Reflection on practice about

Document Sample
Reflection on practice about Powered By Docstoc
					     Reflection on practice about

�   enquire
    a family of typefaces by Antonio Cavedoni

    Submitted in partial fulfilment of the requirements for the
    Master of Arts in Typeface Design at the University of Reading, UK
    July 2009

Readying For Reading       3
The First Unger Workshop   8
Defining A Brief           13
Sketching                  14
Version Control            18
Top Heavy                  20
Defining Serifs            23
Defining Weight            25
Small Caps                 27
Capitals                   29
Italics                    30
Testing                    32
Not yet abandoned          33

The Enquire Journal        35


           As soon as I received an unconditional offer from the University
           of Reading, I started trying to prepare for the course by reading
           the missing materials from course supervisor Gerry Leonidas’
           reading list, and by sketching ideas for new typefaces. The aim
           of my sketching was essentially to gain some momentum and
           to train myself, in order to be fully effective during the year.
           Over the summer I did several small projects that informed and
           shaped a lot what I ended up doing with my practical project.
           The first one, which I called Rest, is a text typeface for setting
           books. Here is a sample of it:

           While I already had some experience with typeface design, I
           never really drew a full old-style serif face before Rest. This gave
           me some ideas about how to place my points, how to establish
           consistency and deviate from it, and it also showed me how poor
           spacing can ruin everything in a text face, despite all the best
           intentions. I now look back at Rest and see a wealth of bad deci-
           sions and just plain bad drawing which I would love to have a go
           at fixing. This is exactly why I designed it, and I am glad I’m able
           to see things differently after my journey in Reading. For Rest, I
           also started experimenting with italics:

           Again, while probably misguided and a bit rushed as an attempt,
           it did raise some fundamental questions that I would rather find
           myself addressing in the Enquire project over at Reading. For
           example, what is the right companion italic for a roman face?
           In the case of Rest, which was based on the aldine typefaces,
           I thought something quite calligraphic was called for. I really
           liked the stark contrast between the roman and italic styles in
           Matthew Carter’s Galliard family, and I thought a chancery was a
           fitting match for a Reinassance roman face.
On the 2nd of June I also started a public blog about my experi-
ence, and I started posting materials from my preparation excer-
cises. The blog was initially titled Readying for Reading and was
eventually re-titled Dispatches from Reading as soon as I moved
to the UK at the end of September. One of the first posts was of
course about Rest:

Having a blog, which is essentially a promise to keep posting
more material, gave me the motivation to keep excercising and
posting the results from my excercises over time. Feedback from
blog readers was welcome and gave me some directions. I would,
for example, receive several critiques on my letterforms over the
few months between the opening of the blog and the actual MA
program. Over the summer I was lucky to have the chance to visit
the library in Fermo, in Italy, which among other typographic
treasures hosted an original copy of Il modo de temperare le penne
con le varie sorti de’ littere ordinato per Ludovico Vicentino. Be-
cause of licensing issues with publishing images taken in public
libraries in Italy I couldn’t show much on my blog for my visit, so
I turned it into another excercise, this time in doing a revival of
an existing typeface. The results, which I called Gratie, are two
typefaces with different degrees of faithfulness to the original,
woodcut and printed sample. I only redrew the characters in the
original without trying to invent the shapes that weren’t there,
therefore the character set is extremely narrow and incomplete.
Nevertheless, it was again helpful input. The aim in this excer-
cise was to try and make the second face more usable in conven-
tional text settings.

I then did another quick weekend excercise, this time on a
sanserif design. In trying to design a new sanserif, one has to of
course come to terms with the fact that the field of so-called “hu-
manist” designs has been overrepresented in the past few years.
Therefore, making an interesting sanserif is a considerable chal-
lenge. I had made at least three attempts at designing a sanserif
family before, and all of them suffered from the same blandeness
and “genericness” – for lack of a better term – that made them
unappealing to me as I was designing them. I was then reluctant
to finish these projects, of course. In any case, I embarked upon
a new sanserif lowercase while doodling some shapes: I came up
with a couple of letterforms thad used broken curves in. I then
tried to counter the sharpness of these by making the instrokes
and outstrokes rounded as if drawn with a brush, and quickly fle-
shed out a complete lowercase. Again, this experiment didn’t go
much further than this initial burst of activity, but looking back
to it I remember how that single decision of using broken curves
informed all the other decisions and made sketching a quick se-
ries of companion letters pretty easy. This is an experience I had
again later on during my MA practical project, Enquire, which is
the main subject of this Reflection on Practice document. Lastly,
towards the beginning of September, I started drawing a new
set of shapes for a sanserif design. I didn’t know at the time, but
the influence of these early drawings in the Enquire project is
very clear to me. In this excercise, I was trying to develop a set
of shapes which would have angular countershapes and smooth,
rounded outer contours. Indeed, the excercise came about
one day when I started drawing some letters by sketching the
counters first, then drawing the “body” around them. I didn’t
complete a character set, but just drew some key characters:

This new set of shapes proved quite challenging to work with,
partly because I had a hard time envisioning how one could get
these to expand into bold weights or work in light weights while
maintaining the somewhat rigid geometry of the countershapes,
and partly because they were very mannered. I do think though
that this was the most promising experiment I did before start-
ing the MA, so much so that I refrained from publishing on the
blog because I kept wanting to go back to it, and still do. As a final
word on my pre-MA preparation, I have to say I looked intensely
at the work of previous year’s MATD students. As every prospec-
tive student I guess I was trying not repeat past errors and to
avoid copying existing designs. While it’s clear to me that origi-
nality is a very complicated subject in a field so conservative such
as typeface design, I nonetheless tried to get intimate with other
student’s shapes in order not to redraw them myself once again.


                                              On the 21st of October we had our first visit by Gerard Unger. He
                                              walked in the room, introduced himself and proceeded to give us
                                              an assignment: we were supposed to come up with a new design
                                              for continuous reading on modern mobile phones screens, such
                                              as the one in Apple’s iPhone (this turned out to be a funny coinci-
                                              dence as I would eventually find myself interning for Apple , but
                                              I didn’t know at the time). I think Professor Unger tried to get us
                                              focused more on the process rather than the end result, so I took
                                              the chance to do something I never do when drawing type, that is
                                              trying to take as many decisions as possible directly on paper. My
                                              first set of sketches, picture below reduced by 75%, was concen-
Additional material about Gerard Unger’s      trating on the letter a, which oftern has many characteristics of a
first workshop with MATD 09 class can be      typeface family. You can see many references in these sketches fo
found on my blog at this address, in a post   my summer experiments, plus ideas I was trying to incorporate
written right after it happened: http://      from some typefaces I have been researching from the Italian           typographic landscape of the 1950s, 1960s and 1970s such as the
workshop.html                                 ones from the Nebiolo or Simoncini type companies.

                                           As soon as I found some lettershapes I hadn’t tried before, espe-
                                           cially the last a you can see in the drawing, I decided to ahead and
                                           try to draw other letters around it and see how the set of shapes
                                           it had would translate to the rest of the alphabet. I tried to stick

above, my second set of drawings for the   to a serif design, because I thought it would have been a chal-
Unger workshop, scaled 70%                 lenge to try and squeeze serifs in such a tiny space. I also tried
                                           to combine several “thumb rules” for small sizes type like big x-
                                           height, low contrast and loose spacing, together with some tricks
                                           such as Dwiggins’ M-formula, broken curves to increase coun-
                                           tershapes size, and lastly a technique I picked up from Francesco
                                           Simoncini’s Delia and Selene typefaces, which is to add stencil-
                                           like cuts in lieu of ink traps at stroke intersections, to try and get
                                           them clearer. On top of this, I added broad-nib pen inspired serifs
                                           and instrokes, to try to remove some stiffness from the design.
                                           Of course these ideas proved to be way too many together, and
                                           the result felt way overdesigned. To try and test the typefaces
                                           at at small sizes, I photocopied my pencil drawings, inked and
                                           retouched them, cut the letters in little pieces of paper which
                                           I then pasted back together to form three lines of text. I then
                                         photocopied these over and tested them first by pasting them to
                                         the wall and looking at them from a long distance and then by
                                         scaling them down with the photocopier. Although both test-
                                         ing methods were fairly primitive, they did reveal that the basic
                                         shapes were working, the letters were legible. They didn’t do
                                         much to reveal the usefulness of all the design features I added
                                         though. Apart from confirming the set of basic proportions was
                                         about right, all the nuances I introduced in the design were gone
                                         as soon as one started looking at them in very small sizes. Also,
                                         because our brief was to design for a mobile phone screen, I was
below, inked and pasted together draw-   getting impatient about trying to see how my shapes would per-
ings, scaled 70%                         form on-screen rather than on paper.

