:

Group	Word	Steno	Pseudosteno	Plover default entries
-end	pend	PEPBD	(PEND)	PEPBD
	spend	SPEPBD	(SPEND)	SPEPBD
	append	^PEPBD	(^PEND)	A(P)/PEPBD
	expend	KPEPBD or ^SPEPBD	(XEND) or (^SPEND)	KPEPBD or EBGS/PEPBD
	extend	^STEPBD	(^STEND)	EBGS/(S)TEPBD
-ent	sent	SEPBT	(SENT)	SEPBT
	assent	^SEPBT	(^SENT)	AS/SEPBT
	ascent	^SKREPBT	(^SCENT)	AS/KREPBT
	accent	^KPEPBT	(^XENT)	ABG/SENT or ABGS/EPBT or KP*EPBT
	stent	STEPBT	(STENT)	STEPBT
	extent	^STEPBT	(^STENT)	EBGS/((S)T)EPBT
	competent	KPEPBT	(KPENT)	KPEPBT or KPE/TEPBT or KOFRP/TEPBT etc.
	scent	SKEPBT or SKREPBT	(SKENT) or (SCENT)	SKEPBT or SKREPBT
	consent	SK*EPBT	(SK*ENT)	SK*EPBT or KAUPB/SEPBT etc.
-ent	intent	SPWEPBT	(INTENT)	SPWEPBT or EUPB/TEPBT
	absent	^SPWEPBT	(^SBENT)	AB/SEPBT

* Note: This is not a new idea; people have implemented similar ideas before in different ways (see phenrsteno for where I got the inspiration). I chose ^ since it looks like both a small superscript A and also the IPA symbol ʌ, which is used to represent a (usually stressed) schwa in English. I like using a punctuation character (rather than e.g. a or e) because I like the ability to easily grep for steno strokes as strings of only uppercase letters or symbols, and because in general-purpose search engines I can search for strokes and it will return results that ignore the symbol (although that could eventually be a footgun).

Sample of other groups of words that ^ trivially or neatly solves:

Group	Word	Steno	Pseudosteno	Plover default entries
1	way	WAEU	(WAI)	WAEU
	away	^WAEU	(^WAI)	W*AEU
	ware	WAEUR	(WAIR)	WAEUR
	aware	^WAEUR	(^WAIR)	A/WAEUR
	long	HROPBG	(LONG)	HROPBG
	along	^HROPBG	(^LONG)	AOLG
2	lesson	HREFPB	(LEFN)	HR*EFPB or HR-FPB
	eleven	^HREFPB	(^LEFN)	HREFPB
3	limit	HREUPLT	(LIMT)	HREUPLT
	eliminate	^HREUPLT	(^LIMT)	HREUPL/TPHAEUT
4	least	HRAOEFT	(LEEFT)	HRAOEFT
	alleviate	^HRAOEFT	(^LEEFT)	A/HRAOEFT or HRAOEF/KWRAEUT
5	motion	PHOEGS	(MOESHUN)	PHOEGS
	emotion	^PHOEGS	(^MOESHUN)	AOEGS or AOE/PHOEGS
6	norms	TPHORPLS	(NORMS)	TPHORPLS
	enormous	^TPHORPLS	(^NORMS)	TPHO*RPLS or TPHOERPLS
7	involve	SROFL	(VOFL)	SROFL
	evolve	^SROFL	(^VOFL)	SRO*FL etc.
8	vice	SRAOEUS	(VIIS)	SRAOEUS¹
	device	TKWAOEUS	(DVIIS)	TKWAOEUS etc.
	advice	^TKWAOEUS	(^DVIIS)	SREUS etc.
9	stipulate	STEUPLT	(STIPLT)	STEUP/HRAEUT
	stimulate	STEUFRLT	(STImLT)	ST*EUPLT
	estimate	^STEUPLT	(^STIMT)	STEUPLT

¹ I call these Kipling-style briefs since they drop the entire first minor syllable

Sample text (schwa manifesto)

IDK ^TK-BG why you would avoid ^SROEUD using pre-initial ^TPHEURBL schwa at^T thisTHEU pointPT ^THEUPT. It’s the easiest and most consistent available ^SRAEUBL iteration ^TRAEUGS upon the ^POPBT Eng^STKPWlishHREURB ^STKPWHREURB layout that you can inc^TKPorporateORPT ^TKPORPT to adv^TKWanceAPBS ^TKWAPBS e^fficieTPEURBncyS ^TPEURBS without un^TPHnecessaryTPHES ^TPHES memorization. Its om^*PHissionEUGS ^PH*EUGS by Ward Stone Ireland seems unusual ^TPHAOURBL in ^TPHretrospectRERPT ^TPHRERPT. Thankfully, it is easy to acquire ^KWAOEUR and accessible ^SEBL, ^KPES/-BL, etc. to exploit ^SPHROEUT on most hobbyist steno keyboards by customizing a few ^TPAOU essential ^SERBL things as^S needTPHAOEDed-T ^STPHAOETD.

Apologies ^POLGS for a^ littleHREUL bitBT ofF ^HREUFBLT innocuous ^TPHOBGS fun. This extra ^KPRA key just makes me want to express ^KPRES emotions ^PHOEGSZ about how I feel so alive ^HRAOEUF and as^S freeTPRAOE asS ^STPRAOES an avian ^SRAOEPB ani^TPHEUmFRalL ^TPHEUFRL! Join the avant ^SRAPBT garde! You’re not alone ^HROEPB and can learn it in a day ^TKAEU or only within a ^W-PB week. I welcome you to opine ^PAOEUPB with your approach ^PROEFP to writing all^HR ofF theT ^HR-FT assorted ^SORTD marked words. Hoping I don’t annoy ^TPHOEU or offend ^TPEPBD a ton of ^TOFPB people by evange^SRAPBGlizLZing/-G ^SRAPBLGZ/-G its appeal ^PAOEL, but its existence ^KPEUFPBS is actually ^TWAEUL a huge asset ^SET.

Tip

Crash-course glossary of phonetic terms used

It is not necessary to know these terms, as their meaning can still be understood in context and through the examples.

Manners of articulation

Stop

Consonant with sudden release after blocked airflow: p, b, t, d, k, g.

Nasal

Consonant with continuous release and airflow through nose: m, n.

Fricative

Consonant with continuous release and turbulent airflow: f, v, s, z, sh, th, h.

Affricate

Consonant that combines stop followed by fricative: ch, j.

Liquid

Consonant with continuous release and non-turbulent airflow: l, r.

Syllable parts

Initial

Consonant (or consonant cluster) forming the onset (beginning part) of a syllable.

Pre-initial

Minor syllable before main syllable part. Usually limited to weak, unstressed light syllables: about, embed.

Schwa

Neutral vowel “uh” (ə) common in English unstressed syllables.

The ex-/comp- conflict, etc.

These dictionaries do not claim to solve the ex-/comp- conflict systematically or completely, but represent the general direction I was going for. In the case of a conflict, usually the leftmost option or the more common word gets priority.

no prefix	prefixed with S	prefixed with ^	prefixed with ^S
T  t-	ST  st-	^T  att-	^ST  est-¹ or ext-
P  p-	SP  sp-	^P  app-	^SP  esp- or exp-
K  c-	SK  sc-	^K  acc-	^SK  esc- or exc-
KP comp- or ex-	SKP and-	^KP ex- or exC-²	^SKP

Note that this table is designed to be mnemonic, not comprehensive. For example:

1. ^ST: “est-” is shorthand for “ast-, aust-, est-, ist-, ost-, etc.” (as in astute, austere, estate, Istanbul, ostensible)
2. ^KP: is ext- or exp- or exc- when used with R (as in extra ^KPRA, express ^KPRES)

Compound initial sounds

In development (not really). Looking for permanent homes for certain starters (marked with ? for now).

For certain other combinations (stop, fricative + stop, stop + fricative, stop + fricative + stop), see the table under the ex-/comp- conflict.

