Cormorant

[Christian Thalmann]

Hi everybody,

better late than never! This is the official thread on the Cormorant typeface project.

Bēhance: https://www.behance.net/gallery/28579883/Cormorant-an-open-source-display-font-family
GitHub: https://github.com/CatharsisFonts/Cormorant

Cheers!

— Christian Thalmann

[Christian Thalmann]

I just issued a new release of Cormorant, including Alexei's overhaul of the Cyrillic and a fix to the missing characters in the small-caps spin-off fonts. A few minor things as well, such as new designs of /A and /R with more relaxed counters for the Garamond style.

https://github.com/CatharsisFonts/Cormorant/releases/tag/v3.3

Can we please push this to Google Fonts as soon as possible? I'm mostly bothered by the missing characters bug in the version currently on GF, but usage-wise, the polished Cyrillic is probably more important...

[Dave Crossland]

Hi

Thanks Christian! Marc will do a final review (along with other Cyrillic improvements projects) and we'll get this live soon :)

Cheers,
Dave

--
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/6d888141-7551-47f2-875a-5a26f1dbd6ad%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Thomas Linard]

Hi,

I don't see Cormorant in today update batch, is it expected soon?

[Christian Thalmann]

The live commit is ready as far as I'm concerned. 

Cheers

Sent from my iPhone

--
You received this message because you are subscribed to a topic in the Google Groups "Google Fonts Discussions" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/googlefonts-discuss/wmHBh2k_eu0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to googlefonts-dis...@googlegroups.com.
To post to this group, send email to googlefon...@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/45d3df45-1c19-4007-87df-f25b0017e846%40googlegroups.com.

[Dave Crossland]

On 17 January 2017 at 03:39, Christian Thalmann <christian....@gmail.com> wrote:

The live commit is ready as far as I'm concerned. 

Marc is in charge of this :D Marc? :) 

[Marc Foley]

Hey folks,

Yeah, I let Christian do some final tweaks. I'm very happy with it!

I'll pr it to fonts/google today

[Christian Thalmann]

Hey guys — thanks for updating Cormorant on Google Fonts! That was quick.  :)

But something seems to have gone wrong in Cormorant Garamond. Many accents seem misaligned compared to the Cormorant forms:

Cormorant:

Cormorant Garamond:

Marc, does this have something to do with your changes to the metrics?

[Christian Thalmann]

On closer inspection, not only are the accents hovering, but also the horizontal metrics are off. The Ć seems wider than the C, for instance. This needs to be fixed ASAP.

This is what /A/Aacute/A looks like in Cormorant Garamond Regular (TTF) after a fresh export. I don't get it — it doesn't look like that in the Glyphs source file, and the default Cormorant cut doesn't have that problem.

[Christian Thalmann]

One more piece of the puzzle: This effect appears in all the interpolated instances, but not the ones corresponding to masters.

[Christian Thalmann]

Mekkablue suggests it's the effect of a bug in Glyphs, which seems to have been fixed in the meantime according to Glyphs' version history. He suggests we restore the pre-scaling version and redo the scaling.

https://forum.glyphsapp.com/t/broken-metrics-in-interpolated-accented-characters/5823/2

Marc, I assume most of your work was figuring out the metric numbers, right? Now that you have them, would it be a reasonably quick matter to redo the scaling?

I'll also have to apply my post-scaling changes again (mostly Garamond's /R), but that's a detail...

[Christian Thalmann]

Actually, now I'm hoping that the error happens in the export stage in Glyphs, and that our source files are actually not corrupted (see Glyphs forum thread). Let's hope this gets fixed soon.

I wonder how many websites now suffer from these broken characters in the meantime...

[Marc Foley]

Hey Christian,

Thanks for reporting this.

This is partly the reason why I scaled it to 1000upm, to try and solve this. :-). Exporting your older version you was <1000upm had even worse resutts.

Imo, this is an issue with .glyphs. We can brute force this as much as want but I think it's best we show Georg this.

[Christian Thalmann]

Agreed. Maybe we can export the fonts with an earlier version of Glyphs as a quick fix. I'll talk to the Glyphs guys.

What's the next possible time for updating GF? I hate the feeling of serving a broken version. :Þ

[Christian Thalmann]

Alright, Georg was very helpful over on the Glyphs boards.

From what he can tell, we ran afoul of a bug in the «Scale to UPM» function that Glyphs briefly had before it got fixed. Marc probably used that function on Cormorant while it was buggy, so a bunch of composed glyphs got their alignment data scrambled in the process. Apparently, simply scrolling through the entire glyph inventory in the font view in a new version of Glyphs is enough to trigger automatic re-alignment and fix the problem.

I have re-exported all fonts and uploaded them to GitHub. The exports look good to me now. Could we please push this version to GF as soon as possible?

[Dave Crossland]

On Jan 21, 2017 10:56 AM, "Christian Thalmann" <christian....@gmail.com> wrote:

Alright, Georg was very helpful over on the Glyphs boards.

From what he can tell, we ran afoul of a bug in the «Scale to UPM» function that Glyphs briefly had before it got fixed. Marc probably used that function on Cormorant while it was buggy, so a bunch of composed glyphs got their alignment data scrambled in the process. Apparently, simply scrolling through the entire glyph inventory in the font view in a new version of Glyphs is enough to trigger automatic re-alignment and fix the problem.

LOL

I have re-exported all fonts and uploaded them to GitHub. The exports look good to me now. Could we please push this version to GF as soon as possible?

Yes, Marc can make a PR and I'll get it pushed as soon as I can. 

Felipe and Marc, I would like you to think about ways to prevent this from happening again for any other updates. I will have to go back to slowly updating fonts after carefully checking them myself by hand if our automation isn't effective.... 

[Christian Thalmann]

Yes, Marc can make a PR and I'll get it pushed as soon as I can.

Phew, good to hear.
 

Felipe and Marc, I would like you to think about ways to prevent this from happening again for any other updates. I will have to go back to slowly updating fonts after carefully checking them myself by hand if our automation isn't effective.... 

Well, a couple of days of a broken version being served is probably still a tolerable risk compared to other fonts with bugs waiting years to be updated due to a slow manual pipeline....

[Marc Foley]

Hey Christian,

I will pr this today for you. I do apologise for what has happened. I can understand this may have caused some stress.

Felipe and Marc, I would like you to think about ways to prevent this from happening again for any other updates.

Hey Dave,

This was an issue with Glyphs, nothing me and Christian did. I think the best solution will be for every new version of Glyphs which is released, I'll do a regression test and use Noto tool's compare_fonts.py to see if Glyphs is exporting fonts consistently.

About a week ago I wrote, fonntbakery-check-vtt-compatibility.py. This will tell me if composite coordinates and points have changed between two different fonts. I may rename this script so it indicates it is far more general purpose than the file names suggests.

I hope this clears things up.

Cheers,

Marc

[Christian Thalmann]

On Monday, January 23, 2017 at 11:14:57 AM UTC+1, Marc Foley wrote:

I will pr this today for you. I do apologise for what has happened. I can understand this may have caused some stress.

No worries; I'm glad we got it under control so quickly.

I've regenerated the exports with new version numbers, so we should now be ready for publication.

[Dave Crossland]

On Jan 23, 2017 5:14 AM, "'Marc Foley' via Google Fonts Discussions" <googlefon...@googlegroups.com> wrote:

Hey Christian,

I will pr this today for you. I do apologise for what has happened. I can understand this may have caused some stress.

Felipe and Marc, I would like you to think about ways to prevent this from happening again for any other updates.

Hey Dave,

This was an issue with Glyphs, nothing me and Christian did. I think the best solution will be for every new version of Glyphs which is released, I'll do a regression test and use Noto tool's compare_fonts.py to see if Glyphs is exporting fonts consistently.

How to get this set up to run automatically? 😁

Perhaps we can ask Georg and his team to do it?

About a week ago I wrote, fonntbakery-check-vtt-compatibility.py. This will tell me if composite coordinates and points have changed between two different fonts. I may rename this script so it indicates it is far more general purpose than the file names suggests.

I agree about rename, and yes, this sounds like a great tool for assuring quality for updates by highlighting regressions

[Marc Foley]

Hey Christian,

Thank you so much for the version number increment. I'll get this out the door now.

Cheers,

Marc

[Christian Thalmann]

How long will it take until the new version is live?

[Dave Crossland]

I can't comment on specifics but soon

Cheers,
Dave

--

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/14b83a5f-f2cd-4265-9a54-9bf411255eba%40googlegroups.com.

[Christian Thalmann]

The fix seems to be live now. Phew! Thanks for that. :)

[Dave Crossland]

Yes it is now live in production. Please let me know if you see any other issues/regressions :) 

[Christian Thalmann]

On Wednesday, January 25, 2017 at 11:01:21 PM UTC+1, Dave Crossland wrote:

Yes it is now live in production. Please let me know if you see any other issues/regressions :) 

Now that you mention it...

https://github.com/CatharsisFonts/Cormorant/issues/31

Arrrgh! It's the same Glyphs bug as before. I must have missed it in my previous scroll-through. Those Cyrillic caps snapped into place when I visited them with the current version of Glyphs. I'm exporting a new version right now.

I'm afraid we'll need another PR as soon as possible. Sorry for the mess.

[Christian Thalmann]

The issue should be fixed in the live version on GitHub now.

[Dave Crossland]

Hi 

Right :) I'm guessing that this will take some time. 

Marc, please add this to your queue, and please write something that will catch this class of regression in future :)

On Feb 4, 2017 5:36 AM, "Christian Thalmann" <christian....@gmail.com> wrote:

The issue should be fixed in the live version on GitHub now.

--

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/6249f59c-302e-49f3-aa03-74b09508f0ff%40googlegroups.com.

[Marc Foley]

Will do it first thing on Monday,

I'll write an accent shifting regression test as well.

[Christian Thalmann]

Thanks Marc! I hope the bug won't turn too many Cyrillic users away from Cormorant in the meantime...

As for an accent shifting test, note that this latest surviving artefact was not related to accents, but simply to components (Latin caps used for Cyrillic caps of the same shape).

[Marc Foley]

Hey Christian,

Thanks for the heads up. I'm on it now.

Cheers,

Marc

[Christian Thalmann]

On Saturday, February 4, 2017 at 7:23:44 PM UTC+1, Dave Crossland wrote:

Right :) I'm guessing that this will take some time.

Longer than last time? How frequently are these GF updates made?

[Dave Crossland]

I can't provide specific future dates but there is a queue of other things that come first and until those are done, Cormorant is in line :)  

[Christian Thalmann]

Thanks for the info, Dave!

[Christian Thalmann]

Wow, this IS taking a long time. :(

[Dave Crossland]

Hang in there :)

On Feb 23, 2017 5:32 PM, "Christian Thalmann" <christian....@gmail.com> wrote:

Wow, this IS taking a long time. :(

--
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/e6ad022d-88ed-455c-81f2-4cc72e1dbf36%40googlegroups.com.

[Christian Thalmann]

Cormorant has just been spotted in print. That's all kinds of awesome! :)

https://twitter.com/NathanZimet/status/835631801976619009

[Thomas Linard]

Congratulations!

[Christian Thalmann]

I just got the book in the mail, and it is gorgeous.  :)

There are just a few instances where the spacing of the apostrophe feels a bit too tight, especially after /f.short (such as in «Chef’s»). I improved that in the live version of Cormorant now.

Since Cormorant is still not updated on GF, I'm hoping that we can include this update in the PR.

BTW, why is it taking so incredibly long? Shouldn't it be a matter of minutes? Can I help with the process? Whom do I have to bribe?  :Þ

[Dave Crossland]

On 15 March 2017 at 15:43, Christian Thalmann <christian....@gmail.com> wrote:

I just got the book in the mail, and it is gorgeous.  :)

There are just a few instances where the spacing of the apostrophe feels a bit too tight, especially after /f.short (such as in «Chef’s»). I improved that in the live version of Cormorant now.

Awesome!

 

Since Cormorant is still not updated on GF, I'm hoping that we can include this update in the PR.

Marc, please queue this up after the other stuff I've asked you to queue up :) 

Christian, I expect Marc will get to it in April. 

 

BTW, why is it taking so incredibly long? Shouldn't it be a matter of minutes? Can I help with the process? Whom do I have to bribe?  :Þ

Running a web API at the scale of Google Fonts (https://fonts.google.com/analytics) is not something you can help with, and is very complex.  

[Christian Thalmann]

BTW, why is it taking so incredibly long? Shouldn't it be a matter of minutes? Can I help with the process? Whom do I have to bribe?  :Þ

Running a web API at the scale of Google Fonts (https://fonts.google.com/analytics) is not something you can help with, and is very complex.  

That is one impressive statistic. :)

I realize it's an incredible amount of work overall. But given the very quick cycle of the previous emergency update for Cormorant, I was under the impression that the infrastructure was so well established already that replacing a file would only take a command line or two. As far as I know, Marc's manual mastering of Cormorant is now part of the repo and won't have to be redone every time there's an update. I guess it's more complicated than that in real life.

[Dave Crossland]

https://youtu.be/R7m5Int1hAA?t=1m4s 

[Christian Thalmann]

To celebrate the official recognition of the ẞ character in German, I've reworked it in Cormorant. For someone who often complains about ugly approaches to ẞ, I sure made a bad example there! New version below.

When/if you update Cormorant on GF, please use the live version rather than a release to ensure these changes are included.

[Adam Twardoch (List)]

Christian,

I actually like the top form more than the bottom form. :)

Sent from my mobile phone.

On 30. Jun 2017, at 20:47, Christian Thalmann <christian....@gmail.com> wrote:

To celebrate the official recognition of the ẞ character in German, I've reworked it in Cormorant. For someone who often complains about ugly approaches to ẞ, I sure made a bad example there! New version below.

<Auto Generated Inline Image 1>

When/if you update Cormorant on GF, please use the live version rather than a release to ensure these changes are included.

--

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-dis...@googlegroups.com.
To post to this group, send email to googlefon...@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/27f099db-6d52-4863-82b9-5d25b11e3be8%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

<Auto Generated Inline Image 1>

[Dave Crossland]

Hi Christian

Great to see your support for this new glyph! 

Marc has said that prepping releases for your project takes up a lot of his time, so I think it's reasonable for us to only update the GF version to tagged releases which you've done QA on to ensure there are no regressions. 

Marc has built a nice regression testing tool and Lasse and Felipe are working to make a general purpose version available with other font bakery checking, so in the future this will be a little easier :)

--

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/27f099db-6d52-4863-82b9-5d25b11e3be8%40googlegroups.com.

[Christian Thalmann]

On Friday, June 30, 2017 at 11:33:59 PM UTC+2, Dave Crossland wrote:

Marc has said that prepping releases for your project takes up a lot of his time,

Isn't the remastering something that only needs to be done once? The UPM is now fixed at 1000 in the Glyphs files, after all.
 

so I think it's reasonable for us to only update the GF version to tagged releases which you've done QA on to ensure there are no regressions.

Fair enough. Please warn me in advance when you're closing in on an update, then, so I can push out a release with the latest changes.
 

Marc has built a nice regression testing tool and Lasse and Felipe are working to make a general purpose version available with other font bakery checking, so in the future this will be a little easier :)

Good to know! 

[Christian Thalmann]

On Friday, June 30, 2017 at 11:29:12 PM UTC+2, Adam Twardoch wrote:

I actually like the top form more than the bottom form. :)

We'll have to disagree on that one, then. :Þ  The previous form is still stored in a layer in the Glyphs file in case you want to make a custom branch of the fonts. ;o)

[Thomas Linard]

Am I the only one to think that the design "Zehlendorf" is more suitable for a Garalde?
https://typography.guru/journal/capital-sharp-s-designs/

[Christian Thalmann]

On Saturday, July 1, 2017 at 10:39:46 AM UTC+2, Thomas Linard wrote:

Am I the only one to think that the design "Zehlendorf" is more suitable for a Garalde?
https://typography.guru/journal/capital-sharp-s-designs/

Zehlendorf is my second-favorite solution. I consider the doubly curved right side a bit too decorative for most «sober» typefaces.

I fully agree with the overall proportions, though, which is why I'm promoting the «Zürich form», which is essentially the Leipzig→Zehlendorf transform applied to Dresden. :)

[Christian Thalmann]

I made a new release for version 3.504. It fixes the previously broken /f_f_f ligature. Can't believe that survived so long! I wonder whether it was a recent regression.

[Thomas Linard]

Hi Christian,

Thanks for your answer!

[Christian Thalmann]

Uploaded a new release of Cormorant. The main addition is support for Pe̍h-ōe-jī (POJ) as requested, and reinstating correct implementation of ligatures in the Italic (the LIGA feature had been reverted to auto-generated for some reason, which doesn't play well with CALT). A few smaller touch-ups here and there as well.