Something about hammers and nails and using the right tools for the job

This sounds great if you are very literate and well-vocabularied.

Most people in the world are not. Or don't speak English at all, another problem with text that you don't find with images.

There is a saying "a picture is worth a thousand words". The brain will recognize a white flurry cat from an image faster than from a phrase "white flurry cat".

> But let's hit the random button on wikipedia and pick a sentence, see if you can draw a picture to convey it, mm?

To be needlessly pedantic, my computer drew the image that conveyed this to me.

Less pedantically text is a medium of exchange for language, and a lossy one just like spoken word. I think there's a lot of power in its flexibility due to that lossiness. It's also one of its subtle weaknesses - we can read text from 5,000 years ago, but there's going to be much debate over understanding the text because of how much context has been lost to time.

As someone who has produced UML diagrams for a living for some time, this really talks to my soul... Thanks for posting.

the phonetic alphabet is a brilliant invention. made by thousands of people over thousands of years.

the least studied 'character' of this alphabet is the most imporant, the most critical character in any phonetic alphabet is the blank space. withoutitphoneticwritingsdoesnotmakesasmuchsesnse.

Obligatory reference to the Unix Koan "Master Foo Discourses on the Graphical User Interface", after reading this post:

> Master Foo said nothing, but pointed at the moon. A nearby dog began to bark at the master's hand.

> “I don't understand you!” said the programmer.

http://www.catb.org/~esr/writings/unix-koans/gui-programmer....

When breathing, always bet on air..

The day will come when we can think an image and then send it. It might change things.

> "Human rights are moral principles or norms that describe certain standards of human behaviour, and are regularly protected as legal rights in national and international law."

image: https://imgur.com/a/8J6yeRe

4000 bytes? That tweeting bird picture really is worth a thousand words ;).

Nitpick:

> It can be translated.

It needs to be translated. No such issue with photos.

It depends, memes, short clips are super effective too.

And yet I can find toilets easily in any country with a simple logo but have to hunt for them if they are spelled out.

Images are wonderful if you don't live in a bubble where everybody speaks the same language.

Dumb post. It's not a contest. Use whichever is appropriate.

I design systems using both diagrams and text sometimes laid out in understood, non-linear patterns. I also enjoy music.

How is neuralink not being discussed here?

What is that image of Gnus doing there?

Ludwigstein cried after reading this

A picture is worth, like, 100 words

Yet Instagram is flourishing.

(2014)

I think text is a powerful mean to express ideas, but pictures are also. And it comes down to how people think: in text or in pictures?

I don't know much about it, but there in a talk, Jordan Peterson addressed that there are people who can think in text, in pictures or even both. And that trait says a lot about ones character.

Interesting stuff

Gopher vs HTTP.

One was only text.

> assuming we treat speech/signing as natural phenomenon -- there are no human societies without it -- whereas textual capability has to be transmitted, taught, acquired

I’ve been watching videos of pets (dogs, cats) being “button trained”—being taught to use (primitive, non-syntactic) language by making associations between things/acts/emotional states and the pressing of one or another audio-playing buttons that has been placed on the floor.

It has very much driven home the point for me, that “spoken language” is actually not something inherent and instinctual to humans (or to any species); but rather is a technology. It’s just a technology that’s rather simple to learn, if you have the right underlying hardware acceleration (e.g. a cerebral cortex)—and, crucially, a teacher. For other intelligent mammals, apparently the teacher is the only component they’re missing!

Language is a technology that humans in particular find very intuitive—at least at a young age when our brains are malleable—but not one we inherently start with. We absorb it easily if we’re immersed in a society where everybody uses language from birth. But in situations where that’s not true (feral children, some very broken homes) we don’t.

In a world where every human being instantly had all entrained structure in their neocortex erased, such that we were “reduced” to being upright hairless apes with the capacity for language but no knowledge of it, I don’t think we’d just instantaneously come up with the idea of language and begin attempting to develop languages to communicate, the way modern people instantly try to develop a creole of the languages they do know, when stuck in a situation with people who share no common language with them.

The idea to associate concepts with specific mouth-noises—and to condition others to use those same mouth-noises for the same concepts, to facilitate transmission of thought—might randomly arise in a few people, but it’d need to catch on and spread from there, just like any other technology. It would either need to be observed and copied, or actively taught.

And I hypothesize that that is what happened (pre)historically: at some point, there were several memetic “waves” spreading increasingly-technologically-advanced (e.g. syntactic, expressive) forms of language across human populations with brains already structurally amenable to them. Of course, each wave would only be a struggle to the generation that pioneered it; the next generation, being immersed in that new, more-complex language form from birth, would find it similarly intuitive.

(This makes me wonder whether we’ve yet hit the “limits of linguistic expressiveness” for our current brain size, such that we’d need to unlimit e.g. average skull diameter at birth to let us get any fancier with language; or whether we’ve still got some, ah, “headroom” left.)

This article annoys me whenever it reappears because it just isn’t clear on what it means by ‘text’.

Sometimes it seems to mean ‘the English language’ or ‘language’ more broadly. Other times it seems to mean ‘strings of ASCII encoded Latin letters’. Sometimes it seems to mean ‘pictures of arrangements of letter like symbols’. In general it just amounts to ‘linear streams of data’.

Sure, if you define text that broadly, it covers a lot of things that are great.

