I'm trying to use #ChatGPT to generate the vocabulary for #GUCAC. It seems like this should be within ChatGPT's wheelhouse, so I may just be asking the wrong questions.

ChatGPT doesn't seem to grok binary or exponents very well. I tried bit positions (5th bit) and hexadecimal (0X20), but 00100000 seems to work best. ChatGPT still isn't right in terms of these words' associations with the taxons; 'Swimming' is a gerund with little in common with meaning and emotion, thought, relationships, or groups of things.

(This will be the only post with the ChatGPT tag, follow #GUCAC if curious)
#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

(this is me)
Create a taxonomy using the 256 most common unique singular nouns and gerunds that follow this fuzzy pattern.
Precede each word with its hexadecimal number starting with zero.
Follow each word with a list of the bits that are on.
In the taxon descriptions below, "more related" means that this word is in the top 50% of the 256 words, in terms of its relation to the taxon via word cosign vectors or similar measures.
The zeroth word is null.
The words will be arranged in an eight dimensional binary array as follows:
Words with bit 10000000 are acts (gerunds).
Words without bit 10000000 are singular nouns.
Words with bit 01000000 are more related to states of being.
Words with bit 00100000 are more related to physical objects.
Words with bit 00010000 are more related to time.
Words with bit 00001000 are more related to meaning and emotion.
Words with bit 00000100 are more related to thought.
Words with bit 00000010 are more related to relationships.
Words with bit 00000001 are more related to groups of things.

(ChatGPT: Close, no cigar)
Creating a list of 256 unique common words that follow this fuzzy pattern is a challenging task, but here's a sample list of words that fit your criteria. I've provided the hexadecimal numbers and the corresponding bit descriptions for each word:

```
0x00: bits 00000000
1. null

0x01: bits 00000001
2. community

0x02: bits 00000010
3. thought

0x03: bits 00000011
4. existence

0x04: bits 00000100
5. history

0x05: bits 00000101
6. mountain

0x06: bits 00000110
7. emotion

0x07: bits 00000111
8. running

0x08: bits 00001000
9. collection

0x09: bits 00001001
10. friendship

0x0A: bits 00001010
11. contemplation

0x0B: bits 00001011
12. happiness

0x0C: bits 00001100
13. time

0x0D: bits 00001101
14. river

0x0E: bits 00001110
15. presence

0x0F: bits 00001111
16. swimming

0x10: bits 00010000
17. assembly

0x11: bits 00010001
18. connection

0x12: bits 00010010
19. reflection

0x13: bits 00010011
20. memory

0x14: bits 00010100
21. today

0x15: bits 00010101
22. forest

0x16: bits 00010110
23. perception

0x17: bits 00010111
24. dancing

...

(Continuing with similar patterns for all 256 words)
```

Please note that generating 256 unique common words that fit the pattern exactly can be challenging, and you may need to adjust or expand the list to meet your specific requirements. Additionally, the categorization of words into different bits is subjective and may require further refinement based on your needs.

Revisited our reviews of an unaccepted reproduction study to a German #HCI conference. While most criticisms were minor, one critique was our use of what appeared to be an outdated method. Upon digging deeper, we found that the referenced handbook had been silently updated online, but without clear versioning or change logs. It's a challenge both for reviewers and us when foundational resources aren't transparent. Now we're left revisiting our work's foundation. #phddiaries #academics

We had a great time at our #datacollection on #StatsraadLehmkuhl last night. We sailed from #Bergen, past #Askoy and towards #Oygarden in great weather.

The bridge crew got to test the #HoloLens #AR application that Floris van den Oever has developed for his #PhD. The crew found the interface to be intuitive, and saw the potential in having information about other ships as overlays in their visual field. Looking forward to more field testing!
#HumanFactors #HCI #MaritimeWork

The ACM Automotive User Interfaces Conference takes place in Ingolstadt next month.

https://www.auto-ui.org/23/

#hci #ux #automotive #conferences #ingolstadt #germany

#GUCAC 'Global' Translator (as a Language Interchange Code)

GUCAC stands for Global Universal Control And Communication. In order to be global, It must provide the ability to communicate across many languages. This will be partially accomplished by providing duplicate vocabularies in 9 widely used languages (Mandarin, Spanish, Hindi/Urdu, Arabic, Bengali, Portuguese, Russian, and Japanese). The current prototype language structure is probably too English-centric, but we have to start somewhere.

