Tierra Nueva - new font inspired by old spanish map - OUT NOW :)

Dear forum-Members,

------------------- edit July 2010 ------------------

Dear members of this forum – Tierra Nueva is finally out, released in July 2010. You can find more information here, and the final specimen here.

Purchase is available via www.fonts.info and myfonts.com.

As this project was started here on typophile.com, I will leave the text below, from 2005 till now in a as-is state for documentation.


in my leisure-time I am working on a font inspired by an spanish map of america made in the year 1561. The Map has hundreds of letters and a lot of illustration (sea-monsters, native americans, etc.) on it, I think it's made by copperplace engraving (or at least by another form of engraving).

There is enough material to make a roman, italic and capital display font with a lot of ligatures, flourishes etc. of it. At the moment, i am experimenting on the roman and have started selecting glyphs for the italic version.

I guess i have worked too long without feedback (as you can see, i have created many ligatures and diacritics already), so it's over time already to show it somebody...

My method is to select some glyphs from the map (they vary, there is no "typical A"), do a 1:1-redraw, adjust and modify it trying to keep the original impression (with some exceptions). I also added some of the letters that are not availiable on the map, as it is in spanish and latin (missing U, missing Z,z, etc.). The original figures are all lowercase, the uppercase ones are derived from them.

There are certainly some problems with some glyphs at the moment (for example C, S, Z, K, W k, s, w, 2, 8, too wide f-ligatures and/or too narrow longs-ligatures, i am sure you'll find more of them).

I would be thankful if you could take a look at it, comparing it with the original letters, and telling me what you would do different, I just need some feedback before I can go further with it...

Last thing: I am planning to call it "Terra Nova". Objection? Suggestions?

Thanks a lot for your help

ps.: if I am already posting: are there any guides, tutorials, FAQs, frequently made errors, etc. how to start working on a font concerning the technical matters (opentype, ligatures, encoding, etc.)? I am trying to lern a lot about it at the moment, but it's just collecting and combining pieces of information and do trial and error...

Sebastian Nagel
8 Oct 2005 — 3:29pm

what kind of file is the attachment supposed to be?

ok, something went wrong with the attachement...

Here's an external link to the PDF: http://www.consolution.at/terranova_v027.pdf

This is pretty cool (and I don't generally like this kind of thing).

"a": The top needs a lot more character.

"g": Too cartoony. Make the ear more irregular, and offet the ends of the bottom bowl so they don't look like they're supposed to be connecting.

"k": Arms no good - try staggering them - and maybe some more flair.

"v"/"w"/"x": Too wide.

"y": That huge leftmost serif is going to give you grief.

"G": Looks dorky with such a prominent top.

"M": Too strange.

"Q": Right on!

"W": Too much.

Cedilla marks: lousy.

O-Slash: the slash could have more zing.

i-dieresis: wassup with the high diacritic?

oe: tilt the bar.

The ligs are cool. For the "fb"/"ffb" though I might leave the bar unconnected to the stem of the "b".

Numerals: I guess you've based them on the original? I would avoid them - make your own.

I think this has good potential - keep it up!


Hello Hrant,

thanks a lot for your feedback.

There is a new Version to look at here:

>> “a”: The top needs a lot more character.
new version, not good yet i think

>> “g”: Too cartoony. Make the ear more irregular, and offet the ends of the bottom bowl so they don’t look like they’re supposed to be connecting.
new version.

>> “k”: Arms no good - try staggering them - and maybe some more flair.
complete redraw, but maybe it's too vivid/thin now.
I never liked the k...

>> “v”/”w”/”x”: Too wide.
new versions.

>> “y”: That huge leftmost serif is going to give you grief.
I shortened a bit now, but i have to compare it to the original yet.
Somehow the guy in 1561 did it with the long one...

>> “G”: Looks dorky with such a prominent top.
tried something. Not sure about it yet...

>> “M”: Too strange.
too wide?

>> “Q”: Right on!
:) not sure whether this Q ist from the italic version, but i think it fits, that's why i took it for roman too.