But it’s a definition that’s so broad it defies its own terms. Text, defined that way, encompasses SVG files. Or even base64 encoded PNG files if you want. So that Twitter logo can be unambiguously shared through ‘text’ too. Look - here’s a tweet-sized version: https://twitter.com/bbcmicrobot/status/1237867433064464394?s...

But there’s a weird cultural bias built in to the assertion that all those things are ‘just text’. Sure, for someone who uses a US keyboard to type Latin alphabet characters left to right, base64 encoded binary, svg, or BBC BASIC, is ‘just text’. But that’s not exactly a universal perspective.

In the limit, this amounts to ‘always bet on data transmission and storage’.

A lot of the listed benefits of text are only realizable when the text is coupled with a specific ‘interpreter’ - be that an SVG renderer, a BBC micro, or a human who speaks English.

Doing stuff with text with computers is hard! Lexers, parsers and tokenizers are probably the most common sources of security bugs in history. And if the text is natural language, we still don’t have reliable computer tools for dealing with it - understanding or generating.

So I just guess I don’t really know what the point of this piece is. Data is all there is. Linear streams of data are often a thing. Because of the history of computing, western language character sets and conventions are often used to capture them in the same format as we use for written language.

I do agree text wins when it comes to expressiveness. However, that expressiveness comes at a cost, just look how difficult it is for beginners to grasp the initial concepts of programming.

I've seen some people pick up this medium very quickly, and others struggle for months with little progress. However, almost anyone can pickup Sketch or Illustrator for creating UI prototypes very quickly.

The expressiveness of text is not always a strength. It's very hard to build programming languages without text, but I strongly believe we still program too much when building UIs. Excel demonstrates that people can quickly pickup a minimal programming language for connecting data to UI, I think an Excel-Sketch hybrid is where the future lies for building applications in particular.

This discussion right here is all text. Imagine if it was pictures instead. Imagine representing any of the comments in form of a picture. Won't be one picture representing a 1000 words. May be the another truth is that "a word is worth a 1000 pictures".

Concerns about 3D multimedia spaces replacing text makes as much sense as worrying about parchment, paper, or rectangle screens replacing text. Virtualized spaces are, much like those others, merely a transmission medium for text (and other media.) As immersive computing arrives (you may not think it is, but I assure you, it is) we will be refactoring a lot of these existing rectangle screens and other contexts where text is projected via physical processes into virtualized ones.

This is why in my 3D virtual space for work, Jel, fully collaborative text panels (synchronized via OT) are the primary kind of element you create: https://jel.app.

What I took away from this is: using combinations of a limited set of symbols to convey meaning is durable (as society advances or retreats and technology changes). This makes sense because human thinking, for the most part, seems to be symbolic (i.e. using a smaller, simpler thing to represent a larger more complex thing). Text is just one such form.

Drawings, sculptures, JPG files, videos etc. are more precise and are better at conveying a specific object or event, but not necessarily its meaning.

If I remember correctly, Neal Stephenson's Anathem deals with some of these themes.

The article is making a weird point and the comments here are talking past each other.

Author should be conveying the unique aspects of Textual media but also it’s shortcomings. Comments here should be discussing the benefits and shortcomings of all kinds of media.

Instead both are discussing which one is better - Text or something else. A lot of it is mutually exclusive. The Tianamen square tank man image is powerful and impossible to encode in text with the same effect, and Nabokov’s prose is impossible to paint a picture of or how we struggle to describe tasting notes.

Text is using a series of efficient pictures to communicate. I liken it to radix-based numbers. I could write 1,749 tallies on a piece of paper, or I could write the base-10 number.

I have to learn decimal numbers: my intended audience needs to learn decimal numbers. But the resulting efficacy is a thousandfold. I think of letters and words like digits, but for communicating broader concepts.

So when we lack the symbols or combinations to express something, we need a unique picture to do so. Pictures are worth a thousand words, but only when a thousand words won’t suffice.

There’s a real ‘ceci n'est pas une pipe’ treachery-of-images philosophical trap here, of course.

If text’s so darn great, after all, why do you need to draw a picture of it before I can understand it?

> “Pictures may be worth a thousand words, when there's a picture to match what you're trying to say.”

The author is saying that because pictures cannot easily capture arbitrary sentences, text is better. But the same applies with the roles reversed: text cannot capture arbitrary pictures! Instead of saying that one is better than the other outright, let’s move to media (for reading, composing, and programming) where the two can be intermingled appropriately, with as little friction as possible. We certainly have the technology for it...

Yes, visual diagrams and shapes are sometimes more efficient. But the main point is that text is the lowest (and cheapest) form of encoding that is still human readable/writeable. Text is always there even when other data formats are not feasible. If you can adequately represent your ideas in text, you are giving it the highest chance of being seen and distributed by other humans. You don't need special tools or artistic skills to replicate the data, anyone with a pen & paper or keyboard can do it!

Text and more broadly natural language is actually a stunningly low-bandwidth form of communication. The reason it works so well despite being low-bandwidth is that we humans share an incredible amount of shared experience and common knowledge, and so the limited amount of information in text can refer to a much wider set of knowledge and assumptions and our brains are able to make inferences and use context in a remarkable way to fill in the blanks.