                                         I then decided to digitise my drawings. I soon found out that
                                         every time I touched the lettershapes, from the original pencil
                                         drawings to the inked ones and finally to the digital shapes, I was
                                         constantly tweaking them and changing details here and there.
                                         The only difference was the speed of operations, definitely faster
                                         with the digital outlines.

After quickly digitising the letters I drew, I generated some test
documents in InDesign to try and print them on a low resolu-
tion printer. The problem was that we couldn’t possibly test our
typefaces on an iPhone or on a real mobile phone, so we relied
on “faking” small sizes and coarse resolution by using a low-end
300dpi printer lying around in the MA studio. Apart from the
dubious methodology, this again gave me some input. It con-
firmed for example that my disconnected strokes were way too
noticeable at 300dpi: they might have worked in the worse print
environments Simoncini had to deal with in he 60s, but not in
the environment I was trying to test against. Frustrated by this
discovery, after consulting Gerard Unger once again, he sug-
gested me to try different details now that I had a digital file to
play with. What I did then was build a multiple master system
with two axes: spikes and cuts, each with two endpoints. The
spikes axis referred to the sharpness of the joins between strokes,
which would become pointy, spike-like on one multiple master
endpoint and regular joins on the other end. The cuts axis was
trying to push the idea of disconnected joins to its most extreme
consequences by either disconnecting them by a huge amount or
having them merge into regular connections again. If combined,
the two axes would produce interesting effects such as artificial
ink traps.
All in all, the instances at coordinates cuts 500 / spikes 1000 of the
multiple master design space proved to be the one that would
read better at around 4.5 points on 300dpi print – on standard,
bleached office paper. Despite the lack of testing in the real
environment we were supposed to designed for, which I found at
times frustrating, I did find the workshop successful in forcing
me to think about my design process, and the various tools and
techniques I can use to tackle a design problems. That the testing
procedure was suboptimal wasn’t that much of a problem for a
short workshop like this. The workshop also gave me some hints
of the hierarchy of decisions one has to make when embarking
on a typeface design project. For fitness for purpose, I found that
proportions and contrast were the biggest factors in my typeface,
and that the detailing I added to the letterforms was essentially
refinement on an existing skeleton rather: small improvements
and adjustments rather than huge leaps forwards in legibility.
This was very helpful as an insight for my practical project, later
in the year.

Instances from the digitised Unger Work-
shop typeface, showing the various edges
of the Multiple Master design space

@@there are several notes about this            My initial brief was to design a typeface for screen usage. I was
phase in my journal entries from October        fascinated by what was going on with @font-face and the Web
and November                                    and I thought I could make a contribution there by creating a
                                                family of Web fonts that didn’t exist. I started right away by
                                                deciding to do a family with a sans and a serif component, in
                                                multiple weights with matching italics. For my I asked around
                                                to some friends which scripts were under-represented, and
                                                Joe Clark gave me the idea of investigating Georgian. Because
                                                the character set of the main Georgian alphabet (mkhedruli) is
                                                compact and the shapes seem to be fairly round-based, I decided
                                                to try and pair it with a matching Cyrillic, which is in contrast
                                                rather square-based. With all this many variables the brief was
                                                a very challenging one and indeed in the end I couldn’t com-
                                                plete it all. First I had to set aside the Georgian, partly for lack of
                                                resources about it within the department, but mostly because
                                                of me landing an internship at Apple starting from the begin-
                                                ning of June, one month before the submission of the practical.
                                                With this reduced time at my disposal I knew something had to
                                                go. The other main problem with the brief was the very notion
                                                of typeface for screen. I kept having conversations about this
                                                with fellow classmates Eben Sorkin and Claus Eggers Sørensen
                                                who were also designing for screen. At the beginning we thought
                                                we would design for a raster pixel grid and then wrap outlines
                                                around it, much like Verdana by Matthew Carter or the Nokia
                                                typeface by Erik Spiekermann were designed. But talking about
                                                this with our supervisor Gerry Leonidas, he strongly adviced us
                                                against it. We came to the conclusion that drawing a typeface for
                                                screen with the current increasing resolution and he constraints
below: lowercase a from an experimental         of a one year program would probably be a bad idea, and that we
version of “Enquire Pixel” where I was          could instead focus on designing an interesting type that would
trying to trigger specific pixels to light up   work at a certain point size regardless of resolution or techni-
at a single point size (11pt) using outlines    cal restrictions on hinting. That way we could concentrate on
                                                getting the lettershapes to work and test them on regular print-
                                                outs, rather than having to go through lengthy hinting process
                                                while at the same time learning the details of typeface design.
                                                With that issue settled, I knew I wanted to do a family that had
                                                many weights and possibly a sans & serif component. The reason
                                                for wanting many weights was because the existing set of “Web
                                                safe” typefaces has a very limited range of weights – typically
                                                just bold and regular – and after having been working as a Web
                                                designer for ten years I felt like some more variety was definitely
                                                needed there. The sans & serif idea was something that I found
                                                fascinating and challenging, and again, a useful tool to have.

            In the first few weeks I made several sketches of existing type-
            faces. The idea being that by redrawing these shapes (without
            tracing them) I could understand how they were put together,
            what was the level of detail in each design and to develop a
            vocabulary of ideas to go back to, eventually. I made a point of

            trying to sketch not only key letters from existing text typefaces,
            but also from more free and expressive display material: this
            is because I thought it was a good idea to not take a too narrow
            approach to my subject, risking imitation of existing designs.
            After reproducing existing designs I eventually moved into my
            own drawings. I remember this feeling of not really being sure
            how to go about it, because my brief didn’t give me that many
            limitations. I then started drawing a rather conventional set of
            serif shapes that made use of very flat and square countershapes.
            I presume the idea was to try to have the counters to raster-
            ize with whole pixels, thus seeming optically bigger and thus
            more readable following Gerard Unger’s advice that opening up

countershapes was beneficial to it. There was a clear influence of
my last pre-reading typeface in there, and I felt slightly uncom-
fortable with it. Was I not able to develop new ideas? In any case,
I drew it on pencil first, in two versions: high contrast and low

contrast, to try and see which one performed better. Type de-
signers generally tend to agree that a lower contrast works better
at small sizes, but I was always curious to understand why then
we were still seeing so many newspaper designs based on the
scotch or modern model: if lower contrast was so clearly better,
why were we still seeing modern type everywhere? Shouldn’t it
have been relegated to the past? It’s clear that my drawings were
rather crude, and especially the “low contrast” version is spaced
so badly that it’s almost impossible to understand if it will work
or not for this first set of drawings. I didn’t realize it at the time,
but this seemingly terrible experiment had a seed of an idea in it
that made its way through the final design, much later: namely,
the s doesn’t have a head nor a foot serif, by virtue of it being so
heavy. I’m not sure why the drawing turned out that way, to be
honest. In any case, after inking this set of sketches, photocopy-
ing them and reducing them several times over, it was clear
that with such bad spacing and drawing quality if would have
been too hard to get an impression of how the typeface actually
looked like at small sizes. Nevertheless, I was determined to try
and keep finding ideas on paper, just because it was so different
by what I had usually done before Reading, where I was jump-
ing on screen from the get go of a project. What stifled this first
attempt was Gerard Unger’s comment about it making use of
William Addison Dwiggins’ M-formula. Looking back at former
Reading students’ work, it was clear to me how Dwigginses ideas
were influential in our school and how this angularity in the
countershapes was a device often tried by students in search
of potentially intersting characteristics to incorporate in their
typeface. Because of this, I thought there was little use in mak-
ing a design much like everyone else before me had done, and so
I rejected it on that basis. The next set of sketches had the same
structure but moved away from the idea of having angular coun-
tershapes completely. With this new drawing, which I was fairly

happy with because I thought it had some flavour of its own, I
proceeded to try and make it bold and light under Gerard Unger’s
guidance. I also produced a quick trial sanserif version, to see
how it could look like. I was very happy with this experiment,
especially with the bold version, and so I proceeded to digitise
the several weights to try and see how they would perform in a
document set in them.