>> “W”: Too much.
new version. simplified.
But maybe i should do a version without the middle serifes.

>> Cedilla marks: lousy.
:) i know... complete redraw. Maybe better now, but i need to think about it.

>> O-Slash: the slash could have more zing.
i hope i got the *zing* :)

>> i-dieresis: wassup with the high diacritic?
some positioning-error. new version.

>> oe: tilt the bar.
done. new version.

>> The ligs are cool. For the “fb”/”ffb” though I might leave the bar unconnected to the stem of the “b”.
The version of fb and ffb now even has the bulge of the b beeing raised. i will undo that i think.
With "bar", do you mean the bar of the "f" that is connected to the stem of the b, or the "error" described the line above? If it's about the f-bar: what is the reason to unconnect them?
fk and ffk longsk and longslongsk is not yet updated with new experimental k.

>> Numerals: I guess you’ve based them on the original? I would avoid them - make your own.
yes, they are based on the original now (the lowercase numerals, the uppercase ones are derived by me).
What is the reason for redrawing (all of) them? I understand about the 2, 6 and 8, maybe even the 5.

Thanks for your corrections, they are a big help and motivation to go on.

There is a new version up at

Many changes (black is new, red is last shown version):
- redone lowercase (unified serifs, up- and downstrokes, rounds, new k, new g
- new numerals (upper- and lowercase)
- some Details in Uppercase
- new dieresis, grave, acute, tilde, cedille

On page 2, there is a very early version of an italic. It's just a redraw of the original, without any changes yet, and there are even some letters missing. So it's just to have a look at it, but not to do a lot of critique at the moment.


  • What do you think of the new h? I'm not sure about that form at all
  • Upstroke of d, k und l are not connected to the stem in the original font, but they are in b and h. Should it be unified?

known bugs:
- Y is completely twisted...
- no updates on ligatures and combined numerals (fractions etc.) unless the main glyphs are quite fixed
- & turns too much to the left

Thanks for your comments

Hi. Going on v. 0.44 --

I generally don't like this sort of face -- too angular, sharp, and compressed for my very subjective tastes -- but you've done quite a good job with this. It has style and a certain charm. The g's are too wide, the a could use a bit more flair, and I feel that v/x/x are too staid. I particularly like your combining f's; they are clear and stylish, and work very well with their partners. Oh, and I like the italic a lot; it's not exactly legible, but its form makes up for any lack of function.

Hi jafo,

thanks for your feedback. Some of your propositions I included in v52, that can be found here:
Terra Nova v52 (black=52, red=44)

new stuff:
- revised the "v, w, x, y"
- changed some numerals again (5 and 8 in upper- and lowercase)
- "a, d, h, q and u" got new downstrokes, they integrate in the line much better now
- the "S, s" are less vivid now
- the "W" is simplified (not good yet)
- changed details in "G, M and e"

  • there is a new ampersand and a first version of an @
  • completed the raw drawing of the italic


A few things I see--

8 needs more stroke contrast.
I liked the old a d u better.
g still seems odd to me. I'd say give the head more stress contrast and squash it a little top-to-bottom; then bring the bottom finial underneath the handle.
s: change the serifs?
Zero needs more stroke contrast.
7 too wide.
Why do the periods and such have a slice taken out of them?
? needs more thick.
What distinction will there be between diereses and double-acutes?
G's top serif should be more like C's.

Italics (even though it's a rough sketch, I'll start critiquing):
Terrific motion. Make sure that as you go, you give each character a definite thick and definite thin area. Like, the z is all thin, no thick. (This applies to some of the Roman characters too.)
X, x don't seem like they should have serifs.
o and s don't fit.
c needs an upper finial.
g too wide.

I like this a lot. Very well done.

Now, keep in mind, I'm still only 16 and this is my... fourth? post on the site. But what I just mentioned is what I see.

Thanks for your feedback.
At the moment i have a lot of other things to do, so Terra Nova has to wait until weekend-sparetime, but anyway i will consider some of your critique.