When a GUCAC message is received, the sent terms will be replaced by the terms in the receiver's chosen language. The receiver's chosen language will be determined during the initial handshake between the GUCAC main server and the receiver. The vocabulary will then be served by a language specific server (or multiple language servers). GUCAC will have it's own syntax and grammar so it will probably sound 'odd' in most languages. Thus, translation will be done with simple replacement, but formatting the initial sentence and parsing the resulting sentence will be left as a task for the users (or, probably some local AIs).

Now, nine languages is hardly 'Global', but the vocabulary for the common conversational vocabulary only requires the first three chords, and only contains about 20 thousand base terms including three thousand phrases. These terms will actually be contained within the user's device. This part of the vocabulary can be created in any language and uploaded to the device. In this way, users can have their native language available for simple conversations and still be able to use the full vocational vocabulary in one of the nine global languages for more technical terms.

This process can also be used for the control division (potentially 65 thousand terms) and the menu (1.7 gigaterms) so that users can access the blocks more easily. The servers that contain these translated terms will be hosted in the countries where those languages are spoken. The effort of translation will be crowd-sourced to increase vocabulary ownership and reduce the load on vocational vocabulary maintenance.

The vocational vocabulary will have regions (areas, divisions, or blocks) marked as modifiable and regions marked as static. The modifiable vocabulary will be subject to change at any time and will include things such as links to commercial sites (which will be leased to support the central servers) and personal space on one's device.

Once the initial static vocabulary has been created, static vocabulary changes should only be promulgated once a year to maintain vocabulary stability. Terms which become obsolete will be saved in a historical archive. Each vocabulary will have a unique version number, and the archive will be used to translate terms from previous versions of the vocabulary.

#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

The #GUCAC Reference Application

Since GUCAC describes a content server; a client app will be needed to handle the mechanics of serving those contents to the user; and the client will also provide facilities for searching, message exchange, user training; and block downloading, uploading, and development. Initially the client app will just serve text.

An open source reference app that runs on desktop Linux will be created to provide a template for developing apps for other platforms. Developers will be encouraged to create proprietary Apps for other platforms and operating systems. The reference app will be available for anyone to use freely, so the proprietary apps should add additional features for usability and user entertainment.

When the security issues have been ironed out, the reference app will also provide facilities for program and script, development and execution. The proprietary apps will have to avoid this area of development until security standards for programming have been developed in the reference app. After that, developers can create competing integrated development environments (IDEs).

#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

The #betwixt #game is amazing. Put experts from #cognitivetherapy, #mindfulness, #CBT, #sciencewriting, #AIcreativity, #hci, and #gamingforwellness into a pot and shake them up, and this is what comes out: a narrative choose-your-own #adventuregame with a strong connection to existing and ongoing research, driven by your own #selfexploration. I am fascinated and learning a lot about myself.

Come for the story, stay for the free therapy. 😃

https://www.betwixt.life/
https://www.betwixt.life/research

We are hiring a Lecturer or Senior Lecturer (Assistant/Associate Professor) in Cyber Security and Digital Harms. We are particularly interested in candidates with research interests in the human side of things (recent hires have been technical) but feel free to talk to us if you have any interest in the role and do pass on to others who you think might be interested: https://www.jobs.ac.uk/job/DCB581/lecturer-senior-lecturer-in-cyber-security-and-digital-harms-550711 #CyberSecurity #Jobs #UKHE #HumanFactors #Cybercrime #CyberLaw #UsableSecurity #HCI #Cyberpsychology

New paper out in #humancomputerinteraction journal: A sensemaking system for grouping and suggesting stories from multiple affective viewpoints in #museums 50 days free access: https://www.tandfonline.com/eprint/AHINMIJGM93A9EM4IQFU/full?target=10.1080/07370024.2023.2242355

#hci #recommendersystems #recsys #affectivecomputing #storytelling #culturalAI

@academicchatter
@cscw @digitalhumanities

https://tandfonline.com/doi/full/10.1080/07370024.2023.2242355

Hey social science research geeks with a niche interest in obscure human computer interface studies, help me out here.