    and sessions hah ides hashheads diane is      and sessions hah ides hashheads       and sessions hah ides
    seasons sassies ohioans ash sis sandiness     diane is seasons sassies ohioans
    assassins dissensions naiades sodas                                                 hashheads diane is
                                                  ash sis sandiness assassins
    nosiness ha dins inn shed shannies aha                                              seasons sassies ohioans
    isidiose naiades dodo nisei deadnesses        dissensions naiades sodas
    a a hissed idaho adenose seine assession      nosiness ha dins inn shed             ash sis sandiness assassins
    anon so indeed adonai soda senonian           shannies aha isidiose naiades         dissensions naiades
    ainoi deadishness indonesians indoin
    so en see snideness dissensions
                                                  dodo nisei deadnesses a a hissed      sodas nosiness ha dins
    hiddenness dionaea node a a ha nannie         idaho adenose seine assession         inn shed shannies aha
    asses sash deans has node insee sash          anon so indeed adonai soda            isidiose naiades dodo nisei
    a donenesses indane a ion odeon               senonian ainoi deadishness
    nondiseased ass aid indonesians anni          indonesians indoin so en see
                                                                                        deadnesses a a hissed idaho
    desanded sod ion handedness hone ha
                                                  snideness dissensions hiddenness      adenose seine assession
    an did ends node hi dianisidine didine
    ones he in shininess sines he anodon          dionaea node a a ha nannie            anon so indeed adonai soda
    dissension said nondiseased ooh hi node       asses sash deans has node insee       senonian ainoi deadishness
    dissensions handedness do sheenies            sash a donenesses indane a
    sassiness ease so end sansei ainoi assassin                                         indonesians indoin so en
                                                  ion odeon nondiseased ass aid
    dies do shininess ossea dieses hood nisei                                           see snideness dissensions
    hen dash adossee son neossin inns as          indonesians anni desanded sod
    daisee inns and shoddiness indenes see        ion handedness hone ha an             hiddenness dionaea node
    diseases one end eased son dianisidin         did ends node hi dianisidine          a a ha nannie asses sash
    nonseasoned a dissensions a hens              didine ones he in shininess           deans has node insee sash
    hoodooed his hasheeshes sissonnes shood
    hi asideness adenoids doodads a ainoi         sines he anodon dissension            a donenesses indane a ion
    hands sans nondenseness hoe asidehand           While the results were encouraging, especially on the bold side,
                                                  said nondiseased ooh hi node          odeon nondiseased ass aid
    a did easies end deed in deadhead               the roman was readable
                                                  dissensions handedness do but had a remarkable absence of features
    dianisidine a deodand ashen sass soon                                               indonesians anni desanded
                                                  sheenies sassiness easebland and unappealing. One could argue that
                                                    that made it rather so end
    dhoon dine sandies seed donnish odessa
                                                  sansei ainoi assassin dies do         sod ion handedness hone
    deeded assishness so en ooh so doss oh a        the whole point of a type design is not to get noticed, but it just
    haine she hanna dosadh nodded do shod         shininess ossea dieses hood nisei     ha an did ends node hi
    donnishness nondiseased nonadhesion
                                                    didn’t adossee son neossin inns another design quite ones he
                                                  hen dashseem the world needed yet dianisidine didineas bland
    deadheads sod diseasedness son hides          asas this to me. I spent several days changing the proportions of as-
                                                     daisee inns and shoddiness         in shininess sines he anodon
    hee nonsaneness heh noons so is nose          indenes see diseases one end experimenting with stroke contrast,
    nosinesses diseasedness sensed nan in
                                                    cenders and descenders and          dissension said nondiseased
    on shoddies ash oh ends ones oisin nod          but son dianisidin nonseasoned
                                                  eased again not much came out of it. In a comparison I did with
                                                  a dissensions a hens hoodooed         ooh hi node dissensions
    adenines edison donnishness disna seas
                                                    other typeface designed for screen my work came across as fairly
                                                                                        handedness do sheenies
    doe anend oh sash sans adenose insession      his hasheeshes sissonnes shood
    shode shahin annie nee inn nosinesses                                               was a consequence of it being
                                                    economical, but I wasn’t sure if it sassiness ease so end sansei
                                                  hi asideness adenoids doodads a
    shin a an dash dens en inane nonene
                                                    rather “small nondenseness          ainoi assassin the do
                                                  ainoi hands sanson the body” or an actual feature of diesletter-
    nones hoed nonseasoned assessed a shed
    donnishness diasene inn dada sins ions        hoe asidehand a did easies end        shininess ossea dieses stroke
                                                    shapes themselves. I also played with the x-height and the hood
    in nonsense add anisidino nosinesses          deed in deadhead dianisidine a
                                                    contrast, trying to understand how they influenced the texture
                                                                                        nisei hen dash adossee son
    hen a sassies doh anoesis a dine adhesion     deodand ashen sass soon dhoon
    disna anisidino disseason in dashed                               It was becoming neossin with as daisee inns
                                                    of the typeface.donnish odessa clear that inns a humanist axis
                                                  dine sandies seed
    dondaine a ha anoesis adds aedes                having a lower so en ooh so
                                                  deeded assishnesscontrast typeface seemed a better choice. I also
                                                                                        and shoddiness indenes see

    deadishness nonadhesion assishness
    isseis shoshonean aid a neossin sandiness       had a a haine she hanna dosadh diseases one end eased son
                                                  doss oh session with Gerard Unger where we opened up the coun-
    do add asonia handiness seasons eddied        nodded do shod donnishness and that did helped make the face
                                                    tershapes of my letterforms,        dianisidin nonseasoned a
    and sessions hah ides hashheads diane is      nondiseased nonadhesion
    seasons sassies ohioans ash sis sandiness       more readable, but again, it removed flavour from an already
                                                                                        dissensions a hens hoodooed
                                                  deadheads sod diseasedness son
                                                    flavour-starved face. I decided his hasheeshes sissonnes
                                                  hides hee nonsaneness heh noons to put aside all these doubts for
    assassins dissensions naiades sodas                         2

    nosiness ha dins inn shed shannies aha
    isidiose naiades dodo nisei deadnesses                                              shood hi asideness being
                                                  soais nose and try and go on digitising the sanserif. While adenoids
                                                      while nosinesses diseasedness
    a a hissed idaho adenose seine assession      sensed nan in in the bold weights, itdoodads a ainoi hands sans
                                                    interesting on shoddies ash          again suffered from some sort
    anon so indeed adonai soda senonian           oh ends ones oisin nod adenines
    ainoi deadishness indonesians indoin so                                             nondenseness and asidehand
                                                    of unevenness and lack of direction in the regular hoe light ones.
    en see snideness dissensions hiddenness
                                                  edison donnishness disna seas
                                                                3                       a did easies end deed in

    dionaea node a a ha nannie asses sash         doe anend oh sash sans adenose
    deans has node insee sash a donenesses        insession shode shahin annie nee
                                                                                        deadhead dianisidine a
    indane a ion odeon nondiseased ass            inn nosinesses shin a an dash         deodand ashen sass soon
    aid indonesians anni desanded sod ion                                               dhoon dine sandies seed
                                                  dens en inane nonene nones
    handedness hone ha an did ends node hi
2                                                               4
2                                                               4

3                                                               5
3                                                               5

                                                                1: m /Users/verbosus/Desktop/Reading/Enquire/enquire-sans-unger-hairline.ufo, 2: m /Users/verbosus/Desktop/Reading/Enquire/enquire-sans-unger-light.ufo, 3: m /Users/verbosus/Desktop/Reading/Enquire/enquire-sans-unger.ufo, 4: m /Users/verbosus/Desktop/Reading/Enquire/Enquire Sans---
                                                                1: m /Users/verbosus/Desktop/Reading/Enquire/enquire-sans-unger-hairline.ufo, 2: m /Users/verbosus/Desktop/Reading/Enquire/enquire-sans-unger-light.ufo, 3: m /Users/verbosus/Desktop/Reading/Enquire/enquire-sans-unger.ufo, 4: m /Users/verbosus/Desktop/Reading/Enquire/Enquire Sans---


           Before I started Reading I had an idea for keeping track of my
           font files. Because editing sessions in FontLab tend to be fairly
           eterogeneous for me, fixing details here and there, and also
           because FontLab tends to be rather unstable on the Mac, I always
           like to keep copies of old font files before starting a new edit. I
           then typically end up with folders full of files in chronological
           order but which are unwieldy to go through and thus practically
           devoid of any usefulness. Because of my background as a devel-
           oper it occurred to me that it might have been useful to have a
           version control tool that would keep track of the various editing
           sessions (revisions) of a typeface and was capable of showing me
           what was changed between revisions and access the contents of
           each revision should I reconsider an editing session and want
           to go back to it, or to a single glyph in a revision. Thanks to the
           open sourcing of a substantial amount of code by Tal Leming and
           to some former experience I had with PyObjC, a bridge between
           the Macintosh’ user interface libraries called Cocoa and the
           Python programming language which typically makes it fairly
           fast to write prototyping code, I quickly sketched and developed
           a first experimental implementation of such a tool. The tool,
           called Typos, uses an actual source code revision control behind
           the scenes, a Python program called Mercurial which uses the
           distributed version control system model.

