New Runic font

dudefellow's picture

This is a new runic style font of the English Roman Alphabet.
The low resolution picture demonstrates a partial character set.

Your genuine views on it would be appreciated.

TextPhotoBackground1.JPG25.52 KB
hrant's picture

Very pointy terminals are problematic, both in
terms of aesthetics and lo-resolution rendering.

The real test will come if/when you make darker weights.


dudefellow's picture

I can make any weight you want. Just let me know the ratio of height to thickness of the lines (and the ratio of height to width into the bargain) and I can make it for you upon provision of a commission. The font is a work in progress (how exciting) in a manner. If you purchase the basic alphabet as it stands now, you can have updates free.

dudefellow's picture

As for the pointy terminals, especially on the A, I have broken the traditional rules, I know, but in fact this isn't supposed to be anything like the Roman typefaces we are familiar with and has its own strict rules of its own, imparting a kind of coherence of form. Runic style scripts present their own challenges in type design.

hrant's picture

For something like Runic (unlike most writing
systems) it might make sense to have a set of
skeletons that you can "expand" as needed, but
the problem you will still have is that the darker
you go the more the strokes start "clotting", and
at varying rates depending on the angle they're
at with each other. This requires what's called
"modulation", which however is not too tricky.
You might even need trapping:

In terms of familiarity, I'm personally "native"
with three scripts, and I've worked on a number
of invented systems* as well; however all that has
only reinforced certain beliefs concerning what
types of forms humans like to look at and what
computer screens like to render.

* Just last month I finished
a font for the Shwa script.


dudefellow's picture

Thank you hrant for your suggestions. I have thought about that problem of what you call "clotting". Because all of the lines in a runic font have to be straight, the risk of the internal spaces becoming too small at greater weights is high compared to scripts where this can be overcome by convex curves, which tend to increase the apparent angle between the lines and thereby avoid the need for trapping. Another solution that can help a little is to widen the runes at greater weights. This would cause a departure from the mostly monospaced font (apart from the I) to proportional. Obviously there are limits to the thickness of the lines in any font, but especially in a runic style. Other than from pen with ink on paper chirographic reasons, we might even speculate that some of the reason for the development of rounded lines from the originally straight ones of the early Etruscan, Greek and Phoenician systems could be to increase legibility and uniformity of the interior and exterior spacing.

I looked at the Shwa system of writing that you linked to. I'll go through it in detail in time, but I recognise the concepts involved and am interested because I happen to have designed myself what I consider to be the best featural phonographic system of writing in the World, that is based on principles some of which are similar to those that the Shwa uses, and is capable of hundreds of phonograms adapted to phonetic representation of global languages. I have been writing a book about it.

hrant's picture

Always interesting to hear of such efforts.

In the case of Shwa, I was commissioned to make a font
mostly because its inventor was wary of being able to
implement the requisite typographic sensitivity. Runic's
forms are less complex so the learning curve is probably
sufficiently manageable - mostly one needs to develop a
grasp of the stroke modulation at acute joins.

Is your own invented system made up only of lines?

> we might even speculate that some of the
> reason for the development of rounded lines

In fact during the development of the Shwa font I
made some suggestions of structural improvements,
although the motivation was typographic; I'd say that
about half my suggestions were accepted. However my
one suggestion of changing a line to a curve was not...
Apparently that would have violated some important
element of linguistic logic particular to Shwa.

So, invented scripts can evolve too! :-)


dudefellow's picture

>Is your own invented system made up only of lines?

Every system of writing for spoken languages in use as far a I know is made up of lines, or is linear, for the practical reason of needing to be writing with a pen. (We can exclude stamps as practical systems for all purposes, and cuneiform because it is no longer used.) If perhaps you meant "straight" lines, the answer is that the system is designed to be executed solely by straight lines if technology and medium requires this, but is capable of being written and easily recognised even with curved lines and is very versatile from the typographic point of view, because its graphemes are based on perceptual principles that allow all reasonable kinds of distortions without affecting their identities.

hrant's picture

Interesting. Please do keep me informed!


dudefellow's picture

I am the kind of person who is reluctant to give things away and I like to phrase things like puzzles because people appreciate knowledge better when they work things out themselves given a few hints, and I do not think it is feasible to start pasting parts of my book here. It has a few hundred pages and is still in progress. I can say that I believe from what I have seen so far that my system looks more acceptable and is more ingenious (if I may say so myself) than other phonetic featural systems, including the Shwa. The system when it is used as an alphabet does not look so different from that from which the western alphabets are descended, and therefore should become culturally acceptable. If I explain the system of writing that I have devised, one might even begin to wonder whether it is the source of those alphabets. When I designed the ideal system of writing I based it not intentionally on any preceding system, but rather on reasonable principles to suit all requirements, and arrived at a system similar to the first true alphabet, which I did not know at the time. I designed it by this processes: I would make a first prototype and then ask "can it be improved in any way?" and if it could be improved, no matter in how small a way, I would make that improvement and then ask again "is there any other way in which this next stage can be improved, even just a tiny bit?" and so on until perfection is reached. A stage is reached where another might stop and be content thinking "this is good", but even then I did not stop searching for something better, even until the stage of perfection being reached. I have not changed the system for at least eight years and have continued to tax my mind to think of an improvement without being able to find any. The intention in the design process was the elimination of arbitrariness, the elimination of creativity, in favour of obeying what is right or correct in every little choice that is made.

