Improving Josefin Sans [Intro + Worklog]

[Thomas Jockin]

After an initial review of Josefin Sans, here's my projected improvements in this final design sprint:

• Darken the LC Bold Master

• Increase x-height of Bold Master

• Check Horizontal contrast for Bold Master

• Check Overshoot for Bold Master

• Redraw Bold and Light Master /S

• Check Bold Master diagonal stroke letters (/M, /N, /A, /V)

• Lighten Bold Master /K

• Review Bold Master Curve velocity of /C, /G, /D, /B, /P

Will post presentation slides with visualization tonight/ tomorrow AM.

[Thomas Jockin]

An overview of the adjustments that will be made in the Bold Master:

Most apparent is the need to increase the joint of the /n family letters. They felt too light next to the /o shapes, no matter what kind of trick I tired in the horizontal stroke of the /n's.

Also applied in this direction:

• Darken the LC Bold Master 

• Increase x-height of Bold Master 

• Check Horizontal contrast for Bold Master 

• Check Overshoot for Bold Master

Considering there is a Roman and Italic to apply these changes to this is the extend I can do in the allotted time.

So the projected timeline is:

Remaining of this week: Josefin Sans Roman Bold Adjustments.

Week Two: Josefin Sans Italic Bold Adjustments.

Week Three: Ext'd Latin for Roman and Italic. 

[Dave Crossland]

On 6 September 2016 at 14:17, Thomas Jockin <tho...@thomasjockin.com> wrote:

An overview of the adjustments that will be made in the Bold Master:

(NB, A good comment also on https://dribbble.com/shots/2945045-Google-Fonts-Improvement-Project-Josefin-Sans-Proposal :)

[Dave Crossland]

Hi Thomas!

I definitely see the need for an art-nouveau geometric sans for text usage, but I'm wary of clipping the distinctive display features of Josefin to make it more versatile... Eg, I would like keep the apex of A pointed, keep the 'g' and 'o' more circular, and for the 'e' tail I could even bring it back (rather than forward) so its terminal is perpendicular to the bar (but that might be going too far :) I'd also seek to make the eye of A and e a little larger. I like the n/m (esp as the contrast is very high in them compared to others in the original) and the tighter spacing and narrower A :)

Cheers

Dave

[Thomas Jockin]

Hi Dave,

My understanding of this project is to improve the Google fonts selected to increase usage.

Is my understanding correct?

[Dave Crossland]

On 6 September 2016 at 20:00, Thomas Jockin <tho...@thomasjockin.com> wrote:

My understanding of this project is to improve the Google fonts selected to increase usage.

Is my understanding correct?

Nearly :) My aim is to improve quality, which will naturally increase usage. To improve quality without changing the essential nature of the typeface is the challenge! :)

[Thomas Jockin]

I agree.

What happens if there are structural differences between the Light and Bold master?

A general observation I'm seeing is compensations that need to done to Bold masters were not taken when they expanded from the Light master.

What is ok for a light weight doesn't work so well in the Bold and that affects quality.

[Dave Crossland]

On 7 September 2016 at 12:32, Thomas Jockin <tho...@thomasjockin.com> wrote:

I agree.

What happens if there are structural differences between the Light and Bold master?

A general observation I'm seeing is compensations that need to done to Bold masters were not taken when they expanded from the Light master.

What is ok for a light weight doesn't work so well in the Bold and that affects quality.

Yep, the original's 'o' is a good example of this, where the darker weights lack optical compensations that are not ideal. 

I guess really all I'm suggesting is a lighter touch :) 

A couple of classic examples spring to mind: 

Futura's Thin is circular while the ExtraBold's 'o' has a more circular/less elliptical form than your proposal:

- http://www.identifont.com/samples2/adobe/FuturaLight.gif

- http://img.font.downloadatoz.com/download/imgs/f/u/t/futura-lh-extra-bold-upper.png

Another classic example, Gill Sans Kayo is even darker but also has a more circular counter contour than Gill Sans Bold:

- https://en.wikipedia.org/wiki/Gill_Sans#/media/File:Gill_Kayo.png

[Thomas Jockin]

Taking in feedback from Reed Reibstein and Dave

(You can see Reed's comments on Dribbble )

Added more joint arching to harmonize the /g with the /n family. 

Added more of an overbite to the /e 

Widened the half round family members [/a/g/d/b]

Thanks Reed and Dave for your input!

[Thomas Jockin]

Got caught up with meetings and traveling today; Not much progress today.