Periods have this slice because the original has them too in the copperplate engraving. I guess this is the only easy way to create a "dot" with the tools available for this technique (doing a "twist" with a beginning and an ending – creating the slice in between). I think at the moment it's too much detail that can't be seen in reading-textsizes. I will try to make it more visible, or, if this does not work, eliminate the slices completely.

I think telling what you see is very important, whether you are 16 or 96. It's just that more people means more thoughts, and thoughts are good :)
Of course, as the designer of this font, i have to keep the decicion which ideas and thoughts i translate into the design, and which i do not, because otherwise it's just implementing *all* the ideas, and in the end the type turns out to be a freak ;-)

very tired, short update:
Terra Nova v59

– redone capital serifs
– minor changes in minuscle (m, n, v, w, x, y)
– some raw alternate capital and minuscle forms
– first drawing of smallcaps
– again new version of grave and acute
– Eth and eth (no experience, so maybe this is crap)
– new alternate & (too thin yet)
– added ft and fft-ligs (raw)
– new alternate form uf long s
– new form of ß
– tt-lig, trashy Th-lig
– re-designed superior-, inferior-, ...-numerals, fractions
– dagger and double-dagger
– new braces and backslash

I love this face.
Doing it as a "flipper font" would be cool.
That would use the OpenType contextual alternates feature to vary alternate glyphs for the same character, so it doesn't appear identical every time.
OpenType would also help manage all your figures, fractions and ligatures -- in fact the ligatures could be made from contextual alternates. I think you would have a lot of fun with that.

Hello Nick Shinn,

I am working on some opentype-features, but as I have found out, this is more complicated than assumed :)

This is my first font that is that far that I can think about opentype-coding, so everything I do is experimenting how it could work.
Ligatures are no problem, figures are (at the moment), as there are more then just two types on them, but there are uppercase, lowercase, table, superior, inferior, etc., and I haven't found out yet how to exchange from every figure-class to any other without problems.
I've tried to look at Adobes Garamond Pro for that, but it's too complicated for me at the moment... Learning by doing will be the key I guess.

On ligatures, there is another little problem: I am German-speaking, and we have a very "nice" ruleset for ligatures: they are only allowed, when they do not spread across so-called "Wortfugen". A "Wortfuge" would for example be here: Mouse|trap. So, let's say if we have the word "Schilfinsel" (Schilf|insel), we are not allews to use a ligature, as it could disturb the legability of this word combined of two seperate words. This is because german language consists of a large amount of these combination of words, and they would not be recognisable anymore as parts of the whole new word if they were connected by ligatures.
Of course, this is a problem for german language only. The only question is: is it useful to full-feature the font with ligature-replacements even of exotic combinations, and annoying german typographers, or should it be reduced, and the additional ligs can be accessed by the glyph-table? Or even do two versions of it :)

Doing contextual alternates is an option, but i first have to re-work them, as they are far less-developed at the moment than the rest of the glyphs.

Thanks for your Feedback

Terra Nova v61, Terra Nova Italic v8

New Version with

  • Uppercase and caps big german sz (controversial topic :)
  • Thorn and thorn (no experience, so maybe crap)
  • some new Lig-ideas and -adaptions (decorative)
  • OE-combination
  • some special chars (R) (too big yet), (c), TM, upside-down ? and !, asterisk (too big yet)

  • redone lowercase-italics

  • redone uppercase-italics A-J

  • some experiments about spacing (which would you choose?)

>I’ve tried to look at Adobes Garamond Pro

One difficulty is that, when FontLab opens an OT font, it gives the glyph classes numerical names.

So, whereas the original code might have been

feature numr { # Numerators
# Latin
sub @Figures by @Numerators;
} numr;

-- which is fairly easy to understand,

you get

feature numr { # Numerators
# Latin
sub @class1401 by @class1402;
} numr;


Thank-you for the explanation of Wortfugen, which is news to me. (Although I had come across the capital "double S" before in a Typophile thread (can someone point that out?), where that phenomenon can occur.