Due to the demands of the academic component of the MA
program, which were starting to increase as the deadline for the
essay at the beginning of January was approaching, work on the
Typos application was quickly put aside. While I still think such
a tool is needed, I realised what I actually needed in the short
term was a reliable backup strategy (which Typos was designed
to provide as well, but couldn’t yet perform) and that I could
live with a folder full of .vfb files for the year without too much
problems. I found a good compromise by using the Dropbox
application, which gives users 2Gb of free online storage, syn-
cronised constantly with one’s computer in real-time. Another
good feature of Dropbox is that each edit to a file is saved as a
revision. While you can’t poke into the revisions without down-
loading the files and individually open them, and while Drop-
box also has no knowledge of what a font actually is, thus being
inable to tell the user which glyphs of properties of a font where
changed between these revisions, it still provides the user with
a safety net should his machine break suddenly or should the
files get corrupted for any reason. The fact that the backups are
online right as one saves a document is a nice reassuring one for
students under stressful deadlines. I’ve been using ever since to
keep all my source files, not only the .vfb files but also encoding
files, my test documents and also the source to the Python scripts
I’ve been developing the design of the Enquire family.
While I still believe a tool like my Typos is sorely needed, and
I’m glad I managed to work on a prototype version of it, given
my workload it was just impossible to make it happen at the
same time as my main practical and theoretical work at Reading,
and has been a dormant project for over 6 months as I write this

            Around the beginning of December I was getting very frustrated
            with my typeface. I just didn’t seem to hit a set of shapes I was
            satisfied with and there wasn’t a clear path for me on how to go
            about it. Everything was put in question: the brief was looking
            less and less convincing, I was having doubts about my approach
            and I kept receiving feedback from my supervisors about my
            shapes being generally okay but utterly devoid of any interest
            – which was exactly the same way I was feeling about them. In
            short, while I had a brief for the “what” my typeface was sup-
            posed to do, I was missing a brief for myself on “how” I was actu-
            ally going to solve it. I was then trying to draw my way out of it
            At the same time I was working on my research for the essay I
            had to write for the beginning of January. My topic was a type-
            face by the Officine Simoncini called Delia, which was originally
            concieved for classified ads and subsequently used in the Italian
            phone books during the 1960s and 1970s. During that research
            I came across many designs related to Delia, including designs
            by Piero De Macchi and Ladislas Mandel. The proportions and
            details of these faces fascinated me. I was also deeply fascinated
            by the work of Adrian Frutiger and especially some of his most
            experimental typefaces like Icone or Breughel, which were
            direct inspirations for the work of another designed I admire,
            Evert Bloemsma and his many idiosincratic typefaces. Bloemsma
            himself was very keen on Frutiger’s work and on the work of
            Roger Excoffon, whose Antique Olive I always found particularly
            well-suited for small sizes, partly because of its huge x-height,
            but also for the way it enhances the horizontals of the letters.
            And lastly, I was always fascinated by the early aldine roman
            types cut by Francesco Griffo da Bologna and the more modern
            interpretations of them such as Bram de Does’ typefaces or some
            of the work by Gerard Unger.
            I can’t really explain why it didn’t happen sooner or why it
            happened the way it happened, but somehow all these sources
            blended together in the back of my head and I remember liter-
            ally having a stroke of inspiration all at once. I was in bed and
            before sleeping I thought I had never seen a top-heavy serif text
            typeface and I wondered how that would look like. As soon as
            that idea crossed my mind I knew I had to put in on paper, which
            I proceeded to do as I woke up the morning afterwards.

As soon as this decision was made, everything fell into place in
a fairly straightforward way, at least for a project with these
premises. I started drawing from the sans serif, taking deci-
sions that would eventually make their way straight to the final
design, such as the reversed stress s or the a with the heavy cut

                                                             top. A quick experiment with trying to add slab serifs to the first
                                                             digitised sans drawings convinced me it was a bad approach, and
                                                             that a new set of drawings was needed. I initially tried to keep
                                                             the lowercase a exactly the same as a sort of “link” between the
                                                             sans and serif, but again, that proved to be a misguided approach
                                                             as both faces needed different stroke characteristics. I also exper-
                                                             imented, on Gerard Unger’s suggestion, with making the strokes
                                                             taper towards the top. I produced several different versions to
                                                             test how much this feature would add to the design without be-
                                                             ing overwhelmed. This is a “super tapering” version I drew:
Enquire Sans Super Tapering ¶ 2008.12.05 ¶ HP 5100

Enquire Sans
and sessions hah ides hashheads
diane is seasons sassies ohioans ash
sis sandiness assassins dissensions
naiades sodas nosiness ha dins inn
                                                             The serif version and sessions hahsassies daysash
                                          and sessions hah ides hashheads
                                          diane is seasons sassies ohioans ash
                                                             experimental lettershapes, anddins inn thin serifs at thenaiades sodas
                                          sis sandiness assassins dissensions
                                          naiades sodas nosiness ha dins inn
                                          shed shannies aha isidiose naiades
                                                                                started four ohioans after the sans, is seasons sassies ohioans
                                                                                diane is seasons
                                                                                                  ides hashheads

                                                                                naiades sodas nosiness ha
                                                                                sis sandiness assassins dissensions

                                                                                shed shannies aha isidiose naiades
                                                                                                                              Antique Olive
                                                                                                                              with more
                                                                                                                       and sessions hah ides hashheads

                                                                                                                       ash sis sandiness assassins

                                                                                                                       nosiness ha dins inn shed
shed shannies aha isidiose naiades
dodo nisei deadnesses a a hissed                                                  preliminar a a hissed idaho
                                                             level. As soon as adodo nisei deadnesses version of the serif was designed,
                                          dodo nisei deadnesses a a hissed                                             shannies aha isidiose naiades
idaho adenose seine assession anon        idaho adenose seine assession anon        adenose seine assession anon so indeed    dodo nisei deadnesses a a
so indeed adonai soda senonian ainoi      so indeed adonai soda senonian ainoi      adonai soda senonian ainoi deadishness    hissed idaho adenose seine
deadishness indonesians indoin so en      deadishness indonesians indoin so en      indonesians indoin so en see snideness    assession anon so indeed
see snideness dissensions hiddenness      see snideness dissensions hiddenness      dissensions hiddenness dionaea node a     adonai soda senonian ainoi
dionaea node a a ha nannie asses          dionaea node a a ha nannie asses          a ha nannie asses sash deans has node     deadishness indonesians indoin
sash deans has node insee sash a          sash deans has node insee sash a          insee sash a donenesses indane a ion      so en see snideness dissensions
donenesses indane a ion odeon             donenesses indane a ion odeon             odeon nondiseased ass aid indonesians     hiddenness dionaea node a a
nondiseased ass aid indonesians anni      nondiseased ass aid indonesians anni      anni desanded sod ion handedness          ha nannie asses sash deans has
desanded sod ion handedness hone          desanded sod ion handedness hone          hone ha an did ends node hi dianisidine   node insee sash a donenesses
ha an did ends node hi dianisidine        ha an did ends node hi dianisidine        didine ones he in shininess sines he      indane a ion odeon nondiseased
didine ones he in shininess sines he      didine ones he in shininess sines he      anodon dissension said nondiseased        ass aid indonesians anni
anodon dissension said nondiseased        anodon dissension said nondiseased        ooh hi node dissensions handedness do     desanded sod ion handedness
ooh hi node dissensions handedness        ooh hi node dissensions handedness        sheenies sassiness ease so end sansei     hone ha an did ends node hi
do sheenies sassiness ease so end         do sheenies sassiness ease so end         ainoi assassin dies do shininess ossea    dianisidine didine ones he in
sansei ainoi assassin dies do shininess   sansei ainoi assassin dies do shininess   dieses hood nisei hen dash adossee        shininess sines he anodon
ossea dieses hood nisei hen dash          ossea dieses hood nisei hen dash          son neossin inns as daisee inns and       dissension said nondiseased ooh
adossee son neossin inns as daisee        adossee son neossin inns as daisee        shoddiness indenes see diseases one       hi node dissensions handedness
inns and shoddiness indenes see           inns and shoddiness indenes see           end eased son dianisidin nonseasoned      do sheenies sassiness ease
diseases one end eased son dianisidin                        test documents with both variants were put together in order to
                                          diseases one end eased son dianisidin     a dissensions a hens hoodooed his
                                                                                    hasheeshes sissonnes shood hi asideness
                                                                                                                              so end sansei ainoi assassin
nonseasoned a dissensions a hens          nonseasoned a dissensions a hens                                                    dies do shininess ossea dieses
                                                             understand how the relationship between the two faces might
                                                             have been developed, how the colour of the two should relate
                                                             and which could be each face’s role.