Has there every been research done on whether the avatar a person is using for an online account influences their online personality and posting style?

#research #SocialSciences #ComputerStudies #HCI

#GUCAC Fuzzy Thematic Sorting

A block of 255 general terms will be sorted using word-vector cosine-similarity as determined by the program word2vec. Each general term will be ranked against each taxon, and the term will get a 1 if it is "more related" to the taxon or it will get a 0 if it is "less related" to the taxon. Half of the terms will be more related and the other half will be less related. Thus, relatedness will be relative rather than absolute. The customary menu term in location 0X00 will not be affected by sorting.

The process will start with twenty or so candidate taxons. A candidate taxon will be selected by the block-developer because of its relationship to some of the terms in the block, and because of its affinity with a trigram in the I Ching.

In the diagram below, we show the result of sorting. Note that the relatedness of taxon7 alternates as 0 1 0 1 0 1, while the relatedness of taxon6 alternates as 0 0 1 1 0 0 1 1. Each successive column will then double the series of ones and zeros so that the taxon0 column will have 128 zeros followed by 128 ones. By looking for these patterns, we can determine which candidate taxons should be final taxons and what their order should be.

This process will be iterative, because the candidate taxons are unlikely to evenly partition the terms on the first try. The developer will probably need to replace any initial taxons that are too similar in their relationships to the terms, with new taxons that are more orthogonal in their relationships. This process is called "Computer Aided Thematic Sorting" or CATS.

Once the final taxons have been selected, any term that doesn't have the correct relationship (either 1 or 0) with a taxon will be adjusted by swapping it with another term that is a better match, or by simply changing the bit's value. The relationships between terms and combinations of taxons are intended to be a mnemonic aid rather than an absolute correspondence, so they may be extremely fuzzy. The relationships don't have to make total sense as long as they are memorable.

The final taxons (the columns) will then be arranged in the order of the trigrams of the I Ching. This will require a change the order of the terms (the rows), but this is the easy part. For example, if the term that was one is now 64, we just copy it to location 64 in a new table.

Finally, the taxons will be copied to the menu term's hints or keywords so that they can be displayed by a GUCAC app to help the user find a given term. The menu term must also given a unique name so that this block can be found by a text search of the menu. The menu term may also contain instructions for using this block.

#algorithm (This will be the only post with the algorithm tag, follow #GUCAC if curious)
#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

#GUCAC Thematic Gradient (this is the fun part)

A thematic gradient will be created within a block using the eight trigrams of the I Ching (https://en.wikipedia.org/wiki/I_Ching). The trigrams (https://en.wikipedia.org/wiki/Bagua) are a situationally-fuzzy generic ontology that can be adapted for roughly classifying the terms.

From the user's point of view, the meanings of the taxons will be added together to describe the term found at that address. Every unique combination of taxons represents one or more words which can be considered an amalgam of their meanings, as shown below (see the attached image). Self plus time is life. Self plus thought is mind. Self plus time plus thought is agenda. Time plus thought is history. These relationships are intended to be a mnemonic aid rather than an absolute correspondence, so they may be extremely fuzzy.

From the developer's point of view, 255 terms will be sorted into taxon order, and then the taxons will be arranged so that they correspond to the ordered trigrams. The fuzzy sorting algorithm will be discussed in the next toot.

The trigrams have been arranged as follows: the passive, internal, or Yin trigrams have all been assigned to the left hand, while the active, external, or Yang trigrams have been assigned to the right hand, assuming that most right hands are more dominant than left hands. These two groups of trigrams have then been arranged vertically to provide a situational control gradient. This organization is intended to help users to associate the binary weight of the trigrams with their fingers.