When I have made the font I can show it to you. I have designed all of the formal characters and worked out their constructions by Euclidean geometry. Even in the glyphs themselves, there is something to be astounded by.

hrant's picture

I certainly agree with you concerning the offering of puzzles!

Best of luck with that project.


dudefellow's picture

Just some clarifications, with the risk of sounding like a one-man conversation .

Braille is an example of a system of writing that is not linear. Incidentaly, I have also designed a phonetic featural system of writing made of dots like Braille.

Some more information about my ideal system of writing: some of the phonetic features represented graphically in my ideal system of writing include:
Place of articulation, anterior versus posterior and everything in between
manner of articulation, e.g. aperture, nasality, form of the tongue
vowels: position on the formant chart, front versus back, high versus low (subtly), centrality, rounding
Suprasegmental features: stress and duration

Now, back to the runic font displayed in this discussion.
The issue of the pointy tips has been of concern to me. Sharp angles between lines behave in the aesthetic sense somewhat similarly to curved lines, in that between both curved lines and corners there needs to be smaller spaces than between parallel straight lines, which is of course why we need kerning (yet to be addressed). For curved lines it is acceptable to ascend below the baseline or above the meanline, but for pointed corners ascent above the topline or below the baseline seems not to be common practice, even though for perceptual reasons it ought to be as necessary as for curved lines. The result therefore of the pointed lines not protruding above or below is illusory manifestation of an uneven baseline or topline (I don't talk about the meanline anymore because the runic script does not have ascenders or descenders), which can become real at low resolutions. That is one problem (hold it in mind). But a runic script is not designed for pixel displays, it is adapted for being carved. For reeader-friendly electronic display, we are better using a different style than runeiform. Runic is perhaps not the most readable, but it has its own kind of aesthetic.

You suggested fixing the dark spaces at corners where there is a small angle between lines, by trapping or modulation. As regards trapping, I do not want to disturb the precise geometry, but I have been thinking that this problem could be circumvented by avoiding excessively small angles between lines. What is the shallowest angle that could be allowed before trapping is required? In my opinion, the problem of clotting between lines should depend only on the angle between them and not on their thickness or weight. I think that any angle less than about thirty degrees would start to cause problems of this kind.

I have been thinking that both of these problems could be addressed partially by special hinting for the low resolutions, without disrupting the precise geometrical ideal (think 2001 Space Odyssey monolith precision). Maybe something like a ClearType encoding. A ClearType would work best for the shallow angles and pointy tips aligned with the vertical elements in the display screen. Shallow angles and sharp corners aligned with the horizontal would be more problematic, but we can avoid them by appealing to a narrow style of width. In fact, perhaps contrary to intuitions of some, increasing the weight of the lines in the runic style, while holding the widths constant, should increase the angles between the vertical and oblique lines, rather than decrease them, and therefore might even improve the issue of clotting there. What do you think about this possibility?

As regards modulation: Did you mean variation of the thicknesses of the lines depending on their slope, or did you mean subtle variation of the thickness within straight lines such that they become slightly lighter as they approach corners? The first is not intended, as this is not supposed to be a style resembling calligraphic forms. The second is not ideal, probably even ugly.

dudefellow's picture

>seems not to be common practice

Within the sample given that is.

It looks as though the S is falling forward, doesn't it?

dudefellow's picture

Another way in which runic differs from Roman is that in the latter the curvature of the bowls at the top or base line tends to be about the same for all such letters and therefore their protrusion can be the same, but in the runic style the acuity of the corners varies with the angles between the lines, and we would expect that the corners with the smaller angles should protrude to a greater extent than the other corners. This leads to an interesting question (interesting to me at least) of how the extent of protrusion should vary as a function of the angle between the lines of the corner and the inclination at which that corner angle approaches the top or baseline. It would be interesting to know whether there have been any experiments done on this kind of perception. I could use artistic judgement, but I would like the font to be designed in such a way as it would satisfy other people and that is why I am asking for your opinions. I put this out as a kind of testing of the ice, as there would be little point in designing a large font only to find when it is finished that its users could point out several holes in it before having to go back and tweak it. I made the font in a single day (except the R which I changed a bit the next day). But of course, I couldn't commission anyone else to do it, because then it is I who would probably be the one who would not like the result.