                                                                 I tried different ideas for Enquire Serif: in the first sketches I
                                                                 used to have very thin, slab-like serifs at the baseline and fairly
                                                                 thick ones at the instrokes. Because of the fairly sinuous style of
                                                                 the family I thought about substituting them for more bracketed
                                                                 serifs, like in this sample from the 22nd of June:

ditchside a saintless colla terroriser ha tulles hind pisci�   ditchside a saintless colla terroriser ha tulles   ditchside a saintless colla terroriser ha
colous orbell hidel do concenter inch it journalist snabbie
antinationalistic phloem mica emancipates all dump un�                                 my supervisor Gerry tulles hind piscicolous orbellcreated
                                                                  According toantinationalistic phloem
                                                               hind piscicolous orbell hidel do concenter inch
                                                               it journalist snabbie
                                                                                                                   Leonidas, though, this hidel do
chanted rooti rush is til mitotic a aphonia dan dehisces un�                                                      concenter inch it journalist snabbie anti�
conspicuous end an limnal damie pupae bolo us rite tanto          problems by dump unchanted rooti
                                                               mica emancipates alldistracting the attention from the x-height level
sire tetartohedral dotlet broodless judah concerti damsel�     rush is til mitotic a aphonia dan dehisces un�     nationalistic phloem mica emancipates
hood cars lustrant pneumocentesis papa ner copperas jots
rampoled en stepper a dis hi caducities tripsill samoa oh
                                                                  to the baseline. During one critique session he cut out the serif mito�
                                                               conspicuous end an limnal damie pupae bolo         all dump unchanted rooti rush is til
                                                               us rite tanto sire tetartohedral dotlet broodless  tic a aphonia dan dehisces unconspicuous
iou put seminaries beparse uni par intrabronchial saumont
portio in mph a circumcincture cultch nontannic palla ball
                                                                                                                   (Jonny Pinhorn) typeface
                                                                  shapes from one of my classmates’end an limnal damie pupae bolo us rite
                                                               judah concerti damselhood cars lustrant pneu�
                                                               mocentesis papa ner copperas jots rampoled en
                                                                  and tucked upon the samoa printout of Enquire Serif to make his
                                                               stepper a dis hi caducities tripsillh of aoh iou
a a bop lie ouch cuisine bid stet sill pens aurin roose men                                                       tanto sire tetartohedral dotlet broodless
emmets me hi hers lithol bremeness nunquam up unpar�
ched am poind poll topos miss tic oh deplore appraise role                                                        judah concerti damselhood cars lustrant
maomao ephoi is ralish hah uh uh a dapson to resuspect a          point:
                                                               put seminaries beparse uni par intrabronchial
jib a hatless per docmac ilima asport pac mesotartaric maid    saumont portio in mph a circumcincture cultch pneumocentesis papa ner copperas jots
scoon it hi dipcoat dilli jassids chloriamb as molle hah hee   nontannic palla ball a a bop lie ouch cuisine bid  rampoled en stepper a dis hi caducities
poll neophilism anaphroditic ream prebind suss presences       stet sill pens aurin roose men emmets me hi hers tripsill samoa oh iou put seminaries
loll bromophenol amphibolous uptrain puris um sic ale          lithol bremeness nunquam up unparched am
tod pot bijouterie amphoteric butt miserableness caballer                                                         beparse uni par intrabronchial saumont
                                                               poind poll topos miss tic oh deplore appraise role
                                                                                                              portio in mph a circumcincture cultch

I thought his point was valid and stuck with simpler, slightly
tapered serifs. I then had to decide how thick to make them and
how wide, a decision I took basing it off another Multiple Mas-
ter experiment. I created a system with tree axes: SerifHeight,
SerifShape and SerifLengh, and I then generated the 27 more
significant instances from the system, printed and compared
them. The variants went from almost sanserif-like all the way to
very thick slab serifs, with tapering variants of different angles.
Once the serifs were established, I also tested them with multi-
ple photocopying of a printout. The idea was trying to see if the
new shapes would survive extreme reproducing conditions. The
results seemed confirming that there weren’t major issues with
the serifs themselves so they weren’t altered much thereafter.

           The digital version of both Enquire sans and Enquire serifs has
           been slightly on the dark side from the beginning. I noticed right
           away that in environments such as Preview (OS X’s PDF reading
           application) or the Iliad, stroke contrast would seem reduced
           than in print and weight would increase. I also thought that for
           fine-tuning screen rendering a graded system of weights would
           have been the correct solution: people would then be able to
           swap one grade for the other in order to accomodate different
           screen environments, reversed settings and the like. I put off
           trying too hard to define weight at an early stage because I knew
           I could have done it much better once I had in place a multiple
           master system with at least a light and an ultrabold weight. I
           then did some studies on existing typeface designs weights,
           trying to understand how light or bold I could go and how were
           the usual relationships between weights in some commercial
           typefaces. This eventually turned into a Numbers spreadsheet
           comparing stroke units in absolute and relative terms:

           While I didn’t have time to implement a complete within system
           before submission, I did investigate the bolder weights for En-
           quire Serif and both the bolder and lighter ones for Enquire Sans,
           trying extreme variants in order to see how much the design
           could bend. Discussing it with my supervisors though revealed
           that in doing so the extreme weights were losing too much of
           the character of the original and were erring too much on the
           display side which would have made them slightly out of place

                                                                                        in a system of interpolated text faces.The sans extreme light and
                                                                                        extreme bold had respectively a stroke unit width of 1 unit (3 at
                                                                                        the x-height level) and a counter unit width of 2 units. Testing




                                                                                        with student licenses of Prepolator and Superpolator helped get-
                                                                                        ting a better idea on how the tools for a multiple multiple master
1: m /Users/verbosus/Desktop/Dropbox/Enquire/sources/enquire-sans-hairline-1.ufo, 2: m /Users/verbosus/Desktop/Dropbox/Enquire/sources/enquire-sans-8.ufo, 3: m /Users/verbosus/Desktop/Dropbox/Enquire/sources/enquire-sans-8-ultrabold.uf

                                                                                        system could have been designed on a technical level.