The weight of each trigram is defined as follows.
Right hand
Trigram Finger Weight Fuzzy Meaning
Supremacy (pointer) 2^7 = 128 You are in control
Moving (index) 2^6 = 64 You are expanding your territory
Resisting (ring) 2^5 = 32 You are fighting a superior force
Peril (pinky) 2^4 = 16 You have lost and are avoiding danger
Left hand
Elegance (pointer) 2^3 = 8 You are a master of some art
Learning (index) 2^2 = 4 You are developing your ability
Flexibility (ring) 2^1 = 2 You are evading control
Submission (pinky) 2^0 = 1 You are being controlled

The explanations of the trigrams above are oversimplified for the sake of clarity. The actual situations may be physical, mental, social, etc; and the subject may be something other than you. I have found that these archetypal taxons may be used to differentiate between many types of related things. The taxon names may be replaced with domain-appropriate names in a given block. These replacement names will be saved in the block's menu term.

You may notice that James Legge's original translation of the Standing trigram has been changed to Resisting and Pleasure has been changed to Learning for this ontology. This is intended to make the trigrams easier to understand and use. Likewise, the weights of the trigrams does not match the classical values of the broken and unbroken lines.

#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

#GUCAC blocks

Each eight finger chord will provide access to a 'block' of 256 terms. The blocks accessed with 1XFF, 2XFF, and 3XFF will be found in non-contiguous locations, but they will appear contiguous to the user. The block will be the user's main method of selecting terms. Handspace can be conceptualized as a collection of 2^24 blocks.

Mechanically, #GUCAC just creates a 32 bit index for a content server. Since a content server can deliver many things, a term can be a word or a phrase, a function, a sound, an image, a variable, a switch, a data channel, a link or a URI, or it may serve some other purpose. Thus a block of terms may be a vocabulary of words and phrases, a toolbox or API of functions, a control panel of switches, a dashboard of data readouts, a folder of bookmarks, or it may provide some combination of these uses.

I have also created a rough design for a first person shooter video-game interface block, which uses the left hand (0x0F) for movement and the right hand (0xF0) for manipulation. When the hands are used separately, each term simply contains independent right and left hand functions, and those individual functions are duplicated in any term that matches the corresponding right or left finger configurations. Thus, not all blocks must be configured as 256 individual terms, because a block can be memory mapped as any interface that can use eight buttons.

#gaming (This will be the only post with the gaming tag, follow #GUCAC if curious)
#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

#GUCAC's menu

The GUCAC integral menu is created by the simple expedient of making the first term in any zero level block a "menu term". A menu term describes the general use of the other terms in it's block and provides 8 taxons that give a thematic weight to each of the eight fingers. This means that as long as the low order chord is zero (0X00), you are viewing the menu.

The top level of the menu for the specialization vocabulary (3XFF) will be organized thematically. In the GUCAC prototype I used OSHA’s Standard Industrial Classification (SIC) codes, and the U.S. Equal Employment Opportunity Commission’s Job Classification Guide. These sources were used on the assumption that vocational domains would have various sets of common and unique specialized terms, and that these groupings would therefore provide the most complete, equitable, and useful categorizations of technical terms.

The second menu level will also be organized thematically. So the top level might be Paleontology and within that, the second level might be Theropoda. This would provide 64K of menu headings.

The third menu level will be organized as standardized categories of blocks, such as introductory, intermediate, specialist, analytical, exploratory, vocational, recreational, vocabulary, toolbox, dashboard, control panel, or some combination of those and other usage oriented descriptions.

There are a few 'areas' where the menu is not needed or used. An area is the top division within 'handspace'. Handspace is the full 4 gigaterm address space indexed by GUCAC. The full breakdown is as follows:

handspace = 2^32 terms - handspace contains 256 areas
area = 2^24 terms - each area contains 256 divisions
division = 2^16 terms - each division contains 256 blocks
block = 2^8 terms - each block contains 256 terms or 16 banks
bank = 2^4 terms - each bank contains 16 terms (addressable with one hand)

Some areas that will not use the menu:
Integers - one area will contain 2^24 positive integers (and possibly 2^24 negative integers)
Unicode - one area will contain Unicode characters in several languages. To save space, characters duplicated in several code pages (like 'a') will only occupy one term.
Conversation - the zeroth area will contain basic words used to construct sentences. This is expected to be the most used area because it only required three chords or less.

#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

@eglassman I also just submitted a grant, so I know a bit about how it feels. The team and idea behind it are great, so lets hope the reviewers agree. Yet, it is always a nightmare to just press the submit button. In the end you feel you just have to get it over with... Good luck with the grant application.

#hci #ux #academia

How #GUCAC works

You can't access 4 gigaterms with 10 binary, on/off buttons without multiplexing some bits. In this case the eight fingers are multiplexed with the two thumbs. In other words, you will need to enter four eight-finger chords, which are numbered in binary with the thumbs. The chords are specified as 0XFF 1XFF 2XFF 3XFF (which is valid hexadecimal notation, it's just seldom used). So, 8 bytes (one nibble for each hand) is 32 bits. Mathematically (2^8)^4 = 2^32 which gives us 4 gigaterms.

For those who are thinking that nobody can remember four eight-finger chords, that is why the #GUCAC protocol specifies an integral menu (described in the next toot). In fact, the remainder of this discussion will be mostly dedicated to explaining the design choices that make #GUCAC usable by individuals who lack an eidetic memory.

The major design choice is that terms requiring the first three chords will be used for the common conversational vocabulary, while terms requiring all four chords will be used for the specialization vocabulary, which contains everything else. The forth chord (3XFF) will contain the top level of the menu for the specialization vocabulary.

The first chord (0XFF) will contain the categorical terms. The second chord (1XFF) will contain metadata terms which can modify other terms or entire sentences. The first and second chords together will expand the categorical terms into basic terms which will be nouns, adjectives and adverbs, and 3036 phrases. The phrases will cover social, business, and scientific situations. The third chord (2XFF) will contain the syntactic terms, which will include prepositions, conjunctions, possessives, etc. The syntactic terms are used in combination with the basic terms to create predicate clauses. The second and third chords together will define the control division where useful personal blocks can be created, and where blocks from the specialization vocabulary can be saved for easy use.

Thus, the user can focus on a single chord at a time, even when that chord is being used in combination with higher or lower order chords.

#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource

Embracing Introverted Strengths, Part 1
From UX Matters

"Many UX professionals are naturally introverted, which likely contributes to their not achieving the same career growth as other more extroverted professionals within an extrovert-biased corporate environment."

https://www.uxmatters.com/mt/archives/2023/07/embracing-introverted-strengths-part-1.php

#ux #hci #jobs #ucd #uxmatters #introvert #extrovert

McCall, Rod, and Lynne Baillie: #ethics #privacy and #trust in #seriousgames

Downloadable chapter on issues relating to serious games.

https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=6e30eeae5d995580792f991e2be7b9dbe409b543

#citeseerx #papers #openscience #games #research
#hci #ux

Introducing #GUCAC

GUCAC (Global Universal Control And Communication) is a human/computer interface that uses 10 binary, on/off buttons to access over four billion things. I will be using the #GUCAC tag to explain how this works and what the ramifications might be.

For anyone familiar with binary math, I need to tell you that I am very well aware that 2^10 = 1024. Also, I have explained this to friends that are familiar with binary math and they have all agreed that this interface can obviously access 4 gig of things (which are called 'terms' in this model).

I will explain how this all works in the next #GUCAC post so that you can try to deduce how the interface operates without any spoilers. Good luck!

Incidentally, GUCAC currently only exists in my mind, so welcome to the ground floor (or more likely, the basement) of the future.

(I'll also be trimming the tags below as they become irrelevant until only #GUCAC remains)
#tech #technology #interface #HCI #data #computer #computers
#internet #webdev #opensource
#introduction

@julian no, you are right, I am not joking. I suspect most people I follow are #journalists or #retrocomputing people, the third group being #hci #ux . I intend to use #mastodon in a constructive way... i.e. for good discussions, helping others and importantly to learn something new. Rather than engaging in bashing.