Perhaps this sounds somewhat unconventional, but instead of thinking of the bowls or angular corners protruding, would it be acceptable to say that it is rather the vertical stems that can be cut back. According to relativity the concept should be the same. I would be more willing to chop the ends of the vertical stems down to size than vary the sizes of the oblique lines. But this approach should only be acceptable if the extent of protrusion, so to speak, of the oblique corners would be uniform throughout the font irrespective of the sharpness or size of the angle. What I am implying is that a kind of "best compromise" fixed and absolutely level (although not necessarily apparently even) extreme for all pointed terminals and corners might work better, with the lengths of the vertical stems somewhat less than the distance between those boundaries?

Any ideas on this by analogy to how the Roman letters A, M, N, V, W have been tackled in other fonts? Truncation of the corners is out of the question.

I would not mind to end the soliloquy.

hrant's picture

It is an interesting question (at least to anybody
interested in the quantification of design). I'm not
aware of any published quantitative studies; Peter
Karow might have done something in this sphere.
Hey, you should work on formalizing it yourself! :-)


dudefellow's picture

Thanks Hrant for that mention of the experimental work of Peter Karow on overshoots for bowls.

Since I first posted it here, I have left the Runic English font as a project on the shelf, by not yet making further adjustments to it, because it is not my primary motivation, but rather really only an offshoot of my principle task. The task of an alphabet is a relatively small one, which I could do for someone when it is demanded.

Hrant, when you suggested

"you should work on formalizing it yourself"

I initially put some thought into the question, reconsidering design strategies that I had rejected at earlier stages of the inception of the font, the considerable thought and calculation that went into the precise letterforms so that I knew exactly what I would be doing by the time of production of the font software itself; I considered such routes to the problem as centres of gravity, and expansion of inlines, which did not seem to offer solutions. So I abandoned those approaches in favour of shortening the vertical stems relatively to the corners of the oblique lines. But it is only recently that I have begun again the design of a runic font, this time continuing one of the styles for the ideal system of writing that I have mentioned.

I have made many of the letterforms for the ideal system of writing now and let you see a very select number of them for inspection. As I am still adding the letterforms, I have not departed from a monospaced font, so the spacing between letters has yet to be tuned. The weight is slightly thicker than in the particular example from the Runic English font shown above. Runic letterforms are generally expected to have quite thin lines to look as though they were stratched with the tip of a sharp pointed implement. In any case, narrower lines for them are necessary because of the tighter enclosed spaces, for example, in such Futhark runes as Mannaz.

As for clotting and modulation of the kind akin to trapping, I think that the only way to encorporate this to my satisfaction into the runic font would be to depart to a space of non-Euclidean hyperbolically curved geometry.

As for overshoot for pointed corners, I have worked out a method for calculating how to design the glyphs to have a desired extent of overshoot as a function of the angle between the lines in question and their thickness, and a variable adjustable to taste. Formulae such as these could be used for making glyphs of varying overshoot to be used in experiments, assessments or judgements by perception of when the baselines or toplines appear to seem even or level.

Despite being now able to choose the overshoot, which I include for some lateral corners, I have clung to the idea to chopping off parts of the termini of the vertical stems in order to make them appear to be more on the same level as the corners of the oblique lines. This solution seems to be acceptably adequate and elegant in the case of the runic style, where horizontal lines are supposed to be forbidden, possibly even at the outlines.

Perhaps you could offer your experienced opinions on their merits from the typographic point of view. If I may offer my opinion, I would say that I am creating certainly one of the better runic fonts of the few that I have managed to see. Many of the runic fonts already in existence seem to me to be quite incoherent, and often quite wrong. Those of Daniel Steven Smith I find admirable, and offer something different; as you can see I have not used them as an inspiration. I think that the academic world of runologists needs a systematic font and one that is not too closely based on and associated with a system from a fantasy fiction, for proprietary reasons alone.

The font that I am creating is to be more standardised, containing up to hundreds of runes, and my system is not limited to runes. This is a far greater number than can be alotted to the Unicode standard slots assigned for the Futhark and related runes, at Runic Unicode range 16A0-16ff. Currently, I am using one of the private use ranges, but for many years I have been interested in and quietly contemplating seeking a dedicated Unicode range for my ideal system of writing. How could someone go about that application; are many supporters required to attain that? I should hardly need to argue for the superlative nature of this ideal system of writing. Anyone with sense will be able to see that clearly. After all, as far as I know, it is the best glottographic system of writing in the World, and may turn out to be of some historical importance.

If some would like to be able to use the font, or influence what runes and weights and measures will be included with it, perhaps some source of funding could be suggested. I have added dozens and dozens of runes, and although I have made designs for each of the Futhark runes, I have not yet bothered to add equivalents of all of them to the font, even though it is very tempting to do so. And the rest I might keep for myself for a while, using the excuse of how controversial they might be. I hope a certain amount of hyperbole is a good form of promotion.

dudefellow's picture

An example of the Runic English font in use [See image DKRunicEnglishFontRingPNG_5524.jpg].

The letters were modulated somewhat by the transformation.

Syndicate content Syndicate content