                               There are two ways of drawing small caps: you either start
                               with capital letters and adapt them for small caps, or you do it
                               the other way around by drawing small caps first and adapting
                               them as capitals later. While course supervisor Gerry Leonidas
                               advocated strongly the latter approach, arguing that the limited
                               space of the small caps makes it easier to take decisions that can
                               be easily applied in the caps, professor Gerard Unger suggested
                               the former one. I decided to make up my own mind by drawing
                               the small caps first in Enquire serif and the caps first in Enquire
                               sans, and then compare the two. My first sketch for Enquire serif
                               small caps was somewhat goofy: I like to think of it as an experi-
                               ment in designing a text version of Estro, a typeface by Aldo
                               Novarese echoing reverse-stressed clarendon-like faces popular
                               in spaghetti western movies from the 1960s. These were later

Enquire 12 small caps ¶ HP 5100 ¶ 2009.01.30

20pt italian

il dottore mi raccomando di non ostinarmi
a guardare tanto lontano. anche le cose
recenti sono preziose per essi e sopra tutto le
immaginazioni e i sogni della notte prima. ma
un po di ordine pur dovrebbe esserci e per poter
cominciare ab ovo, appena abbandonato il
dottore che di questi giorni e per lungo tempo
lascia trieste, solo per facilitargli il compito,
comperai e lessi un trattato di psicoanalisi.
non e difficile di intenderlo, ma molto noioso.
dopo pranzato, sdraiato comodamente su

20pt english

we have not cared to live in the place
ourselves, said lord canterville, since my                                                      27

grandaunt, the dowager duchess of bolton,
     redrawn completely: because small capitals only appear on their
     own and they rarely occurr within a line of text, they can be
     soemwhat more free and playful than capitals that can be both at
     the beginning of a line but also within a line of lowercase, at the
     beginning of names and have therefore to be better behaved. My
     final result, shown below, incorporates more calligraphic fea-
     tures and more open countershapes than their capital sisters:

     I wasn’t able to fully answer my original question on the design
     methodology, though, because the small capitals of Enquire sans
     had to be rushed before the submission deadline by scaling and
     widening hastily the capitals. The results are highly unsatisfac-
     tory and will have to be revised soon.

                                   I started by drawing Enquire sans capitals inspired by the Roman
Enquire Sans 5 ¶ HP 5100 ¶ 2009.02.23
                                epigraphic capitals – as many a student has done in the past:

                                   Apart from the highly decorative tail of the Q, this set of caps
20pt italian                       quickly proved to be too wide with a lowercase of these propor-
                                   tions. It was then revised to be more tame and generally narrow.
So perfettamente come mio padre mi guari anche di
                                   Ideas for Enquire serif caps were first sketched on paper and
questa abitudine. Un giorno di estate ero ritornato a
                                   then redrawn on screen. The biggest departure from the lower-

casa da una escursione scolastica, stanco e bagnato
di sudore. Mia madre mi aveva aiutato a spogliarmi e,
avvoltomi in un accappatoio, mi aveva messo a dormire
su un sofa sul quale essa stessa sedette occupata a certo
lavoro di cucito. Ero prossimo al sonno, ma avevo gli
occhi tuttavia pieni di sole e tardavo a perdere i sensi.
La dolcezza che in quella eta si accompagna al riposo
dopo una grande stanchezza, mi e evidente come una
immagine a se, tanto evidente come se fossi adesso li
accanto a quel caro corpo che piu non esiste. Ricordo
la stanza fresca e grande ove noi bambini si giuocava
e che ora, in questi tempi avari di spazio, e divisa in
due parti. In quella scena mio fratello non appare, cio
che mi sorprende perche penso che egli pur deve aver
preso parte a quella escursione e avrebbe dovuto poi
partecipare al riposo. Che abbia dormito anche lui allo
                       case was add back
                                  Io guardo quel posto, ma expect-
                                                      letters where they
altro capo del grande sofatoC, S and G.the serifs tokept fairly light andweremi
                       ed such as           They were                     extremely
                       vertical, partly me, la dolcezza del riposo,
sembra vuoto. Non vedo che to have them render better on screen and partly
                       to avoid them creating
                                              cui sento their intersection
mia 53 madre, eppoi mio padre di too dark an area atecheggiarewith
                       the main strokes.
le parole. Egli era entrato e non mi aveva subito visto
perche ad alta voce chiamo:
          Finding companion italics for the Enquire romans proved to be
          a challenging endeavor. By observing the quietness of Enquire
          sans texture and the sparkle of the serif one, I came up with the
          concept of trying to reverse the relationships in the italics. In
          order to serve as a secondary face for Enquire serif, Enquire serif
          italic should have been tame and quiet, while Enquire sans italic
          should have been more active and interesting. I started by creat-
          ing a test sheet and sketching the two versions together, but that
          didn’t take me very far. This sketch, though, did:

          I went digital soon after, and tried different variants for Enquire
          serif italic, from more angular to more plain:

     I decided to keep the angle equal from the serif italic to the sans
     italic, for them to keep some relationship with one another. The
     device I tried on the sans italic to make it stand out and be more
     “active” than its serif counterpart was to make it much more
     narrow, including chopping of the left part of the t or f crossbar
     in order to make the letters take even more space. This gave a
     very different flavour to it than the serif italic, which being al-
     most a corrected slanted roman was supposed to have much less
     eye-catching characteristics.


          The Enquire family was tested throughout its design process in
          three main environments: on various Apple computers run-
          ning Mac OS X 10.5, on paper in laser printouts at 300, 600 and
          1200dpi resolution and on an ebook reading device, the iRex
          Iliad, with an e-ink screen. Different testing documents were
          developed over the course of the design. At the beginning of the
          project it was mostly documents containing the word adhesion
          in different point sizes, with fake text generated through the
          Web site Then, as the character set increased,
          I started setting test documents in actual text, at first in either
          English and Italian and then in different European languages
          as soon as I had enough accents and diacritics to support them.
          This was done at several point sizes: printouts would generally
          be from 9 to 13 points, with occasional printouts at 18 and much
          larger point sizes for critique / hanging purposes. Where pos-
          sible, the test documents would mix styles (sans & serif, italic &
          roman, caps & lowercase, small caps & caps, regular & extra bold
          etc.) to test relationships between the elements of the family.
          I also eventually created a single, long test document with spac-
          ing tests, language tests and the full character sets of each font,
          which I kept on reprinting with newer versions of the font files.

           During the TypeCon conference in Atlanta I was told a quote,
           allegedly by Matthew Carter, according to whom “typefaces are
           never finished, only abandoned”. While Enquire is most defi-
           nitely not a finished project, it isn’t at the stage where it can be
           abandoned either.
           Notwithstanding of all the work that went into it, as I write this
           (2009.07.21) Enquire still requires major kerning work and the
           weights have to be completely reimplemented; spacing needs
           revising and the Cyrillic character set has yet to be fleshed out.
           Accents, punctuation, diacritics need a serious re-thinking, too.
           But the laying out work of the family is complete: the “flavour”
           has been established and as they stand the fonts have already
           reached a status of bare usability. I look forward to finishing this
           family and hopefully get it into the hands of potential users.

                                                        Following the example of former MATD student Titus Nemeth,
                                                        I decided to keep a journal throughout the design of the Enquire
                                                        project. While things got hasty towards submission date and
                                                        the journal hasn’t been updated much as the project grew over
                                                        time, I did manage to keep fairly detailed notes about the first
                                                        few weeks of the projects. Looking back I think it was a helpful
                                                        experiment: it provided me with context and perspective on the
                                                        early choices I took. Here it is in its original and unedited form.