Did do sketching that some swash caps for the italics would work well with this design.

My plan is to complete Roman work Wednesday and start Italics.

-TJ

[Thomas Jockin]

Worked on spacing and weight/ contrast work for the Romans today.

I'll do the fancy animation demos tomorrow as I do last touches and expand the character set.

-TJ

[Thomas Jockin]

Expanded the Bold master to Google Pro Character Set.

Taking long than planned. Will need the weekend to complete Bold Master set.

Working on Light master next week.

[Dave Crossland]

Please share some images :)

[Thomas Jockin]

Started working on the Character Set Expansion for The Light Master today.

[Thomas Jockin]

Wrapping Up the Character Set work for the Light Mater today.

Will do kerning + weight instances tomorrow. 

[Dave Crossland]

Great, thank you Thomas!

[Beka Buckley]

While you may be delighted, as a user tapping into Google's auto fonts this is a design aesthetic disaster!

This overwrite has totally defaced my website brand aesthetic.

Why did you have to overwrite the original design and not leave it as a newer, optional version for folks to *choose* rather than be dictated to.

Maybe why we chose it was because we liked + accepted it's original quirkiness. I cannot see how this will make it more appealing. You killed off a niche. Congratulations!

Now it is ruined.

[Dave Crossland]

Hi Beka

On Sep 6, 2017 7:31 PM, "Beka Buckley" <bekajb...@gmail.com> wrote:

While you may be delighted, as a user tapping into Google's auto fonts this is a design aesthetic disaster!

This overwrite has totally defaced my website brand aesthetic.

Thanks for making an effort to provide this feedback here :)

Why did you have to overwrite the original design and not leave it as a newer, optional version for folks to *choose* rather than be dictated to.

The old version had many problems which are now fixed, along with much wider language support. 

Maybe why we chose it was because we liked + accepted it's original quirkiness. I cannot see how this will make it more appealing. You killed off a niche. Congratulations!

Now it is ruined.

Would you be willing to describe the specific features of the old version that you are missing? 

--

Cheers
Dave

[Dave Crossland]

Hi all

I chatted with Beka offline and the light weight had the style she was looking for :)

[Thomas Linard]

Hi,

There is definitively a problem with Josefin Sans v2.00.

I checked the different binaries in https://github.com/ThomasJockin/JosefinSansFont-master (four different sets!) and the new binaries in https://github.com/google/fonts/tree/master/ofl/josefinsans

The binaries in "instances" and "for iKern" are the same (loose kerning).

The binaries in "iKern" and "version-1.000" are the same (better kerning, except for Bold version, which has the kerning of "instances" version).

The v2.00 has a more loose kerning than the loose kerning of the "instances" version, and the weight distribution is a mess.

See the example PDFs (Thin, Light, Regular, SemiBold, Bold, at 60 pt, metrical kerning, same InDesign source document).

[Dave Crossland]

What are the 5 lines of each PDF? 

[Thomas Linard]

Hi Dave,

Thin, Light, Regular, SemiBold, Bold. Always in the same order (same InDesign source document).

[Dave Crossland]

Hi

In your PDF for v2 I see the Bold being much narrower than on the web.... 

On 12 September 2017 at 16:29, Thomas Linard <thli...@gmail.com> wrote:

Hi Dave,

Thin, Light, Regular, SemiBold, Bold. Always in the same order.

On Tuesday, September 12, 2017 at 4:27:07 PM UTC+2, Dave Crossland wrote:

What are the 5 lines of each PDF? 

--
You received this message because you are subscribed to the Google Groups "Google Fonts Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to googlefonts-discuss+unsub...@googlegroups.com.
To post to this group, send email to googlefonts-discuss@googlegroups.com.
Visit this group at https://groups.google.com/group/googlefonts-discuss.
To view this discussion on the web visit https://groups.google.com/d/msgid/googlefonts-discuss/dad3ca0a-97c7-4776-95a1-74697b357b7d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Cheers
Dave

[Thomas Linard]

Yes, I'm sorry. I did my tests again: I had mishandled a cache phenomenon. I remade the documents V100 and V200: it's less bad than I had initially reported, but the kerning and the distribution of weights is still, in my opinion, not good.

Attached: the new PDFs.

On Tuesday, September 12, 2017 at 4:59:26 PM UTC+2, Dave Crossland wrote:

Hi

In your PDF for v2 I see the Bold being much narrower than on the web....