It is possible to include alternate sets of rules within an OpenType feature, distinguished by a "language tag". You can see this at work in the "locl" feature of A Garamond Pro.

I wonder if any foundries, such as Linotype, have implemented language-specific OT ligature substitutions?

Here I am, sounding like an authority on OpenType (and I have even given a workshop on contextual alternates), however, I have only been making OT fonts for a year, and there are huge gaps in my knowledge, as I have only "mastered" some capabilities of some features. What's more, there are new capabilities to be invented/designed all the time, such as dealing with Wortfugen.

So don't be discouraged, make this an OpenType font from the outset.
I expect that quite soon, as OpenType becomes the norm (no longer "terra nova"), typographers will have zero tolerance for fiddling around with glyph substitutions and font changes, but will expect to handle things like compound fractions and complex ligatures by going straight to the OT palette. You might think "I can retrofit this font with OT later", but if you're anything like me, you will have new fish to fry by then, and it won't get done.

I'm not saying OT features are a necessity of every font, but I do believe that it is more important to make special ligatures accessible by OT, rather than to just make them.

Hmm… it seems to me there's too much inconsistence in and among stem widths if the face is meant to be used extensively.
I'm not talking about irregularity or degree of finish but just of internal cohesion as a whole, especially in the Italic.
You should post an example of textsetting, if the spacing is already enough OK to allow it.
And of course Nick's suggestion is agreeable.

Nick Shinn:

Thanks for the reveal of this Fontlab-behavior, I already was wondering about Adobe-programmer's super-high IQ so that they can remember classes by numbers :)

Capital double s is quite a controversial thing: german double s is a ligature of longs and s (other say longs and z... another topic), and it's a letter by its own today, used when the s is spoken sharply, but the vocal in front of it is long, not shortened (otherwise, ss is used).
When writing in capital letters (for titling or advertising, etc.), the problem was that there is no equivalent for ß in capitals. This leads to typographical uglyness like "MAßNAHME". Correctly, ß is replaced by SS in capitals, so it becomes "MASSNAHME".
But there are cases where this leads to misunderstanding. In lowercase writing, the words "Maße" (measures) and "Masse" (mass) can be indentified as two different things, but in uppercase, both become "MASSE".
There are examples of a uppercase equivalent of ß, like in inscriptions of family names on tombstones, and there were also tries in DDR (eastern germany) to create a uppercase ß, but until now this was not successful. There is no place for it in unicase, and it is not used in dictionaries either. So there is no big hope that it will ever be "standard".

There is no doubt that I will try to do a lot of opentype-Scripting with Terra Nova, as this is also a project to learn new things. But I know already that it will be difficult :)
I think the problem with Wortfugen can't be solved until Opentype supports dictionaries. Until now I haven't found fonts that have solved this problem (not even linotype, that have done things like Zapfino...). It's the same problem as scripting the usage of s/longs: too many (or no) rules, too many exceptions.


I don't understand your annotation about stem widths (but i think this might be a lingual problem). Could you explain it to me in more detail?

Is it about cohesion between roman and italic (as italic is thinner than roman at the moment), or is it a problem of each of them by their own (if so, I don't understand yet)?

Short textsetting is seen in the last version (v61 iv8) (on page 2 and page 3), that can be seen in the very first posting of this thread. This is done by indesigns optical kerning (instead of using font-metrics, as this is not very usable yet). I can post a longer text when I get back home next week.

Thanks for your posting anyway

good work!

Regarding Wortfugen: I suspect that getting it correct is up to the application (e.g., InDesign) just as hyphenation is. I don't think there would be a way to implement this kind of context-sensitive functionality in OpenType anyway since it depends on something analogous to a hyphenation dictionary.

Incidentally, what do you do in the case of a font like Bembo, where the top of the f hangs over and would collide with the l if the ligature is not used? There are also some sans serif typefaces in which the fi and fl ligatures actually pull back the top of the f rather than make a connection. My understanding has always been that such ligatures exist to solve spacing problems and do not have any sort of semantic function, so I was a little alarmed when you described how things are done in Germany.

(Nice looking font, by the way.)

One of the most appreciated typographers in Germany, H.P. Willberg, solves this problem of f and l colliding where no ligs are allowed with give more spacing so no collision occurs. When done well, this works, but it's a bit strange to look at in the beginning.

Edit: It also works because there are only whole words seperated, like "book|shelf", not parts of them. So no words themselves are destroyed.

So after more than two months, I am happy to present a major update.
Please consider this as a completely new version , there are too many small and bigger chances to compare it to older versions.

glpyh list
quick and dirty presentation

  • a lot of finetuning based on largescale printouts
  • new e and g
  • new diacritic and extended latin characters
  • revised and many new ligatures
  • new (and partially experimental, unfinshed) punctuation
  • roman numerals
  • completely revised spacing (this now is "numeric spacing" in indesign, completely based on the font, no indesign optical features used)
  • no kerning yet
  • currently working on some opentype-features (revised numeral-features, capital spacing, case-sensitive positioning, etc.)

  • some changes in italic (not good yet but better than before)

  • started working on titling capitals and ornaments (digitizing phase)

I think the font is entering the "detail phase", so if you feel like niggling on details, this is just the opportunity to do so extensively ;-). I would appreciate any comment!

Thanks for your help

At the moment, I am more concentrating on Sofa/Canapé, but on request, here is Version 0.100

Please forget about the currency symbols at the moment, they are work in progress. Most of the overlapping-effects can be avoided by zooming in a bit. Sorry for the inconvenience.

I've finally found a link to the map again, it turned out to be quite famous... Map by Diego Gutiérrez



Terra nova is fabulous!


Thanks, dezcom :)

For myself, I was in doubt about Terra Nova for the last weeks, I think I need a break before I can start working on it again, but the comments today motivate me to do so soon.


I know what you mean Sebilar, about taking a break. It is a good idea. It gives us fresh eyes to see things we were blind to before. I need to do the same.


A question about your term "fabulous": Is it fabulous because it is rich of glyphs not every font has, or is it fabulous concerning quality?

I am afraid that I was trapped into expanding the glyph count (not for numeric reasons, but because I had so many ideas what could yet be integrated), and lost focus on quality management.

But as I said: I think after taking a break, it will be more easy to concentrate on quality in detail again. *And* I need a printer... Driving to University for every test printing is not very motivating...


Terra Nova

Looks like I need a new name (isn't it?).
And I guess "Terranova" instead of "Terra Nova" won't be enough?



For a Spanish font of that era, you really need to consider including several glyphs which were much used in that era. There were ligatured-character and other glyphs for de, del, and several other common "glue words". Unfortunately, there are no ISO designations for these characters yet. The closest one for "de" is Ð (an old Norse glyph; in most typefaces the horizontal also extends leftward from the vertical, which wasn't the case for the Spanish ligature). There isn't anything even close for the other combo characters commonly used by the Spanish back then.

Cheers, Warren

Thanks for your suggestion. Is there a list somewhere on the web, where I can see how the glyphs looked like, and what they meant?

It would not be a big problem to include them as a opentype stylistic set.


BTW: I've just re-opened Tierra Nueva (Terra Nova) today, and there were such obvious flaws... Let it rest for several months was a good decision indeed.
I did some major changes that calm the forms down and make the font much more readable. More to come soon.

Lovely! I enjoyed reading the critiques, and look forward to your update.


Very nice, but I don't like the double connections on the "sch" ligature. they seem to be overwhelming and there just because you think they must be.

Ahoy There! What a wonderful discovery in the treasure map. http://ian-albert.com/misc/bigimages/gutierrez/diego-gutierrez-map-whole...
The original is in the Library of Congress. Well its not exactly a treasure map but the font may be a treasure! The dots of the ? ! ; look like glittering thumbtacks too regular and distracting- why not use short strokes like you did for the j and i .
Since this is a "map font" why not include special characters for the cardinal directions North South East West, complete with arrows and florishes?

Vladimir, your link isn’t working. Here’s another one: Diego Gutierrez Map

It would certainly be nice to see this released!

I'll upload an up to date pdf this evening (europe).
It is far cleaner and calmer, there is kerning, redesigned numerals, etc.

Yes, there should be some "nautic" features like cardinal directions, and flourishes too. And I am still interested in this: "There were ligatured-character and other glyphs for de, del, and several other common glue words.", but I can't find information on this...

The most common abbreviation (and seen still a good bit here in Spain) is the DE ligature. It's very simple to make, you just take the midddle bar of the E and stick it off onto the D:

That's pretty much the only ligature you see in Spanish other than the ampersand as "et" used to be used instead of Modern Spanish "y" for "and". I've read in several places that there used to be a ligature for LL or ll, but in my research I've yet to find a single document that actually used such a ligature, although there are some interesting posibilities. The "ch" is occasionally kerned closer but seeing them ligatured is not common practice.

A good font for Spanish needs a,o,e,z,d,n,r,s,c,ñ in superscript off the top of my head (see this thread on ordinal numbers which you may or may not want their combinations ligatured. Best bet for them is do several bar/underline characters that can underline one, two, three, four of them at a time. Likewise,the tilde was not just reserved for the ñ, but could be used over any vowel as well where it meant "this letter plus an m or n" (eg, cõtar = contar, ãpliaciõ = ampliación). It could be used over multiple letters to signify that several letters (usually vowels) had been omitted.

Of course, it wouldn't hurt to have some nice capital Qs designed for the words Que and Qué, which do appear quite frequently.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

Thanks guifa for that. That's just the type of information I need at the moment. The Basics are settled, the more specific typographic features are on the rise...

I've now uploaded a new version (see very first posting of this thread).
I'm redesigning the astrological symbols etc. at the moment.

Looking very nice. One thing on your numbers and some other glyphs: check the outline direction. Some of them are going backwards which causes some interesting white-outline artifacts on Mac OS X. I would probably given the style of the font use a · under your ª and º (including in the combination Nº) instead of nothing/underline. I think it fits the font style a little better, and is certainly well represented in Spanish typography/calligraphy.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

Thanks Florian for the link to the marvelous map. One could zoom in and inspect the various portions of the map. The lettering on the map seems more tight and regular than Sebastian's interpretation- I could see how hard it was to create a font that encompassed both this regularity and the many idiosyncratic elements that give it its special flavor. I compared the new version of the font with a copy of the pdf I saved last week. I was sorry to see the small g lose its "cartoony" shape as Hrant has remarked with some justification. Sebastian, how about using the g found on the map in the word "Magallanes" just to the left of the Southern tip of S. America? It has dignity and a nice flourish to it. Also the new version of the font seems a bit too staid, perhaps because the text sample lacked the exciting ch and ft and other ligatures that added so much to the character of the font.

One thing that stood out for me was the fj ligature. I really dislike that near colliding dot. Other than that I'd love to see this get out there: it was one of the first things that I read about in Typophile before I joined.

Here's a new testing and quick-and-dirty-presentation document...

Version 0.144

Guifa, Vladimir, Eluard: your comments are not forgotten, even though most of them are not yet included in this document. I'm just consolidating this at the moment to get a new overview.

Nice presentation - that is a lot of work, but well worth it. Keep at it until you hear the cry from the crow's nest "Land Ho!" I always wondered about situations with new fonts such as this one, of how to teach the new user the keystroke sequence required to produce such ligatures as the ship or cardinal point glyphs.

Looking good, Sebastian!


That italic is to die for.

I love the presentation of it.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

So when can I buy it? Looks really good to me. I think you've captured the spirit and flavour of the old cartographers well.

Love it, Seb

Is it available yet?..

Great typeface, congratulations on your beautiful work.

Holy sh** - what a lovely specimen book....I'm in a puddle.

Hi Seb,
Is there any chance we could please please please buy this gorgeous face? Pretty please??

Many thanks!

View original article