2008-11-17 21:38 (Mon) Hello there, this is the start of the Enquire    both around 160dpi). He also said to concentrate more on making an
 project documentation. I’ve been here for almost two months now and    interesting design, instead of making it fit for the screen. He said
 just last week I finally started drawing something for Enquire. In a   optimising for OS X/Quartz or ClearType from the get go wasn’t a
 way, there are some ideas from the first Gerard Unger workshop         wise idea, and that we could look into it further on in the project.
 (21/10–24/10) I might get back to, but the bulk of the new drawings
 started last week. I first redrew some key glyphs from some screen      With this in mind, after having digested it, I started drawing
 types I liked (Calibri, Corbel, Sys, Verdana) and before that, in      general shapes that should go well together and work in text, and
 early October I drew key glyphs from general text faces I found        here is where I am now. I’m trying to keep all ideas open at the
 engaging, like Dante, Zeno, etc.                                       time, and I’ll be keeping on sketching in the next few weeks, even
                                                                        though it’s relatively late, MATD-speaking. I haven’t settled on a
 I dug back some sketches I made in early September, before coming      sans or a serif or on any details in particular, that will come from
here, for some letterforms with very square countershapes, where the    the testing I’ll do on screen and on printing tests in the
outside shapes where round and organic and the insides had a            department in the following days.
somewhat sharp corner at the top left of the counters, to make them
very big and easy to render on screen with pixel grids. Although I      2008-11-21 12:36 (Fri) Yesterday I started digitising my own
thought my first inking and reducing tests with the photocopier          drawings. I started from the extrabold then proceeded with the
where pretty good – because it looked like the deep cut was turned       regular. While I had relatively few problems getting the extrabold
into the natural contrast of the letter – Gerard commented that this     “adhesion” to look like I wanted, I’m still struggling with the
feature might have been too obtrusive. I then worked on a new set of     regular. It just isn’t as interesting a design as the extrabold,
sketches, these ones with more round shapes, then I experimented         which by the way is something more like a bold and then some, but
with creating extra bold, light and sans versions of the same            not quite extra. As usual I’ve been fiddling with my toolchain,
design.                                                                  exporting all the files to UFOs and correcting them with Preoplator
                                                                         just in case.
 I have been holding off from digitising my sketches so far because
that’s my usual design process, the one Claudio keeps criticising me    2008-11-26 19:57 (Wed) In the past few days I’ve been talking with
for and which I’m not satisfied with myself: I have an idea for a        fellow type designers outside Reading about many things. The topic
typeface and then in an afternoon I can whip up a whole lowercase        naturally came to my project, and my trying to get away from the
alphabet very fast. The problem is it’s full of inconsistencies and      “Reading house style”. While one can’t generalise too much on the
inaccuracies, and then fixing them becomes a pain. Which is why I        issue, I find it indeed true that most student projects from Reading
never finished a typeface so far, and which is also why I’m trying       tend to share many similar traits (one could argue these traits are
to stick with the “adhesion” method Gerry suggested.                     the ones of “good” typefaces in general, of course). What I’m trying
                                                                         to do, especially since having started the Rest project this summer
  Part of the reason I didn’t want to go digital is also because I       – which was supposed to be my pre-Reading Reading type – is to try
wanted to get basic functionality going with my font version control     and achieve simplicity and clarity in my project. The brief is
app, which I started truly working on last week. Now I can add a         complicated as it is, I don’t want to fiddle with asymmetric serif
font to a repository, track its revision and get a nice live preview     shapes or other fussy details to make my typeface unique or
of what changed when, including commit notes. It still has a long        original. I just want it to be good, in a very straightforward and
way to go before it’s usable by other people, but for me that’s all      uncomplicated way as I can make it.
I need at the moment. It works with UFO files I can export from
FontLab, which is good. It will make for a nice tool once the RoP         So even though I already digitised some drawings, I’ve been
time comes. Being based on Mercurial, it can track any kind of file     keeping on sketching and trying different ideas. On monday I went to
apart from font files, including this log. That’s a good thing.         London and sketched on the train, going and coming. I drew several
                                                                        variations, then settled on one I liked and that sort of echoed my
 And while I’m at it, I think it’s worth documenting also the           current sketches, and drew several glyphs in a sans variant, a bold
struggle in the first few weeks in Reading regarding my practical       and two trial italics (more like scripts, actually). I’m pleased
brief. I came up with three possible projects and after Gerry           with the results, as some shapes keep coming up all the time I think
essentially killed two of them I settled on my second brief – a Web     I’m finally starting to settle on something I like and that speaks
typeface. The struggle was related from not really knowing how to       with the voice I want it to have. Yesterday I was passing by a shop
approach the design of such a thing. I did have ideas about how to      window in Reading and I saw something written in Optima – which did
go at it, talking about it with Eben and Claus, but when I finally      have some resemblance to what I was doing. I was worried of having
got to talk with Gerry I got feedback I couldn’t understand. (Sorry     been subconsciously redrawing Optima all this time, so when I got
Gerry, I have to paraphrase here.) I/we wanted to approach the          home I fired up FontExplorer and was happy to see it’s different
project by shooting for some sort of low resolution first, by           enough from what I’m doing, not just in “feel” (its strokes taper
drawing bitmaps for certain sizes and then walk our way up with         slightly), but mostly even at the basic shapes level, which is the
outlines, but Gerry suggested that approach was doomed to be            difference that matters to me. Phew!
irrelevant, with the screen resolutions getting better and
better. He suggested – twice, actually, because I didn’t get much        I wouldn’t mind standing on the shoulders of giant Zapfs, but I
from our first talk – to shoot for a point size for text in a           would mind redoing other people’s faces just for lack of
common, high enough resolution all the screens where converging to,     insipration.
something like 200dpi which cheap laser printers can do but which
also devices like the iPhone or my Iliad are approaching (they are

2008-12-03 01:41 (Wed) Today Gerard was in. During the weekend I        devastating. Legato in particular is completely blowing my socks,
 redrew my base serif version based on the new “train” sketches. This   but it has to be said it’s also slightly darker so maybe the
 morning I showed it to Gerard, we agreed to make some changes which    comparison wasn’t as good as it could have been. Eben argues that
 where very beneficial because he thought I was going in the Italian    the thing with Legato is that the different axes of the outside and
 Old Style (not the Goudy one, the Veronese) direction – which isn’t    counter shapes are playing together to improve the salience of each
 what I am trying to do at all. I then started redrawing the sans. I    letter, and that’s what is making it much easier to read than
 figured I might try and see what could a sans variant look like, and   Enquire Sans.
 if it would perform better than this one. After Gerard’s advice I
 also contacted Mary Dyson about eventually testing my typeface,         I think this version of my typeface reads okay, but 1. it performs
 maybe in January or so. She says it’s feasible, and that the easiest   very poorly on-screen and 2. on 1200 dpi print it’s much worse than
 thing to test would be a comparison of my type with other in the       some of the C fonts. So I have a bittersweet feeling towards it: on
 same category. But before doing that I need to have a lowercase and    one hand I think it’s a very interesting direction to pursue and I’m
 uppercase, I need to have established the main roman weight, and I     eager to try to apply the same principle on a serif design, which is
 need to think with Mary of a way to test this thing that will          yet another unexplored area, on the other hand I can’t ignore the
 actually make sense.                                                   evidence of the other typefaces being just plain better. To be fair,
                                                                        the C fonts in print don’t look very good, and Legato doesn’t look
 Meanwhile I’ve been moving other stuff: I started working on my        quite good on screen either, so maybe it’s a matter of striking the
essay and dissertation, and also looked into my two non-latins. I       right balance between the two in my own design.
now have my Cyrillic “adhesion” (нобельфайк) courtesy of Vera
Evstafieva, and I also have some texts in Georgian on which I            In a way I wish there was a magic ball that would tell me if I was
performed a simple letter frequency analysis to see which letters       right in pursuing this idea or not, on the other I guess I have to
where the most used and which combinations occurred most                travel this path myself and see what happens.
frequently. The answer is these letters, in order: ა ი ე მ ს რ ნ ვ დ.
                                                                        2008-12-29 01:59 (Mon) I’m at home in Sassuolo. I should actually go
  Yesterday night I sketched a combined roman + cyrillic + georgian,     to bed since tomorrow I have to go to the bank at 8:30 (eek!) then
and I figured I would try to do it with a vertical axis, because         at the museum to do some research again. Since I came home I’ve been
that’s what I always see in Georgian types. Even Sylfaen, which has      doing an awful lot of research, but basically didn’t do much for my
a sloped axis, has a vertical axis for Georgian. But a change of         typeface: I did manage to tweak the weight of the stems a bit and to
axis would be a major edit to my current sketches for the roman, so      change the terminations after feedback from Claudio Piccinini and
I need to check with Fiona and possibly a native speaker before          Eben. Today I also got very encouraging feedback from Piero De
doing it. I also have had contacts with Andrea Tinnes, who designed      Macchi: he says I should try to make the current serifs more organic
a titling Georgian typeface, and who gave me the email address of a      and more related to the nice flow of the e and s.
client of hers, from Tbilsi. Tomorrow I’ll check my multiple weights
of the sans, which got somewhat negative comments from Kris: he says     I haven’t printed the new serif version yet, and the first
they all feel quite uneven and it’s not sure where they’re going. I     printouts I did in Reading of my early top-heavy serif where okay
figured I would make the axis vertical on the sans to follow the        but not great. Despite all this I think the idea is starting to have
Franklin Gothic model, which I like very much in publication            legs and I’m willing to go forward with it. I spent a little bit of
use. But yeah, the arch stem joins don’t quite match in the various     time tonight re-evaluating what makes Legato work and what is it I’m
weights, the contrast constantly moves all over the place, and the      missing in Enquire Sans. I also found other typefaces in the same
weight distribution in the various instances is just not smooth, but    category (although without the black-white play) like Bitstream
it’s fixable.                                                           Chianti, which is surprisingly similar to Legato in its finish. I
                                                                        took a closer look to Frutiger’s Icone, although I should really try
 Now I’m off to write an email to the yellow pages guy, for my          to get my hands on his new book to understand its ideas about it.
essay on the Simoncini directories typeface, Delia.
                                                                         One funny thing about Enquire Serif and Enquire Sans is that if
2008-12-04 01:02 (Thu) This afternoon (wednesday) I had a talk with     you set a line of Enquire Sans within the Serif, it’s *very* hard to
 Gerard about the sans I drew yesterday and the various weights. He     tell them apart because the colour and texture is quite similar. The
 keeps trying to gently push me towards experimenting more. He always   rhythm is very different though due to the different axis of the
 has positive comments about my shapes, but seems to think I can do     Sans (vertical), the different fitting (looser) and the overall
 better than this. His argument is that since I’m here I should         contrast (much lower), so you do get a little bit of leeway between
 really take advantage of the safey net of a year of time to explore    the two. The question is: is the sans even needed, seeing how close
 different directions, even if they’re dead ends on paper. This makes   it is? I mean, isn’t the purpose of an accompanying sans to provide
 sense, but somewhat clashes with my intentions to “just” make a good   some differentiation and contrast? At the moment I’m having formal
 text typeface. Thinking about it, though, Gerard’s approach            harmony between the two but no contrast whatsoever. Maybe the
 definitely makes more sense: his own is a way of thinking about any    contrast can be introduced with the weights? Maybe one could use
 design problem, really, whereas mine is very narrow-minded about one   Enquire Sans Extrabold for titles and Enquire Serif Regular for text
 specific goal. I’ll follow his advice and push things around. So far   and the two would live merrily together on the same page without
 I feel like I have designed at least 5 completely different adhesion   needing an “external” sans? That does sound quite monomaniachal…
 variants already – including the first Unger workshop. I can do
 better!                                                                2009-01-20 22:13 (Tue) Essay done, it was painful. Today I showed my
                                                                         workfile to Victor Gaultney and while I was explaining all the ideas
2008-12-06 22:27 (Sat) So I took Gerard’s advice. That night, before     I had (the Web font, Georgian & Cyrillic, the top-heavyness, etc.)
 sleeping, I had the idea of trying to draw a version of Enquire Sans    it occurred to me like I might have cut myself a brief that’s way
 that would be top-heavy, like Antique Olive is. The thinking being      beyond my possibilities. The main thing Victor said about it is that
 that since the top of the letters is the bit that helps us read         having the weight at the top might give me some serious hinting
 them, by adjusting it ever so slightly I could help the eye a little    problems, which would kill the on-screen readability. On top of
 bit. This is super unscientific, but I thought it was worth a           that, last week I spoke about screen fonts with Adam Twardoch, who
 shot. So on thursday morning I started a new file and redrew my         essentially suggested he would aim for a particular pixel
 “adhesion” on the same skeleton of Enquire, but with a weight           reproduction at a certain point size & start from there, maybe
 distribution all at the top. Guess what, it worked! The design          distorting the outlines to get a specific rendering. That’s opposite
 immediately gained personality and meaning, and I was very, very        from what I’m doing now. Dunno!
 glad about it. Even Gerard liked it! (Update 2008-12-26: I should
 note that Gerard’s first comment when seeing it was “now, *this* is    2009-02-01 19:42 (Sun) Last wednesday, after finishing the essay, I
 a typeface!”) After that I spent the afternoon trying to get the        frantically rushed a lowercase for Enquire serif for Gerry’s
 texture to look right, which I found more difficult than expected       thursday interim review session. I was the last of the day, and all
 and it’s still not done. I also produced a version with stems           in all I thought it went well. He pointed out my serifs being of the
 tapering to the top, another unexpected feature which might be          wrong shape (I had them bracketed), because all my shapes look like
 interesting to play with.                                               they’re trying to push out of their outline, and with the weight
                                                                         distribution at the top, these serifs where a bit “stealing the
 Yesterday though I tried comparing my top-heavy version with            show”, by making the bottom heavier than necessary. I talked about
several other regular and top-heavy sanserifs and the results where      it with Gerard this week, and he agrees that bracketed serifs don’t