Type	Starter	Steno	Pseudo	Starter	Steno	Pseudo
	no prefix			prefixed with com- con- K-
Nasal + nasal
MM	amm- emm- imm-	^PH	(^M)	comm-	KPH	(KM)
NN	ann- enn- inn-	^TPH	(^N)	conn-	K	(K)
Nasal + stop
MP	amp- emp- imp-	KPW	(IMP)	comp-	KP	(KP)
MB	amb- emb- imb-	KPW	(IMP)	comb-	KPW	(KB)
NT	ant- ent- int-	SPW	(INT)	cont-	K	(K)
ND	and- end- ind-	SKP	(AND)	cond-	K	(K)
NK	anc- enc- inc- ank- enk- ink-	^TKP	(INK)	conc- conk-	K	(K)
NG	ang- eng- ing-	^STKPW	(ING)	cong-	K	(K)
Nasal + affricate
NJ	enj- inj- ang- eng- ing-³	^SKWHR	(INJ)	conj- cong-³	?
Nasal + fricative
NF	enf- inf-	TPW	(INF)	comf- conf-	TKP or TKP*¹	(KF) or (KF*)¹
NV	env- inv-	TPWH	(NW)	conv-	KW	(KW)
NS	ans- ens- ins- anc- enc- inc-³	STPH	(SN)	cons- conc-³	SK	(SK)
Nasal + fricative + stop
NSP	insp-	STPH	(SN)	consp-	SKP	(KSP)
NST	inst-	STPH	(SN)	const-	?
NSK	insc-	STPH	(SN)	consc-	?
Stop + fricative
BS	abs- obs-	SPW or ^SPW²	(SB) or (^SB)²	n/a
DF	def- dif-	TKP or TKW¹	(DF) or (DW)¹
DV	dev- div-	TKW	(DW)
DS	des- dis- dec-³ desc- disc-³	STK	(DS)
Fricative + fricative
SF	suff- surf-	STP	(SF)	n/a
SV	subv- surv-	SW	(SW)

1. when it conflicts with dep- TKP (DP)
2. when it conflicts with int- SPW (INT)
3. followed by a front vowel (e, i, y) only, i.e. “soft” pronunciations of c or g

Note that NK can optionally come before W to form a starter for NQ: inquire ^TKPWAOEUR (INKWIIR).

Example mnemonics for a few starters:

Starter	Steno	Explanation	Shape	Resembles
NK	^TKP	(K) + ^TP, like (N) shifted left	▜▘	⚓️ anchor 💂🏿‍♂️ Mr. T, who encouraged kids to stay in class
NG	^STKPW	(G) + ^S	█▌	🔳 quadrangle ingot
NJ	^SKWHR	(J) + ^H	▙▟	🌉 engineer’s suspension bridge 🚂 train engine
NF	TPW	(F) + W	▜	ꟻ  an inverted F 7️⃣ infamous favorite number
NV	TPWH	(NW)	▝▛	✉️ envelope invitation

The pattern for the prefix con- plus a consonant can be thought of as K alone when the consonant is a non-labial non-continuant, and K overlapped with the consonant otherwise. This is because in the former case, consonant chords already overlap with K. Alternatively, it can be thought of as dropping the consonant after K when it is a non-labial non-continuant.

		Labial		Coronal		Velar
Non-continuants	Nasal		KPH comm-		K conn-
	Stop	KP comp-	KPW comb-	K cont-	K cond-	K conc- conk-	K cong-
	Affricate				K conj- cong-³
Continuants	Fricative	TKP comf- conf-	KW conv-	SK cons- conc-³
	Liquid			KHR coll-	KR corr-

As usual, the asterisk is used for conflict resolution here, but K* can also be thought of as representing a doubling or strengthening of K for words that would require two Ks, such as concord KO*RD (≠ cord KORD) or concave KA*EUF (≠ cave KAEUF).

More complex compound starters?

Recently thinking about adding a few more starters, inspired by wooningeire. Most seem rare though (not counting productive prefixes like in-). My dictionary contains some ad-hoc solutions, e.g. anthology SPWO*LG (INT*OLG) that try to get away with dropping consonants, e.g. enthusiast ^THAOUFT (^THUUſT), which I mostly dislike.

Type	Starter	Steno	Explanation	Example
MF	amph- emph-	KPWH TPW	KPW + H	amphibious, emphatic
NTH	anth- enth-	SPWH	SPW + H	anthem, anthology, anthropology, enthrone
NKP	incap- incomp-	^TKP	^TKP (P reused)	incapable, incapacitate, incompatible, incompliant
NKSP	inconsp-	^STKP	^TKP + S	inconspicuous
NKV	inconv-	^TKPW	^TKP + W	inconvenient
NCH	ench- incoh-	^TKPH	^TKP + H	enchant, incoherent

Caveats

Sometimes the dictionary includes:

• shifty inversions, such as album ^PWHRUPL (^BLUM) or abduct ^TKPWUBGT (^DBUKT)
• fingery conveniences, such as elected ^HREBGD (^LEKD) for canonical ^HREBGTD (^LEKTD) – as per the main.json entry AOE/HREBGD (EE/LEKD)
• misstrokes actually just manual data entry errors at this point
• generous brainfarty fallbacks, such as active ^TWEUF for canonical ^TW (core of the activ- word family)
• redundant redundancies that I’ll never actually use but are there for completionism’s sake, such as upon ^POPB (^PON)
• opinionated briefs, such as pretty much most of the dictionary (I had an example but it slipped my mind)
• or orphans, such as apple ^PHR (^PL) because before I got deep into this whole rabbit hole I tried defining it as WAPL although I guess that conflicts with wam so it was cute while it lasted

Stressed pre-initials

Sometimes the conceit is stretched to include any pre-initial vowel, even stressed, if convenient, such as era ^RA, or if it would be annoying to write out. But words like efficacy are not really in scope. Without a lot of repetition, I’ll usually reach for briefs along the lines of APLZ rather than ^PHAFPB for Amazon due to the stress mismatch. (I’ll move these into caret-stressed.json, okay)

Words without initial stress take precedence, so even the misstroke eleventh ^HREFPBT precedes elephant.

Word families are kept together in the core part of the dictionary, even if one word happens to be stress-initial. For example, illustrate and illustrative. Similarly, words with ambiguous initial stress are kept in core, such as apparatus.

Word family starters

Note that word family starters usually contain the primary stress (e.g. elèctr). This is different from most other caret entries, which conventionally have unstressed starters.

Starter	Steno	Pseudosteno	Example
activ	^TW	(^TW)	activate ^TWAEUT activity ^TWEUT
autom autonom	^TPWH	(^TWM)	automate ^TPWHAEUT autonomy ^TPWHEU
electr	^TWR	(^TWR)	electron ^TWROPB electric ^TWREUBG

Conflict theory

Tries to use * to more or less resolve conflicts based on an informal hierarchy of differently colored schwas:

• Spelled with a etc. or pronounced with a low-mid vowel or schwa (“schwa”) /ə æ ʌ/ etc.
• Spelled with e, i, etc. or pronounced with a front vowel or i-colored schwa (“schwi”) /i ɪ ɨ ᵻ ɛ/ etc.
• Spelled with o, au, etc. or pronounced with a back vowel or u-colored schwa (“schwu”) /ɵ ᵿ oʊ ɔ ɒ/ etc.

So for example:

• addition ^TKEUGS (^DISHUN)
• edition ^TK*EUGS (^D*ISHUN)
• audition would be next in line but alas the overworked * key only allows distinguishing one bit of conflict

Another example:

• a mission would be first in line if this phrase were actually needed, but the two worthier words below take more priority
• emission ^PHEUGS (^MISHUN)
• omission ^PH*EUGS (^M*ISHUN)

The farther the vowel is from mid central, the lower the priority:

• elicit ^HREUS/EUT (^LIS/IT)
• illicit ^HR*EUS/EUT (^L*IS/IT)

But derivations of asterisked words can be unasterisked if there is no longer a conflict:

• allusion ^HRAOUGS (^LUUSHUN)
• illusion ^HRAO*UGS (^L*UUSHUN)
• illusory: both ^HRAOUS/REU (^LUUS/RI) and ^HRAO*US/REU (^L*UUS/RI) because *allusory isn’t a word

The a lot/allot and a lot of/aloft conflicts are resolved by using AU for the latters. Although per the above, allotment is both ^HRAUPLT (^LAUMT) and ^HROPLT (^LOMT).

Not yet consistent, might try to keep word families together. TODO figure out better efficient/officiate and acceptable/accessible

Pseudosteno

When relevant, ^ can be written in pseudosteno as initial a e or o instead. For example, ^DISHUN → aDISHUN, ^D*ISHUN → eD*ISHUN or whatever. I guess ^sT for ext ^ST can work too

*-prefix-word-parts.json

Contains non-word prefixes such as extens ^STENS to make longer words from parts. For example, extensible does not need to be explicitly defined as it can be formed from extens + -able -BL. (Wait, this is a bad example oof)

Though some entries are arguably independent words, such as anon ^TPHOPB (^NON) or admin ^TKPHEUPB (^DMIN), I put them here due to their intentional function of taking suffixes. For this reason, these entries are not literally prefixes of the form {blah^}. Those affixes (prefixes and suffixes) are in *-affixes.json.

*-phrasing.json

Multi-word entries that could be considered part of a more memorization-heavy “phrasing” system.

Half the entries are just short mandatories + ^ that give the word a appended on either before or after, such as a bit ^PWEUT (^BIT) or from a ^TPR (^FR)

Highly attested phrasings that seem wholly uncontroversial, such as akin to ^KEUPBT (^KINT) or according to ^KORGT, are in the non-phrasing dictionaries instead.

*-product-names.json, etc.

Keeps words with stylized official spellings in the same place to make it easy to review, find patterns, etc.

caret-as-s.json

Shape-based outlines that relied on top S have been redefined with ^ in place of S

caret-trying-out.json

Testing ground for some guys I’m trying out, will see how it feels

caret-ipa-fingerspelling.json

Fingerspelling dictionary for IPA symbols used in (American) English as well as a selection of commonly used IPA symbols in general. Omits ordinary letters, like s, as they can be fingerspelled normally. Intended to be iconic and useful, rather than comprehensive.

Like ordinary fingerspelling, the left bank is used for consonants and the vowel bank is used for vowels, but the right bank is used for some vowel variants and suprasegmentals as well.

^*- is used for IPA fingerspelling, and ^*-P for suprasegmental symbols. The consonants follow the convention of H for fricative (if the base is plosive) or palato-alveolar (if the base is already fricative), R for retroflex, and HR for palatal (i.e. alveolo-palatal). The vowels follow the convention of -F for turned or barred variants.

IPA	Chord	Base	Explanation
Suprasegmentals
ʰ	^-FPD	-P	-FD (-H)
ˈ	^-PD	-P	-D top
ː	^-PDZ	-P	-DZ top and bottom
ˌ	^-PZ	-P	-Z bottom
Consonants
ɸ	^TP*	TP (F)	F, since ^PH* is used for ɱ
β	^PWH*	PW (B)	B + fricative H
θ	^TH*	T (T)	T + fricative H
ʈ	^TR*	T (T)	T + retroflex R
ɬ	^THR*	T (T)	T + HR (L)
ð	^TKH*	TK (D)	D + fricative H
ɖ	^TKR*	TK (D)	D + retroflex R
ɟ	^TKHR*	TK (D)	D + palatal HR
dʒ	^SKWR*	SKWR (J)	convenient for d TK* + ʒ
ʃ	^SH*	S (S)	S + fricative H
ʂ	^SR*	S (S)	S + retroflex R
ɕ	^SHR*	S (S)	S + palatal HR
ʒ	^STKPWH*	STKPW (Z)	Z + fricative H
ʐ	^STKPWR*	STKPW (Z)	Z + retroflex R
ʑ	^STKPWHR*	STKPW (Z)	Z + palatal HR
χ	^KH*	K (K)	K + fricative H
ç	^KHR*	K (K)	K + palatal HR
ɡ	^TKPW*	TKPW (G)
ɣ	^TKPWH*	TKPW (G)	G + fricative H
ʁ	^TKPWR*	TKPW (G)	ɢ’s uvularity + rhotic R
ʔ	^TKPWHR*	TKPWHR (GL)	GLottal
ɱ	^PH*	PH (M)
ŋ	^TPH*	TPH (N)
ɳ	^TPHR*	TPH (N)	N + retroflex R
ɹ	^R*	R (R)
Semivowels
ʍ	^WH*	WH (WH)
ɥ	^WHR*	W (W)	W + palatal HR
ɰ	^WR*	WR
Long vowels and diphthongs
iː	^AO*E	AOE (EE)
ɪəɹ	^AO*ER	AOE (EE)	+ R
eɪ	^A*EU	AEU (AI)
ɛəɹ	^A*EUR	AEU (AI)	+ R
aɪ	^AO*EU	AOEU (II)
aɪəɹ	^AO*EUR	AOEU (II)	+ R
uː	^AO*U	AOU (UU)
ʊəɹ	^AO*UR	AOU (UU)	+ R
oʊ	^O*E	OE (OE)
ɔəɹ	^O*ER	OE (OE)	+ R
aʊ	^O*U	OU (OU)
aʊɹ	^O*UR	OU (OU)	+ R
ɔɪ	^O*EU	OEU (OI)
Short vowels
ɪ	^*EU	EU (I)
ɪəɹ	^*EUR	EU (I)	+ R
ɨ	^*EUF	EU (I)	+ F for barred
ɛ	^*E	E (E)
əɹ	^*ER	E (E)	+ R
ə	^*EF	E (E)	+ F for turned
æ	^A*	A (A)
æ	^A*E	AE (AE)	(redundant)
ɑɹ	^A*R	A (A)	+ R (redundant)
ɐ	^A*F	A (A)	+ F for turned
ɑ	^A*U	AU (AU)
ɑɹ	^A*UR	AU (AU)	+ R
ɒ	^A*UF	AU (AU)	+ F for turned
ɔ	^O*	O (O)
ɔɹ	^O*R	O (O)	+ R
ɵ	^O*F	O (O)	+ F for barred
œ	^O*EF	OE (OE)	+ F for ligature
ʊ	^AO*	AO (OO)
ʌ	^*U	U (U)
ɜːɹ	^*UR	U (U)	+ R
ʉ	^*UF	U (U)	+ F for barred
ɯ	^PH*F	PH (M)	+ F for turned

user-diacritics.json

Contains words spelled with diacritics (e.g. café) as well as multi-stroke fingerspelling entries for letters with diacritics (e.g. ä = A*/^-FL) based on iconic right-hand shapes. See another precedent of this.

Diacritic	Symbol	Chord	Shape
Acute	´	^-RP	⠐⠁⠀
Grave	`	^-PG	⠀⠑⠀
Circumflex	ˆ	^-RPG	⠐⠑⠀
Háček	ˇ	^-FBL	⠈⠊⠀
Macron	¯	^-FPL	⠈⠉⠀
Breve	˘	^-FRBLG	⠘⠚⠀
Tilde	˜	^-RPBL	⠐⠋⠀
Umlaut	¨	^-FL	⠈⠈⠀
Ring	˚	^-P	⠀⠁⠀
Slash	̷	^-RL	⠐⠈⠀
Cedilla	¸	^-RPB	⠐⠃⠀

user-*.json

I mean sure? but probably of little interest until Chordial exists

user-main-overrides.json

Cause I really don’t like these defaults. Also includes some prefixes, such as zee SAO*E (ZEE) or ob OB (OB), but I put them in this file because the intent is entirely to cover up annoying defaults that confuse me when typing stroking outlining writing longer words

user-punctuation.json

“Sentence” means sentence-contextual (built-in feature that attaches punctuation to preceding word). Add * to the first four (Period, Comma, Question, Exclamation) for unspaced versions.

I envision ? KW-PL and ! TP-BG as shape-based complements of each other: the shape of KW-PL ascends from left to right like a question’s intonation rising ↗︎, and the shape of TP-BG descends like the inflection of an exclamation falling ↘︎. (See also)

Code	Command	Spacing	Steno	Shape	Explanation	Plover default
{.}	Period	Sentence	TP-PL	⠈⠁⠀⠀⠉⠀	middle+ring, also FuLL stoP	yes
{,}	Comma	Sentence	KW-BG	⠐⠂⠀⠀⠒⠀	low mnemonic for descender	yes
{?}	Question	Sentence	KW-PL	⠐⠂⠀⠀⠉⠀	shaped like rising intonation, also quasi-p. QUestion Mark	yes
{!}	Exclamation	Sentence	TP-BG	⠈⠁⠀⠀⠒⠀	shaped like falling intonation, also quasi-phonetic BanG	yes
{^:^}	Colon	Unspaced	TH-FL	⠈⠈⠀⠈⠈⠀	dot pairs, on upper row like .	KHR-PB (CoLoN)
{:}	Colon	Sentence	^P-PT	⠁⠁⠀⠀⠁⠁	same but further out	STPH-FPLT
{^;^}	Semicolon	Unspaced	KR-RG	⠐⠐⠀⠐⠐⠀	dot pairs, on lower row like ,	TH-FL
{;}	Semicolon	Sentence	SW-BS	⠂⠂⠀⠀⠂⠂	same but further out	STPH*FPLT, SP-PT
{^…}	Ellipsis	After	SW*	⠂⠂⠂⠀⠀⠀	three dots on lower left row	HR-PS (eLliPSis)
{^_^}	Underscore	Unspaced	KWR-RBG	⠐⠒⠀⠐⠒⠀	low line, index-middle-ring	RUPBD (UNDeR)
{^-^}	Hyphen	Unspaced	H-F	⠀⠈⠀⠁⠀⠀	tiny line across upper row	H-PB (HypheN)
-	Dash	Spaced	PH-FP	⠀⠉⠀⠉⠀⠀	short line across upper row	H*PB
–	En dash	Spaced	TPH-FPL	⠈⠉⠀⠉⠁⠀	longer line across upper row	TPH-RB (NdaSH)
—	Em dash	Spaced	^TPH-FPLT	⠉⠉⠀⠉⠉⠀	longest line across upper row	PH-RB (MdaSH)

user-punctuation-paired.json

For paired punctuation, such as (opening and closing) brackets and quotation marks.

I prefer outlines based on iconic symbol shapes to outlines based on symbol names. Outlines for the opening/closing square brackets, curly braces, and angle brackets use exclusively the left/right key banks respectively. To avoid conflicts, the main outlines for opening/closing parentheses use both banks, with vowels A and U meaning opening/closing respectively. When I want to type symbols, I don’t want my hands to feel like they’re typing real words, and vice versa.

For aesthetic reasons, the vowel keys in the diagrams below are positioned to line up with the WR and RG keys, although by default the Uni lines them up with R*₂ and *₄R. This layout decision does not apply to other sections of this document.

Code	Command	Spacing	Steno	Shape	Explanation	Plover default
{~|(^}	Paren open	Before	KPAPG	⠐⠅⠀⠀⠑⠀	shaped like parenthesis, left vowel for open	PREPB (PaREN)
{^~|)}	Paren close	After	KPUPG	⠐⠁⠀⠀⠕⠀	shaped like parenthesis, right vowel for close	PR*EPB
{~|()^} {#Left}	Paren both	Before	KPAUPG	⠐⠅⠀⠀⠕⠀	shaped like parentheses, cursor in between	PR*EPBS (meant for code, not prose)
{[^}	Bracket open	Before	TKPA	⠘⠅⠀⠀⠀⠀	shaped like square bracket	PWR-BGT (BRacKeT)
{^]}	Bracket close	After	UPLG	⠀⠀⠀⠀⠝⠀	shaped like square bracket	PWR*BGT
{~|[]^} {#Left}	Bracket both	Before	TKPAUPLG	⠘⠅⠀⠀⠝⠀	shaped like square bracket	none
{[^}	Brace open	Before	SKPWA	⠒⠅⠀⠀⠀⠀	left brace protruding out	TPR-BGT (FRench bracKeT)
{^]}	Brace close	After	UPBGS	⠀⠀⠀⠀⠕⠂	right brace protruding out	TPR*BGT
<{^}	Angle open	Before	TKWA	⠘⠆⠀⠀⠀⠀	left angle tilted 45° CCW ⤺	PWRABG (BRACKet)
{^}>	Angle close	After	UBLG	⠀⠀⠀⠀⠞⠀	right angle tilted 45° CW ⤸	PWRA*BG
{'^}	Quote open	Before	KW-LT	⠐⠂⠀⠀⠈⠁	analog of KW-GS for " open, top row mnemonic for ' form; also quasi-p. QUote LefT	SKW-T (Single QUoTe)
{^'}	Quote close	After	KR-LT	⠐⠐⠀⠀⠈⠁	analog of KR-GS for " close, top row mnemonic for ' form; also quasi-p. Quote RighT	SKW*T, -PBT

Some alternate versions:

Code	Command	Spacing	Steno	Shape	Explanation
{~|(^}	Paren open	Before	STKPWA	⠚⠇⠀⠀⠀⠀	shaped like thicker paren
{^~|)}	Paren close	After	UPBLGS	⠀⠀⠀⠀⠟⠂	shaped like thicker paren
<{^}	Angle open	Before	TPWA	⠈⠇⠀⠀⠀⠀	left angle tilted 135° CW ⤸
{^}>	Angle close	After	UPBL	⠀⠀⠀⠀⠏⠀	right angle tilted 135° CCW ⤺

Add * for unspaced versions.

Code	Command	Spacing	Steno	Shape	Explanation	Plover default
{^}({^}	Paren open	Unspaced	STKPWA*	⠚⠇⠀⠂⠀⠀	shaped like thicker paren	P*PB (PareN)
{^}){^}	Paren close	Unspaced	*UPBLGS	⠀⠀⠂⠀⠟⠂	shaped like thicker paren	P*PBT (PareNThesis)
etc.

user-commands.json

Retroactives

Left side of the board is mnemonic for “retroactive.”

Code	Command	Steno	Pseudosteno	Shape	Explanation
{*+}	Repeat Last Stroke	#	#		wiki default
{*}	Toggle Asterisk	#*	#*		wiki default
{*?}	Retro Add Space	#SK	i.e. #SK	⠒⠄⠀⠀⠀⠀	horizontal is mnemonic for add space
{*!}	Retro Delete Space	2K	i.e. #TK	⠘⠄⠀⠀⠀⠀	vertical is mnemonic for no gap
#NOOP		2K3W	i.e. #TKPW	⠘⠇⠀⠀⠀⠀	used for notes-to-self

Spacing control

Need to add backspace

Code	Command	Steno	Shape	Explanation	Plover default
{^^}	Suppress Next Space	TK-TS	⠘⠀⠀⠀⠀⠃	2nd-furthest columns	TK-LS (DeLete Space)
{^^}	Suppress Next Space	TK-LG	⠘⠀⠀⠀⠘⠀	both ring fingers	TK-LS (DeLete Space)

Capitalization

I'm going to change most of these into shape-based briefs

Code	Command	Steno	Pseudosteno	Explanation	Plover default
{*-|}	Retro Capitalize	KPA*D	KPA*D	+ -D past tense	yes
{>}	Next Lower	HRO*ER	LO*ER	LOwERcase	yes
{*>}	Retro Lower	HRO*ERD	LO*ERD	+ -D past tense	yes
{<}	Next All Caps	KPA*L	KPA*L	CaP ALL	yes
{*<}	Retro All Caps	*UPD	*UPD	UPpercase + -D	yes
{MODE:CAPS}	All Caps On	KA*PS	KA*PS	all CAPS	yes

OS cursor movement and shortcuts

Code	Command	Steno	Pseudosteno	Explanation
{#Control_L(a)}	⌃A (like Home on Mac)	^H*FT	^H*FT	* + SH-FT for Home, with ^ for top S
{#Control_L(e)}	⌃E (like End on Mac)	SR*RS	SR*RS	* + SR-RS for End
{#Shift_L(Tab)}	⇧⇥	STA*B	STA*B	Shift + Tab
{#Super_L(f)}	⌘F	KPH-F	KPH-F
{#Super_L(s)}	⌘S	KPH-S	KPH-S

Plover dictionary and GUI/plugin manipulation

Code	Command	Steno	Pseudo/Shape	Explanation	Plover default
{PLOVER:LOOKUP}	Lookup	HR-FR	⠀⠘⠀⠃⠀⠀	both index fingers	PHR*UP
{PLOVER:ADD_T…}	Add Translation	PWHR-FRPB	⠀⠛⠀⠛⠀⠀	both index+middle	TKUPT
{PLOVER:FOCUS}	Show Plover	PHROERB	PLOESH	PLover SHOW	PHROFBGS
{PLOVER:FOCUS}{# Super(Shift(K))}	Open Word Tray†	PHRORTD	PLORDT	PLover wORD Tray
{PLOVER:FOCUS}{# Super(Shift(L))}	Open Spectra Lexer	PHRERBGT	PLERKT	PLover spEKTRa
{PLOVER:FOCUS}{# Super(Shift(S))}	Open SVG Layout†	PHROFGS	PLOSFG	PLover SVG
=wt_prev_page	Word Tray Prev	#-RB	⠀⠄⠀⠐⠂⠀	plugin-suggested
=wt_next_page	Word Tray Next	#-GS	⠀⠄⠀⠀⠐⠂	plugin-suggested
=wt_reload	Word Tray Reload	#-RBGS	⠀⠄⠀⠐⠒⠂	plugin-suggested

† To use, patch word_tray_ui.py by adding SHORTCUT = "Ctrl+Shift+K" in class WordTrayUI and SHORTCUT = "Ctrl+Shift+S" in layout_ui.py

License

Code in plugins/, third-party-dictionaries/, and main.json is not my own. It’s just version controlled here to record my modifications in order to prevent any updates from overwriting them.

OERZ TKO WHAFRT TPUBG UPT

^TR-BGS WELG

Contributing

KR-BGS WELG

TODO

• Document repo contents
• Methodology (how I went about finding what words to include)
• Split user.json into phrasing, plurals/edge-cases, prefixes, glaring-omissions
• Add one-liners, scripts, exploratory research (with documentation)
• Dictionary generator/preprocessor to define e.g. accustom in terms of ^ + {custom}
• If I ever do all that I’ll name my baby theory plonkver thoeur (pronounced /θwɑːɹ/)

See also

• Declarative open phrasing engine, my WIP rewrite of Jeff phrasing