To take an analogy, think of writing text as analogous to compressing data using a Huffman code. Our ideas correspond to the initial uncompressed data, natural language text corresponds to compressions of those source data, while our brains correspond to the Huffman tree that tells you how to decompress. With our brains/context we can recover the initial ideas, just like with the Huffman tree we can recover the uncompressed data. Without the Huffman tree, the compressed data are gibberish.

On the one hand this means that text is really powerful as the author says; we can store an incredible amount of information in a small amount space, and can reconstruct the source idea the text represents efficiently (if with some amount of ambiguity and error).

On the other hand text is very far from universal. Anyone who sees the Twitter logo sees the exact same thing (interpreting it as a bird, of course, requires the prior knowledge of what a bird is and looks like). However, anyone who sees a piece of text not only needs to understand the language it's written in, but also all of the ideas that it refers to. That's why we still have many examples of languages and texts that are undecipherable: we've lost the context they were originally written in. Even Egyptian hieroglyphics were undecipherable until the 1800s when people used the Rosetta stone to provide context to decipher it.

Text has further issues as well, chief among them its ambiguity. Not only is it easy to under-specify things in text, but it's also possible for the same piece of text to mean different things at different times and places.

As a culture these issues may be strengths; poetry and literature derive strength from this openness to interpretation, and many would argue so does law where statutes written centuries ago can be adapted to our time. But from a purely data storage and transmission perspective, these are clear weaknesses.

I think the one weakness of text is context. In 2,000 years, if someone finds an inscription reading "Thanks Obama", what will they make of it? They'll might know that Obama was a president of a country. Without context it would be hard to tell if this was 1) genuine sentiment 2) snarky criticism or 3) ironic. Good luck with "Covefe" or whatever.

My point is that context matters a lot, and (I assume) that it can only be reconstructed through a lot of text.

I’m so glad this post is still around. I cite it a lot when discussing creative tools with people who think that visual is the only way to go.

Obviously the best computer tools offer both visual and textual ways of working. But if you have to pick one, bet on text.

I was about to add a caveat about pure visual tasks, like image composition, but advances like DALL-E are starting to put those tasks into question as well. If I were making a photo editor today, I would bet on text.

I love Emacs and plain text. But claiming that text is everything is ignoring the power of images. The brain can immediately grasp the meaning of an image, visuals are immediately understood, some of the most remarkable things human beings have created is art, that is visual imagery.

Just because our current technology is best able to handle and deal with text says nothing about the power of sound, images, sketches, drawings and handwritten notes.

He uses a mathematical formula as an example of text, with its symbols and whatnot. I don't count that as "text", to me, its closer to a UML diagram or flowchart than it is to the text I'm using to write this comment. If you just conflate "all things represented through glyphs and symbols", then I find that too broad to be useful.

This looks like a specialization of Tim Berners-Lee's Rule of Least Power[1]:

  > suggests choosing the least powerful [computer] language suitable for a given purpose

https://en.wikipedia.org/wiki/Rule_of_least_power

The author seems to be arguing that "most general" = "best" - because text is the most general communication format, it's the best.

This argument is pretty trivially false.

By this logic, the best programming language is assembly, because it's the most general. Actually, writing opcodes directly might be more general, in case your assembler doesn't have translations for some undocumented opcodes. The best editor would be a hex editor. The best web browser would be netcat piped into a hex editor, as well. The best OS would be no OS at all (because OSes impose restrictions in order to make programming easier and safer). The best application, for any kind of application, would be an interpreter that would allow you to create your own, however you wanted.

Engineering is necessarily a tradeoff between generality and efficiency. Technology is largely used to make things more efficient, and so our tools always impose some constraints on the problem/solution space in order to be more efficient than not using that tool.

A common design pattern is a specialized fast-path and slower (but more general) fallback.

In addition, as many others have stated here: "the right tool for the right job". Videos, audio, pictures, and interactive tools will always be more efficient for certain problems. If your sole concern is generality, then yes, by all means, use text. However, this will almost never happen; your design space will almost always necessitate a tradeoff of generality with efficiency - in which case, pure plain text is rarely the solution.

It's a good post, and he makes excellent points.

Like any "hard and fast" rule, however, it tends to go a bit pear-shaped, when viewed from certain contexts.

As someone that has made the mistake of designing a "pure" iconic interface, I can tell you that alternatives to text UI can be quite difficult to implement[0].

But a well-designed symbolic UX can be leaps and bounds more effective than text.

In some contexts.

Basically, YMMV.

The main issue with text, is that is assumes that:

1) Everybody can read, and

B) Everybody is on the same page.

In any given day, I notice written signs everywhere. But the really important stuff tends to be done symbolically.

Notably, caution/danger signs and other warnings.

Road signs are almost always text, but the same thing I just mentioned, applies to important cautionary road signs. You can assume that anyone driving can read (written road test), so why use icons?

That's because we can process symbols much more quickly and effectively than text. A well-designed icon can be instantly recognizable. Take, for example, the classic radiation or biohazard icons.

They still need "training" to properly interpret; but nothing like the level of education required to simply read (and understand) the word "BIOHAZARD."

    ⠀⠀⠀⠀⠀⠀⠀⠀⠀⣠⡖⠁⠀⠀⠀⠀⠀⠀⠈⢲⣄⠀⠀⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⠀⣼⡏⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢹⣧⠀⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⣸⣿⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⣿⣇⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⣿⣿⡇⠀⢀⣀⣤⣤⣤⣤⣀⡀⠀⢸⣿⣿⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⠀⠀⠀⢻⣿⣿⣔⢿⡿⠟⠛⠛⠻⢿⡿⣢⣿⣿⡟⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠀⣀⣤⣶⣾⣿⣿⣿⣷⣤⣀⡀⢀⣀⣤⣾⣿⣿⣿⣷⣶⣤⡀⠀⠀⠀⠀
    ⠀⠀⢠⣾⣿⡿⠿⠿⠿⣿⣿⣿⣿⡿⠏⠻⢿⣿⣿⣿⣿⠿⠿⠿⢿⣿⣷⡀⠀⠀
    ⠀⢠⡿⠋⠁⠀⠀⢸⣿⡇⠉⠻⣿⠇⠀⠀⠸⣿⡿⠋⢰⣿⡇⠀⠀⠈⠙⢿⡄⠀
    ⠀⡿⠁⠀⠀⠀⠀⠘⣿⣷⡀⠀⠰⣿⣶⣶⣿⡎⠀⢀⣾⣿⠇⠀⠀⠀⠀⠈⢿⠀
    ⠀⡇⠀⠀⠀⠀⠀⠀⠹⣿⣷⣄⠀⣿⣿⣿⣿⠀⣠⣾⣿⠏⠀⠀⠀⠀⠀⠀⢸⠀
    ⠀⠁⠀⠀⠀⠀⠀⠀⠀⠈⠻⢿⢇⣿⣿⣿⣿⡸⣿⠟⠁⠀⠀⠀⠀⠀⠀⠀⠈⠀
    ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣼⣿⣿⣿⣿⣧⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
    ⠀⠀⠀⠐⢤⣀⣀⢀⣀⣠⣴⣿⣿⠿⠋⠙⠿⣿⣿⣦⣄⣀⠀⠀⣀⡠⠂⠀⠀⠀
    ⠀⠀⠀⠀⠀⠈⠉⠛⠛⠛⠛⠉⠀⠀⠀⠀⠀⠈⠉⠛⠛⠛⠛⠋⠁⠀⠀⠀⠀⠀

And, of course, the classic skull tends to convey a message that even the uneducated can understand.

I have learned the hard way, not to get too creative, when presenting GUI. I've learned to use platform conventions, and ISO symbols[1], where possible; even if I am not that thrilled with the aesthetics.

[0] https://littlegreenviper.com/miscellany/the-road-most-travel...

[1] https://www.iso.org/obp/ui/

Author says text is the oldest and most stable communication technology but his example is some pictures on rock.

FYI: I collect bet on text goodies in the Awesome .TXT / Text page [1].

Yes, Text, Text, Text - The past, present and future of writing (and publishing).

[1] https://github.com/officetxt/awesome-txt

It is interesting to think about it in the context of coding.

There is a strong opinion (which I share) that text coding is much more efficient than 'graphical' alternatives due to text flexibility and nice features ( easy compare, universal medium...)

While agree with the basic premise of the author, I find it amusing that I found the black background and font selection difficult to read. When I switched to reader view, I was experiencing pure text, and it was better for me.

Context < Subtext < Plaintext

Cute, overly self congratulatory turns of word aside, this had me look up the etymology of text. I hadn’t put it together before, but woven is such a fitting root for the fabric of our thoughts made manifest.

So happy to be on a website w/o any pictures or videos or bright emojis. To tell the truth I have some problems of communication in messengers where everyone considers cool to send animated sticker instead of words.

Actually a single letter is "a picture". So Text is a collection of small pictures. There is no such thing as "Text"- we give a meaning. Otherwise it is bunch of weird arrows and curves - pictures.

I don't think they're different enough to refute my point. You see Chinese characters in a great variety of typographic and handwritten styles, all able to be read by fluent people.

It is equally true of the Chinese character system.

They have more "letters", but the point is still to unambiguously map a drawing of a character to a specific, discrete value.

There is a well known fact in the field of psychometrics: human general intelligence g can be modeled as two relatively independent sub-factors: verbal intelligence and spatial intelligence. Other sub-factors, if any, are way more speculative.

As a human being, it is quite possible to have verbal intelligence higher than spatial intelligence, and indeed I know some people with verbal tilt and some other people with spatial tilt. These people tend to think & approach problems differently, while having different strengths and weaknesses. Naturally, similar people cluster together, and some professions (e.g. lawyers, journalists, writers, programmers) are more amenable to verbally tilted persons, while other professions (mechanical engineering, airplane piloting, architecture) are amenable to spatially tilted persons.

Looking around via this lens, discerning cognitive styles inherent in design of the human experience is enlightening. One can see that our physical and social, educational environments and governing institutions are designed with one cognitive style in mind at the expense of the other: and this privilege goes to verbal cognitive style.

Let me offer a different perspective: to a person with higher spatial and weaker verbal cognition, this environment looks physically simplistic, tasteless, sometimes outright boring, often suffocatingly so. Utilitarian safety & simplicity prevailing over beauty and shape-being, denying the inhabitants possibilities of space meaningful by itself. Letters, words, strings of words are everywhere, starting with high-school where the recent historical trend of increasing verbalization of curriculum continues, and on to adult life where verbally intensive professions pay more and command significantly more power (note how in the aforementioned occupation list the second one contains less status-worthy & more specialized occupations). Limitless possibilities of rendered worlds on megapixel screens collapse into a flat-designed abstract hellscape of recursively composed words and menus. The brain of the child - a pinnacle of neuroplasticity - adapts, as relentless march of critical developmental periods continues, unnecessary white matter pathways wither away, while economically useful ones are potentiated and strengthened for the forthcoming endless competition with similar human beings, similarly shaped. The best and brightest in this game of words become lawyers and career politicians, movers and shakers of our world; but are they truly our best, and do they truly imagine the referents of their symbols ? Are we led by seeing or blind ?

Much could be said about benefits of verbal thinking, endless composability (to some, vacuous, denying interesting constrained structure) of syntax & grammar & semantics, and rightly so. Yet one wonders, which avenues of thought, of being, both alone and together as a people, were not taken. How a civlization of prevailing gestalt could look like ? Confronted with this state of affairs, one wonders, if "Always betting on text", pedal-to-the-metal, more-of-the-same is really going to bring us somewhere at all ?

If you, dear reader, have some latent spatial/geometrical imagination which was pushed away by economically profitable word manipulation engines that grew through you, maybe you too wonder about this question.

Can someone give me a tl;dr?

(Just kidding. Fantastic article.)

> We can read texts from five thousand years ago,

What's interesting is pretty much all file formats are still readable (The media they are stored on is often not)

If you have the file it should be readable with a program you can find on Google.

On topic, No, although the article might be technically correct the biggest thing holding back many consumer products is lack of ability for non-text.

Signal had massive issues for years because it didn't have emoji. Now it's up with the rest. It's so much easier now you can copy a photo from Signal.

Like all these apps it needs work, like where's my real gun emoji. But they'll get there.

This feels like an arbitrary perspective.

Text has all the same properties as an image.

It’s a composite of elements of varying height, width, and meaning to the whole.

Some sentences can be longer, or shorter. One element can be overloaded with meaning more than another.

I’m not really sure if there’s a point here at all.

Text is actually made of a bunch of little pictures. Pictures: check and mate!

Completely fails to mention speech.

And from the claims it makes ("text is the oldest") it seems either forgot about it, or is not even aware such thing exists.

Except nobody really reads text anymore. If you want to deliver a powerful message pics and vids have the most bang for the byte.

Yeah, it definitely depends for me too, but getting the nuances right from a video wouldn't work for me at all, even in a thing like dancing. Live instructor is an entirely different proposition (you can choose to focus on the bits that you care about, move around to get a better view, etc.). And I can stop them and ask for them to repeat, slow down or split their movement into more granular "part-steps".

Videos are simply not that for me. Perhaps with actual 3D videos and improved tooling we might get there (eg. ability to zoom in on an instructor's hips in "infinite" resolution if you are having trouble following that particular bit of their motion).

And I mean, you can sure learn from a video, but I just find it harder. Eg. I can go look at dance or basketball moves and try to reconstruct them, but it's usually a couple of core points that need to be taken in to achieve them that are much easier to grasp when broken into fundamentals.

However, videos where the gist of the work is on the student, vs on the instructor, are much easier to make, which I believe is part of their popularity!

You can teach someone to paint or take photos, with text. You can't teach someone to write with paintings, or with text-less video.

> Words leave nothing to the imagination.

If only. Differing interpretations of words are one of the dominant sources of political discord in the US. The things words leave to the imagination are killing people.

They weren't obnoxious, just subscribed to a more restricted interpretation of "text" than yourself, and to which many people seem to agree.

It's clearly a continuum from 3D objects, to pictures, to icons and ideographs, to highly abstract sings representing words and ultimately sounds. That fact that some designer of character sets decided to put the limit somewhere and include a bird icon and not an Obama icon does not invalidate other interpretations.

By the way, there is a thing called Emojicon where a character for 'Albert Einstein' was proposed as valid 'text'.

Cool, that's a very interesting observation, so thanks for elaborating!

As originally worded, it only sounded like "hey, this dude is famous" (for HN value of famous :-), so I would appreciate the commentary you just shared right off the bat! Thanks again for the clarification!

Sure, but the English dictionary is a bit longer than the Unicode spec :)

My point was that there is a lot of implied data that undermines this "compression" of information into "simply bytes" today: I am pointing out the implicit in OP's assumption.

Maybe I did not choose the right addendum, but a pretty comprehensive addendum is needed (in all likelihood, an "ancient English dictionary" is needed anyways).

> I don’t think hybrid or non-textual methods have really been given their due or worked on so much,

The problem is that they really have. Visual thinkers in academia have been aggressively pursuing this in CS PhDs and research since the 80s or even 70s.

The results haven't been great. Visual programming works well for some limited cases. Rational Rose, which produces code from UML diagrams, came out in 1994 and has been in continuous development since, has not taken over the world.

The big problem is high information density. With text you can go through it in a linear fashion, even if it's very long. Diagrams tend to hit a breaking point where they're too complex to take in. There's no clear place to start. Then suddenly they are worthless.

I take your point, as I often sketch things myself to help me think, but only pencil-on-paper. I would guess there are tools for sketching with a stylus on a tablet, but I have never felt a pressing-enough need to push me into checking them out.

Chat has other advantages. Lets say you're trying to agree on a restaurant for dinner:

- If I'm busy right now I don't need to stop what I'm doing to answer the call, or you don't need to try again later.

- Need 2 minutes to look up your schedule and suggest a time or find a restaurant? That's fine, I can do something else until you reply and you don't feel like you need to rush the process.

- 6 hours later I forgot which restaurant we agreed on? It's ok, just open the chat, it's right there.

- The chat is searchable, if 2 months later I don't remember the restaurant name I can probably find in a few seconds by searching some related keywords like "food", "restaurant", "dinner", etc.

This depends on the people involved. I find most of my programmer friends prefer text in most circumstances, but most people in sales, marketing, accounting, HR and management that I've worked with tend to prefer voice, video, face to face. As do people in other industries (eg my mother loves voice and video and while she does use text a lot, mostly to talk to me, often finds it tedious).

This is why my default is text until a call becomes obviously better. "Where do you want to go for dinner?" ten messages of hemming and hawing, "Call me when you're free". A call forces an interruption on the person, or a game of phone tag. Texting can be answered when each person is able, and can be used to coordinate the call.

A circle has a fixed level of abstraction, and representing a complex thought requires many circles. Importantly, in a complex image each individual circle doesn't convey much information, it's the composition that encodes any information.

Words have a near arbitrary level of abstraction, so that single words can convey immense collections of ideas, so much so that new words can be defined that represent the ideas that would otherwise take many words to express.

There's a lot of selectivity happening on this post. Nobody here has worked in an office where the management prefer video calls instead of Slack messaging? I've seen no shortage of people on HN argue that this kind of text-based communication is inferior to video calls or screen shares when solving tech problems.

It seems pretty obvious to me that text, video, audio, and images are good for different things, and that diversity is reflected in the ecosystem. There isn't a universal trend towards or away from text, not really. People are making fewer voice calls, but they're also consuming more content on Youtube instead of on blogs. They're also moving a lot of their social life onto platforms like TikTok and Instagram, which are primarily video/image.

I don't see a universal trend here in either direction. I think people are cherry-picking a couple of isolated trends or points and making really broad, generalized statements that aren't warranted.

The whole idea of picking a "best" medium here is silly anyway. It's childish, in the same way that people used to argue about whether books, movies, or games were the 'best' artistic medium. Different mediums are good for different things.

Hieroglyph writers have to resort to tricks like punning to express ideas that are not simple pictures. There's a reason why written language became more abstract.

There's an entire dialect of English that has evolved just to handle expressiveness in text formats. There's a standard and reproducible grammar, and most people (who do or have spent any time on the internet in the last 15 years) use it a second language. Messenger services allow the flexibility of asynchronous communication with the added benefit of synchronous communication when it's important (that's why it's preferred often to email, for example).

Voice notes, like email, are fully async and require actually listening to someone talk, which changes the communication media that you're working in.

Messaging has risen to the prevalence it has because of the features it brings to the table, not in spite of them.

Imagine you have to communicate something precise to me and I am stuck in a meeting:

- you send me a picture, it may be ambiguous;

- you leave me a voice call, I have to go out of the meeting room to listen to it

- you write a message, it just works. If there is info to use, I will easily copy your message in any application (Excel, Word, SAP...).

Also, text can be stored easily, so I will keep an archive of your text, probably far less an archive of your voice message or diagram, as they are very heavy.

> I am a programmer and like many of my peers I cannot stand those systems

I am also a programmer but I love my graphical shader editor in Octane [0]! It's the perfect interface for defining a graphical data flow. For example, I can implement really complex spatial conditionals by creating a merge node controlled by a greyscale bitmap. E.g. to create a gold thread on a silk material, I can merge existing gold and silk shaders through a bitmap of a lace pattern, itself edited in Photoshop. Just by connecting some nodes.

[0] https://home.otoy.com/render/octane-render/

This, exactly.

I think there are three reasons why non-textual programming is constricted to a few domains, and is not widespread as general-purpose:

- As you point out, all the tooling around programming like editors, exchange formats, version control, debuggers etc. are only available for text, rendering everything else generally unusable.

- There are many failed attempts, with many still being textual in the end, only with over-the-top syntax highlighting, such as Scratch. These attempts have paved the way of non-textual programming with gravestones, making the field look like a graveyard, so nobody wants to get into it anymore. Scratch might still have its place in education as an gentle introduction to the concepts of programming, but no software developer is seriously going to use it for any big project.

- People are used to text and they will reject other approaches instinctively. You can read many statements in this discussion here like: "Graphical programming would make your head explode", which I think is similar to how people where afraid of using trains in the early days: "Driving to fast will kill you" and "Nobody can survive such speeds". It might take a whole new generation of programmers to accept anything non-textual as intuitive (which is why it could become popular with artists and designers I think).

> Because graphical programming are just textual programming languages which are just overly complex (xml)

i don't think that's fair. i mean, on a computer, you have to serialize everything into sequences of bytes. you could also encode images into XML:

  <pixel r="255" g="80" b="110" />
  <pixel r="11" g="0" b="203" />
  ...

but i wouldn't say that makes them "textual".

Yes, it does help when you can read between the lines . . .

you somehow managed to describe the pictures without showing them

It's probably not the only removable part, the icon can be 414 bytes: https://github.com/edent/SuperTinyIcons#how-small

Can you elaborate?

Outside of bugs, I was under the impression that SVG was actually pretty well specified and that (except perhaps for pathological examples created intentionally) the rendering of an SVG was quite consistent.

What am I not aware of?

Most people don't speak fluid tok pisin either; does that make it less of a language?

Also, you'd be surprised by the amount of people that write SVG by hand, which I doubt you'll see with PNG anytime soon.

> We tend to severely underestimate oral "technologies." Scholarship existed before writing.

This comes up a lot in Bible scholarship.

Many modern people assume the words of Jesus, for example, were lost to a game of telephone before being written down years later in the form we have today.

But conveying the teachers words with true fidelity to what he said was very important to a rabbi's students, and they had a culture and techniques to make sure they did this with high reliability.

That's a beautiful sentence, especially in this here context, but it flew over my head. Would you care to expand?

Cause that phrase is just 7 words. A picture would have 993 other unnecessary words.

My kids have no idea what a codec is. Still they record and play videos and audio messages all the time.

I don't know in which world or time you are living, it's been a long time since I touch anything related to codecs.

So Lisp? You still type it in as text.

SVG is definitely human-readable. It's just drawing instructions, draw a line from there to there, then another one from there to there, etc. I can read it to someone over the phone with a pen and graph paper, and he can draw it exactly like it should be represented, with zero loss of quality. It's no different than someone looking at list of cash transactions. Numbers are text too.

Of course, to extract meaning from this text (what is it?) requires drawing it. But you don't have to, you can just name it. You could just say "I'm sending you the stylised image of a blue bird, with the instructions on how to draw it using the SVG text standard". Or just, more clearly even, the following string of characters: "twitter_logo.svg"

Is math that you not understand, really text?

Is human comprehensibility a pre-requisive to be called text? What if there was a written language no one understood, would that be text? What if someone broke the code for it? Would it suddenly become text?

This is why fax machines persist in Japan.

Some of us are on a desktop computer, all of us are on a mobile phone.

You do realize how cumbersome that mobile keyboard is right. With a click of a button, you can record audio and/or video.

But still, text messages rule them all.

Who sends texts on desktops anyway? A small minority.

Not by themselves, but maybe they didn't use an optimiser before. Or used a far better optimiser now. Or tweaked the logo, and now it's more amenable.to optimisation.

I don't know, but if they do, they seem to have a half-life of about 2.431184 years. That's interesting news.

I don't think it counts for the purposes of this blog post.

You can't translate the Wikipedia quote in the article into an SVG except by just rendering the bare text.

Checking programmatically that text is about "bird" or "Twitter" is very easy. Checking that an SVG is about "bird" or "Twitter" is very hard.

It's not even efficient. It conveys the same message as the word "Twitter" in a hundred times the size. And it wouldn't be less efficient if it were a binary format.

It's maybe more suitable for carving into granite than a binary format, but you'd be better off carving the logo itself.

How far do you take that line of reasoning? Base64 is text?

it's not consumed as text

He would have to use words for that... Could you help him showing some image of "powerful" and "effective"?

I am not the source of the original statement, but my own interpretation of "powerful" aligns with simplicity and flexibility. Plain text has only slightly more structure than a stream of bytes, meaning it retains a lot of simplicity and flexibility. Yes plain text can be inefficient and is overused (a proprietary unpublished wire-format does not need to use JSON). However, 50 years from now, the only data I feel comfortable knowing I'll be able to read is plain text (and possibly also JPEG and a few other well-specified and simple binary formats). Many binary formats are effectively defined by large complex and transient code bases that target particular tool chains and APIs. A new binary format needs to justify itself, less so plain text.

Some SVG is very readable and easy to modify. I'm no designer but I've opened up SVG files and made modifications to get what I want (remove elements, change colors, etc).

Not all SVG authoring tools make SVG's easy to read/modify. Tools in the 90's used to the same thing with HTML. Are you going to argue HTML isn't text?

All three of your examples concern people with a vested interest in arriving at a certain conclusion established long before they ever actually look at the text in question. No communication medium will solve the problem of humans trying to bend that communication to fit their pre-existing opinions.

For neutral observers, I would definitely argue that text is far more concrete than photos. There's a whole host of reasons for this in my opinion, but one that hasn't been mentioned in this thread is that humans are used to seeing images in three dimensions, whereas photos are typically two-dimensional. By removing an entire axis of data you're introducing far more room for ambiguity than could have existed previously.

The ideological message in 1984 is anything but clear when you compare it to what the message would look like had George Orwell merely written a nonfiction treatise, without resorting to any literary devices.

If you know nothing of Orwell or the historical importance of 1984 and you read the book, it certainly wouldn't be clear that Orwell was trying to make a point, let alone what that point is. Some people would guess it does, others would say "It's just a story." Orwell intentionally obfuscated his point in choosing the mode of delivery (fiction). Indeed, this intentional obfuscation/ambiguity by fiction writers is what keeps literature professors employed.

The difference is that you can read a work like 1984 and at least know it is fiction, and that the message, if any, is hidden. Whereas you cannot look at a photo and easily tell if it's documenting what you think it is, no matter how much the photographer tries. At best you can see the shapes, and to a much lesser extent, things like colors/contrast. What the photographer was trying to convey is much more open to interpretation than 1984 is.

Your reply is in text. Would you care to link a set of paintings to express your argument instead?

The funny thing about XML is that it was explicitly conceived to solve problems of representing, translating and transmitting component models and configuration.

The sad thing is that it was abused, horribly, in the early naughts when people tried to take it too far. On the one hand it was rushed out without the proper tooling and you had humans typing raw xml without the right amount of support from intelligent autocomplete coming from schema-aware editors. On the other, you had systems that tried to force XML into roles that are better suited for a programming language (I'm looking at you, ANT). And worst of all, it became the centerpiece of a bunch of tech that people really hated-- remember all that crazy WS-* shit, and "SOAP" (which could have stood for "Complex Object Access Protocol". All this wasn't the "fault" of XML which, IMHO, has rock-solid core concepts. XML got an undeserved bad rep from association to the failure of the things it was used with.

XML still survives, though. The tooling is good. If you type XAML or HTML in the right editors, it's serves the purpose competently and without drama. People aren't going crazy finding new XML based applications anymore. It's just doing it's job. I prefer pure XML for config files even now-- of course I never expect users to see it or type it. XML, at the end of the day, is intended to be manipulated by tools.

I wish I could say the same for yaml and to some extent json. I feel like these things are about to be abused like XML was.

Images are the best way to describe images. Many ideas are better expressed in text than anything else we have.

HN doesn't render ASCII or ANSI art correctly however both methods can serve as representation of Twitter in text and graphic less then <700 characters.

     ,~
    ('v)__
   (/ (``/
    \__>' hjw
     ^^
   Twitter

Bird kindly borrowed from: https://www.asciiart.eu/animals/birds-land

Just an argument against audio and video. With a stylus instead of a keyboard you can still scribble amd draw naturally. That was my original example.

You’re comparing incredibly arbitrary things. Is a photo comparable to a tweet, sentence, paragraph or novel? Is a video equatable to an essay?

You’ve convinced me that it’s all incredibly relative and complex, very much impossible to set up an equation with which to even compare it, and there’s no clear cut answer - and further, I don’t see why one side even needs to win, or why anyone cares beyond some personal attachment or game or identity pride or maybe just a childhood love of Wittgenstein.

Are symbols text? All of math could be represented as symbol. Are spoken words text? What about animal calls? The intro to Wall-E conveys so much, did I need text to understand that? I’m persuaded by a seeing a video of beautiful village in its natural state, or nature. Li Ziqi is maybe the most compelling video on all of YouTube, and it has no audio or text whatsoever.

> > XML was inspired by HTML

> No. HTML used to be "an application" of SGML.

The “No” is wrong. It's true that HTML used to be an SGML application, but that doesn't contradict that XML was inspired by HTML.

Specifically:

(1) First, HTML was an SGML application.

(2) Then the Web took over, making HTML a very popular SGML application.

(3) Then, inspired by the design of HTML, but seeking broader application without the complexity of unconstrained SGML, XML was created.

I'm reminded of a story I heard where someone responded to news of a death in the family with what they took to be a crying face, but what the recipients took to be a crying laughing face.

> why the Unicode consortium thought that putting emojis in a character set would be a good idea.

That puts blame on the wrong organisation. The consortium is to the most part only standardising existing character repertoires. Emojis were popularised by a mobile phone equipment manufacturer as a marketing stunt. http://enwp.org/Emoji#History

Emoji are modern hieroglyphs. It's just an additional means of conveying meaning.

> font designers now have control over the way I get to express my emotions

This could have been a major issue, but media platforms recognized it and have provided their own pictograms for years, mostly compatible with each other.

> It really is beyond me why the Unicode consortium thought that putting emojis in a character set would be a good idea.

iirc some japanese character sets already had them (introduced by telecoms for text messaging?), so originally they were included for compatibility [i.e. to fulfill Unicode's goal of being able to encode everything]. then people discovered the emoji keyboard on iphones and it got popular

They did it for compatibility with Japanese encodings.

That sounds like a you problem. This site is one of a dwindling number that work flawlessly with a text-only browser, so if you view it in Links or similar (as I am right now) you can have a pure text experience without any of those pesky non-text technologies getting in the way!

In the USA the USPS usually does not fail to deliver a text document, no superior nor inferior technology required.

And they still reach places having no other form of comunication.

If it was not a post card, you would need to open the envelope please.

If you would like to read that kind of text off your screen, you might just need to scan it yourself or find someone who can.

What does the display (or TTS) technology used to read it matter in this discussion?

Obviously you need ways to display text in the physical world, be it a screen or a piece of handwritten paper. It's still text.

If it was really Unicode I could use an eye roll emoji to reply with a succinct image, instead of all these words to describe my reaction.

The Softbank one looks like someone grinding their teeth in frustration to me.

But those weapons don't have the same highly political negative associations as guns do. Knives are tools more than weapons; bombs can kill a lot of people (see the 2013 Boston Marathon attack or IEDs in Afghanistan), but their typical stylization is also very cartoonish and there isn't an NRA for bombs; swords are archaic and mostly shown as medieval or pirate weapons; and bows and arrows as weapons are also archaic, and they're only used today for hunting.

Sure, but in the case of words, both sender and recipient at least agree about the content of the message, even if their interpretation is different.

In the case of emojis, both the sender's and the recipient's software agree on the message content but the actual persons no longer do because the content is suddenly being displayed very differently.