follow from my arch-stem connections, which aren’t bracketed at all.     to do, but it needed to get sorted eventually. Now I know which
                                                                         languages will (hopefully) be supported by Enquire, which glyphs I
 Gerard hadn’t seen the serif before this week. The second crit he       still need to draw, which accents, etc. and both the sans, serif and
gave me started by saying that the design was growing on him, which      the matching italics have proper encodings. Hope to produce the
was encouraging. Myself, I think I’m finally committed to the idea       various bold versions is fading as I will struggle just to produce
and I really like what I have so far. Marco argues it’s a bit too        the regular ones I think.
informal and wouldn’t be appropriate for an annual report or
something like that. It’s a valid point, which I’m not sure how to
address right now. Gerard also commented on the fact that without
serifs on a, c and s I’m going towards a true informal and maybe I
should go all the way and introduce upright italic features in the
design. We also designed together a version which had Swift-like
serifs on said letters: it’s much more serious indeed, but it
doesn’t sing to me. I think there’s a combination of issues at play,
not least the fact that I’m starting to get too attached to the
shapes I have now, because I like them so much.

2009-02-04 00:23 (Wed) We have small caps! I started working on them
 in the afternoon after coming home from James Mosley’s lecture and I
 have the full A-Z done now. Many letters are just horrible,
 especially C, G, X, Z and S, but at least it’s a start. Tobias is
 coming on monday so I’d like to show him some uppercase and some
 text set in Enquire, possibly. I need to go to the dept tomorrow and
 print some of this to see how it’s coming out. I also need to print
 a new copy to hang in my room as I have a pretty old version
 now. Next up: numbers, weights, sans or cyrillic + georgian?

2009-02-05 18:49 (Thu) Yesterday afternoon a cranked out the full
 lowercase for the sans, then started showing it around together with
 the serif. The feedback I got so far has been all negative:
 Francesco, one of my partners in the studio, said he hates it and he
 would never use it (on the serif). Marina, commenting on the sans,
 says the weight distribution is wrong and “ugly” at display
 size. Claudio Piccinini started by saying the shapes are very
 elegant – he particularly loved the small caps! – but then said he
 would probably never use it on screen due to the top-heaviness which
 bothers him. I wrote an email to Gerard Unger about it, who replied
 I should go ahead with my design anyway. This is the second crisis I
 go through with Enquire, but I just realised it comes with the
 territory: I’m not used to go to unconventional places and I should
 accept that people will not like what I come up with. I think
 Enquire in its current state has a lot of potential, I want to go
 ahead with it!

2009-04-03 22:38 (Fri) It occurred to me I might have to update this
 log since I haven’t been touching it for two months! I have worked
 on Enquire in the past two months, although I’m still nowhere where
 I wish I would be. The updated “internal” deadline for the practical
 for me is the 5th of June, because I’ll be moving to California
 after that. I respaced the serif, redrew the small caps, drew caps,
 somewhat fixed some details for screen rendering after Microsoft
 feedback (horizontal stroke in bottom bowl of g, etc.), and drew an
 ultra bold weight for the sans, in order to interpolate the thing. I
 also have ranging figures for both sans and serif, which I’m pretty
 happy with. Unfortunately the serif has huge colour/stroke
 inconsistency issues at the moment, so I can’t really move to the
 non-latin as I would like. I should point out that since the
 California news, I decided together with Gerry to drop the
 Georgian. I still think in the back of my head I could probably do
 cyrillic + greek, but since now it’s the beginning of April and I
 have like less than two months to finish the thing, I don’t think I
 can do it.

 Meanwhile I set up a test document I’ll be using for checking
spacing and stuff as I go along. I’m still unsure what to put in it

2009-06-21 10:32 (Sun) Wow, another two months without updates. I’m in
 California now. Enquire now has lowercase and uppercase italics in
 both sans & serifs. It also has an ultra bold in the serif
 lowercase. The deadline is in two weeks and this is the stuff I have
 to go through: 1. make a new pass at spacing the sans, 2. extend the
 character set of the sans & serif romans, with diacritics & all,
 3. finish the character set of the cyrillics, 4. extend the
 character set of the italics, possibly including cyrillic italics,
 5. finish uppercase for the extra bolds and extend character set,
 6. kern everything. Given the time I have I will have to cut
 corners, of course. Wish me luck!

2009-06-28 19:44 (Sun) I spent the week mucking about with encodings,
 naming files, aliases and unicodes. Not exactly the most fun thing


Shared By: