Google Pics

Floor van Steeg

Simon Cozens said:

But good enough is good enough. People don't want beautiful, efficient code; they want to solve a problem.

This is the point — it applies to any machine generated work. Generated typography is good enough for some, generated code is good enough for some. Saying "AI is not useful for <X> (I'm an expert in X), but it  is useful for <Y> (I'm no expert in Y)" just means the output is good enough for you personally. It doesn't mean that AI is in general more useful for Y.

Matthijs Herzberg

Highly recommend this essay by Agyei Archer on Craft. His views are pragmatic but empathetic, and put into words a lot of things I've felt but haven't been able to express. It is very relevant to design in the age of AI, and our latest round of enclosure — the commons once more being privatized, or scraped in this case.

An excerpt: 

"The first craft ideology grew up alongside the first enclosure. William Morris and the Arts and Crafts movement built the modern idea of craft as a critique of industrial deskilling: the alienated factory hand against the fulfilled artisan. It resolved, famously, into luxury goods. Morris's firm could only furnish the wealthy, and he knew it, and the knowledge sits in his writing like a stone. Every revival since has followed the same arc: letterpress, hand lettering, artisanal type, slow design. Each begins as resistance to a deskilling and ends as a price tier. Resistance functions as research and development for the next enclosure."

This is a potential future of typeface design, at least in the sense that we are familiar with it now. There will be space for it in a high-end, luxury market. That's already true to some degree — the bulk of foundries on Creativemarket and my fonts aren't making substantial figures to begin with, whereas those at the top are doing very well. AI will simply exacerbate that by pulling the rug from the lower rungs, much like the steam engine replacing 98% of textile weavers, yet textile weaving still being a viable option for a select group of artisans who position themselves well in their market. Perhaps AI will create new opportunities too, as techno-optimists would have you believe.

That's not a future I endorse or want, but we need to be serious in our analysis of capitalism, which has shown the same patterns time and time again.

John Hudson

@Qwerasd

Some people don't care, perhaps it's even the majority of people, just like some people don't care about good typography or thoughtful design in any sort of discipline. However, I am people and I do want beautiful, efficient code. I care a great deal about it, and want to share that passion with others; I want people to see the beauty I see.

I am always sympathetic to craft, and the point is certainly well-made that the code generated by current models of AI is based on training on vast amounts of code written by humans (and yet still manages to be ‘hideous and harebrained’ — perhaps these are the ‘transformative’ qualities that AI brings to its derivations from other people’s work and ideas?).

I think there remains a distinction, though, that is particular to software, which is that the value in software is in what it does, while the craft that goes into writing code is invisible to the vast majority of people who use the software. People are unlikely to see the beauty that you see, because they are unlikely to ever see the code.

Anything that human beings do can embody craft, and anything that human beings make can possess that extra-utile quality of beauty that we call art. We are a crafty, art-making species. But craft and art are often hidden and hence not valued in the utile valuation of things used.

I know some very good electricians, who practice their profession with skill and considerable knowledge and practical craft. If one were to strip away the drywall, one might even admire a kind of beauty in the way they have wired a house efficiently and tidily. But the inhabitants of the house only care that the light goes on when they flip a switch, and that their appliances work when plugged in. Not only are these people not aware of how well or poorly the house has been wired — presuming everything works —, they are mostly ignorant of electricity and would struggle to explain it to someone.

I have invented some clever ways to do things in OpenType Layout — programming of sorts, and something I’ve been told is ‘Turing complete’ —: someone looking at the source for my GSUB and GPOS tables might admire how I have handled kerning to zero-width marks in Telugu script, or contextual variation of ornamentation on otherwise identical words in some stylistic sets in the Gabriola display font. But most users of these fonts are never going to see those sources and, insofar as they use the fonts, only care what outcomes they produce. What they are going to see is the typeface design, which is the primary distinctive aspect of any font: the first reason they have chosen to use this font rather than another, and whose visible, graphical aspect is, hence, where the primary value of the font subsists. Other aspects of font functionality have value, of course: the character set coverage, the accessibility to all the forms needed to typeset a particular document in a particular language, etc., but all of these are experienced in the visible outcomes of using the font, not in how it was made or how the tables are structured and active programmed aspects were written.

This is true of all software I can think of, from fonts, to games, to any kind of app, to the back-end of a website: the value is in the use, not in the code. [This is why intellectual property law around software is messed up in the USA and many other places: it seeks to apply protection somewhere other than where the value subsists.]

Circling back to AI: I acknowledged above that the code generated by current models of AI is based on — i.e. looted from — vast amounts of code written by humans. That is how these models work, and it is very much in keeping with the kind of digital colonialism and extraction that characterises the whole baleful AI plunder of human creativity by capital. It seems to me, though, by way of a thought experiment, that a programming language, as a way of directly or indirectly instructing a computer to perform certain functions and tasks, could be taught to a machine from first principles of programming and then language-specific ways to do things, in the same way that a human being would learn to program. That’s not the quick-and-dirty (oh, so dirty!) kind of LLM approach in current use, but it doesn’t seem inconceivable in a different model. This is not to say that programming languages are fully determinative, such that there is only one way to affect a function or task in the computer in any given language, but they are more so than e.g. natural language in human communications, precisely because the latter is not directly functional: outside of the military, what someone says does not compel what another person does in the way that a command in a line of code compels what the computer does. So, yes, code can be beautiful and elegant, and code can be hideous and harebrained — and I have witnessed plenty of cases of human programmers criticising each others’ code, over the years —, but ultimately its value is going to subsist in its outcomes, not in how it was written.

That seems to me to be what makes it easier for people to think AI ‘is actually good if I use it for this other thing that I know little about but I find it useful for’, when that thing is writing code. And I can totally see why that is frustrating to people who have found humanly fulfilling experience in learning to program and in the practice of doing it well, for whom that fulfillment is not limited to what the software does but also found in how the code is written. Sadly, the world is full of things that used to be culturally vital activities, through which people not only found personal fulfillment but earned a decent living, but which are now niche hobbies and only rarely a livelihood, because new technologies provided more efficient ways to achieve the same or similar utility. Calligraphy is an obvious example in the context of this thread. No one is denying that calligraphy is an activity in which artistic excellence and human dignity find expression, or is something that someone could sit down and learn to do — just as they could learn to program —, but today’s civilisation doesn’t need a lot of professional scribes, or for this to be a skill that the majority of literate people need to possess, because we have computers and, before that, typewriters.

[AI is] trained on an enormous trove of code written by people who generally had no way of anticipating that sharing their code the way they did could have allowed for it...

There was a poll question and discussion on typo.social specifically about this, asking how people who had contributed fonts to Google Fonts (and hence any OFL-released or similar open source or libre license font release) felt about them now being used to train AI. The responses are interesting — not least the observation that OFL-derivative works are required to include acknowledgement of the original copyright holder and be released under the OFL, so AI-derived output that fails to do so may be in license violation —, but my immediate thought was that the history of unintended consequences is, well, what we commonly call history.

John Hudson

For the record: I absolutely think that people should defend their livelihoods against the predations of capital, and not take it lying down when their work and ideas — in whatever medium or form — are being looted by companies and individuals intent on extracting the value of that creativity to increase their own wealth. Insisting on the cultural value of the specific in the face of the general seems to me one way to state the grounds of that defense.

John Hudson

@Peter Constable

When mechanical printing was introduced in Europe (or elsewhere), were there professional scribes who considered what they did in producing manuscripts as having qualities and intrinsic "creative" value that was missing in the outputs of mechanical printing?

Circling back to that parenthetical ‘or elsewhere’: this was absolutely the case when printing from type was introduced for Arabic script, where the quality gap between the work of professional scribes and what was possible in typography was very much wider than it had been in Europe. If you’ve ever seen a copy of Gutenberg’s 42-line Bible in person, you’ll know just how good it was in terms of the existing canons of text and book production within his culture. The was no equivalent work in print in Arabic — arguably, there still isn’t —, and a culture that valued calligraphy over basically all other art forms was very quick to point this out, and relegated typography to a very minor role when it didn’t reject the technology outright.

C.Fransen

John Hudson said:

We are a crafty, art-making species. But craft and art are often hidden and hence not valued in the utile valuation of things used.

I know some very good electricians, who practice their profession with skill and considerable knowledge and practical craft. If one were to strip away the drywall, one might even admire a kind of beauty in the way they have wired a house efficiently and tidily. But the inhabitants of the house only care that the light goes on when they flip a switch, and that their appliances work when plugged in. Not only are these people not aware of how well or poorly the house has been wired — presuming everything works —, they are mostly ignorant of electricity and would struggle to explain it to someone.

 

—, but ultimately its value is going to subsist in its outcomes, not in how it was written. 

@John Hudson

I have to disagree with you on these. (Thank you on your lengthy reply earlier by the way)

The electricians will probably tell you that this is the way it has to be done. This is the way they've learned it from their teachers. It's the way they have been doing it for years and years without additional problems and according to the needed standards. It's the way they get certified and insurance companies and customers will be happy not to be electrocuted or to have a fire break out.

They'll also tell you it's needed because of the fact when troubles occur he/she can troubleshoot the problem better and faster. Less work equals less costs. 
Then there are the electricians who come after; The successors, the colleagues, the fellow professionals (competitor-colleagues), etc. They too want to work with installations which are in order.

Sure, some people won't think about the work behind it and just care about the light bulb. But there are many people who value the idea that things are well organised behind the scenes as well.

Value is not only going to subsist in its outcomes. Value is, in my opinion, a package deal, not just a fancy wrapper.

John Hudson

I don’t think we’re disagreeing, or are disagreeing only on what this means with regard to human versus AI coding. I am struck that you describe a lot of aspects of my electrical installation metaphor as broadly determinative — they have to be done that way, especially where standards are involved — because that tallies with my sense that best practices in programming may be largely rule based.* Your point about ‘successors’ — that people who might have to upgrade or extend the installation want to work with something that is tidy and in order — obviously applies directly to code.

But those seem to me kinds of value that only come into play in specific circumstances, and the value of that whole package is to people needing to work on the building/code, while the value to the inhabitant/user is experienced essentially entirely in the functional usefulness. That value is, to my mind, something very different from a mere ‘fancy wrapper’: it’s the difference between an inhabitable house and somewhere you are going to freeze to death in the dark.

[I am reminded of the different ways that people reckon or experience value in open source software. Advocates of software freedom with whom I have talked tend to stress the value in the freedom to fork, modify, or contribute to the code, while the majority of the people who use open source software never fork, modify, or contribute to it; for them, the value subsists entirely in the fact that they don’t have to pay for it. This is even more true for open source fonts than for other kinds of software.]

I also think the argument from bad code is a losing proposition, just as the argument from too-many-fingers or incorrect letter shape have proven with regard to AI image and text generation. Tools like Claude Code are going to get better, probably to the point where they can refactor previous AI-generated code and make it progressively less hideous and harebrained.

Again, this doesn’t mean that AI-generated code is good in anything other than an instrumental, functional sense, no matter how good it gets in that regard, or even how good it gets in terms of being well-ordered and efficient. Its social, ethical, and environmental impacts remain deeply troubling. It is blatant looting by digital colonialists, and we’re all facing becoming the colonised peoples of their imperial domains.

_____

* I concluded one of my chapters in the Bloomsbury Handbook of Global Typography with a reflection on future AI typography (this was written three years ago), with reference to the two distinct periods of Jan Tschichold’s work. The later, classical typography that Tschichold was able to describe in rules for the Penguin typesetters to follow seems to me the sort of thing that the machine can also learn to produce, while the experimental modernist typography and layouts of his earlier career can only be pastiched by AI, because they are individualistic and personal and not bound by expressible rules. I wonder if any programming language permits of anything so individualistic and personal?

Ray Larabie

This post is in reference to the AI assisted Cyrillic glyph additions a customer previously mentioned.

I got permission to post these images and tell you how the fonts were modified. The basic situation is that an indie game developer was contacted by a Ukrainian gamer who wanted to add a Ukrainian language pack to the game.

The indie game dev Claude code, which hacked together the new glyphs from existing parts. You can see the original Dwarven Stonecraft font in this image. It's an old freeware font with no known license restrictions.



The Ukrainian helper opened the resulting font in FontForge, took screenshots and highlighted some problems with the generated glyphs.








The Ukrainian helper used ChatGPT-6 Astra to fix the problems. You can see before/after images of some of the changes.




And here you can see how this Ukrainian helper modified an old freeware font called "He's Dead Jim". This type of modification was not prohibited by the license for that font.