The FONT Project

[The Hobbyist]

I am looking for people who either have exceptional computer coding skills, or at least spare money to HIRE such people, in order to successfully create The FONT Project.

ALL FONTS are just shapes, regardless of their size or proportions, despite being italicized, or stretched, or otherwise distorted. ALL FONTS can be drawn, scaled and manipulated by a computer when they are in VECTOR format.

We are constantly fighting the "What FONT is this?" battle, over and over again, endlessly. I want to end this battle.

I am looking for people who are crackerjack coders, people who THINK in computer languages. The FONT Project's mission is to examine every character of every FONT, to accurately plot that individual character's digital fingerprint, which would consist of a digital array of values (1's and 0's).

Then, when searching for the name of a font, you would upload a single character to a website. The character would be loaded into a grid that would do a forensic examination of the character's shape, to determine it's fingerprint. From there, it should be a simple matter of searching a database to identify the character, and thus, the FONT.

I do not know enough about coding to do this myself. I am left-handed, so I think in pictures. Right-handed people think in LOGIC, 1's and 0s !! They are the people who could make this project a reality.

Imagine a site, where you load a single character, or even a jpg image of a character, and the site quickly returns,

Font Name: Almost Heaven
Character: Lowercase 'r'
Status: Bold Face
Alteration: Italics
Stroke: 2%
Other: Stretch Height 15%

I believe this CAN be accomplished. After all, the FBI can find one person among seven billion people, by digitally plotting key points in a fingerprint, and then returning the name(s) of the closest matches, IF that fingerprint is in the database. I do not see this as being more complicated than that.

Joe

 

[Pauly]

There's websites like this already:
https://www.whatfontis.com/
https://www.myfonts.com/WhatTheFont/

The thing is, There are so many custom fonts out there created by different people. A font finder wont know EVERY font. It wont know the fonts Johnny created if they're not in the data base.
For example, some fonts are from adobe. If you tried to search an adobe font on one of those sites, there's a change it'll come up empty as those sites may not have the licencing to use the adobe fonts.

For your idea to work, You would need a licence for every single font type kit available and constantly updated with new fonts.
Theoretically it's possible....
Logically, you'll be running into limitations quickly, especially with licencing.

 

[Bobby H]

What you're asking for is not realistic at all.

The "What The Font" tool at MyFonts' web site already does the things you describe. I think there are other tools online that serve a similar purpose. I'm skeptical a crowd-sourced rival to MyFonts' What The Font tool is going to be any better.

There are multiple variables why this font project effort can't succeed. First, the sample images posted in this forum vary widely in quality. Some would work in automated tools like What the Font. But other images are so poor in quality that it depends on the human mind to make abstract judgments that are impossible for computers.

Ultimately this font project thing would depend on all typefaces being coded into some kind of grand database. There are many commercial vendors of type. I think MyFonts is the biggest online store of commercial type, but it doesn't cover everything. That can lead to tools like What the Font not bringing up a correct match for certain uploaded type samples. Then there is a bigger problem of all the many thousands of free fonts being produced and hosted at sites like DaFont. None of those fonts are covered by tools like What the Font, which only search commercially sold typefaces available on their web site. There is absolutely no way anyone could expect all people who design typefaces, free or commercial, to submit their designs to some ultimate database of fonts. It's just not going to happen.

 

[jtiii]

I am looking for people who either have exceptional computer coding skills, or at least spare money to HIRE such people, in order to successfully create The FONT Project.

This is a joke, right?

Right-handed people think in LOGIC, 1's and 0s !!

Okay this is obviously a joke.

I do not know enough about coding to do this myself. I am left-handed, so I think in pictures.

I'm guessing thousands of left handed programmers would disagree with you here.

 

[Gino]

Cool idea..... let's re-invent the wheel, too.

 

[James Burke]

Cool idea..... let's re-invent the wheel, too.

Sorry Gino....that's already been done.

JB

 

[Gino]

Oh, that must be in that section about people who can't do anything but wanna be a part. Just read the policies.

 

[Humble PM]

Cool idea..... let's re-invent the wheel, too.

All right, Mr. Wiseguy,” she said, “you’re so clever, you tell us what color it should be.”

 

[Notarealsignguy]

Sorry Gino....that's already been done.

JB

View attachment 159271

Those are texas tires

 

[Gino]

barkleysumt......mrs humblebee

 

[Fred Weiss]

Reentered in next post

 

[Fred Weiss]

I am looking for people who either have exceptional computer coding skills, or at least spare money to HIRE such people, in order to successfully create The FONT Project.

ALL FONTS are just shapes, regardless of their size or proportions, despite being italicized, or stretched, or otherwise distorted. ALL FONTS can be drawn, scaled and manipulated by a computer when they are in VECTOR format.

We are constantly fighting the "What FONT is this?" battle, over and over again, endlessly. I want to end this battle.

I am looking for people who are crackerjack coders, people who THINK in computer languages. The FONT Project's mission is to examine every character of every FONT, to accurately plot that individual character's digital fingerprint, which would consist of a digital array of values (1's and 0's).

Then, when searching for the name of a font, you would upload a single character to a website. The character would be loaded into a grid that would do a forensic examination of the character's shape, to determine it's fingerprint. From there, it should be a simple matter of searching a database to identify the character, and thus, the FONT.

I do not know enough about coding to do this myself. I am left-handed, so I think in pictures. Right-handed people think in LOGIC, 1's and 0s !! They are the people who could make this project a reality.

Imagine a site, where you load a single character, or even a jpg image of a character, and the site quickly returns,

Font Name: Almost Heaven
Character: Lowercase 'r'
Status: Bold Face
Alteration: Italics
Stroke: 2%
Other: Stretch Height 15%

I believe this CAN be accomplished. After all, the FBI can find one person among seven billion people, by digitally plotting key points in a fingerprint, and then returning the name(s) of the closest matches, IF that fingerprint is in the database. I do not see this as being more complicated than that.

Joe

Just to bring you up to date ... In 1988, with the cooperation of most of the type foundries, my company brought a program to market named FontFinder for the purpose you suggest. It worked just fine and we advertised it at trade shows and in various trade magazines. Sign people in general shrugged and said "so what". The only folks who recognized its value were the professional document examiners and law enforcement. We sold licenses to the FBI, the Secret Service and lots of police departments. But sign people, printers and graphic artists sat by and watched.


FontFinder was eventually discontinued and a couple of others have found their way to market. The current program of choice is Find My Font.
As the owner and manager of Signs 101 for 14 years, I received many complaints from members about the lack of a need to have font identification at Signs 101. I disagreed and to this day you can upload a type sample and get it Identified in short order.


There is no need to develop a new way to identify fonts.

 

[The Hobbyist]

There's websites like this already:
https://www.whatfontis.com/
https://www.myfonts.com/WhatTheFont/

The thing is, There are so many custom fonts out there created by different people. A font finder wont know EVERY font. It wont know the fonts Johnny created if they're not in the data base.

Here in the little town of Franklin, KY., there is the old jail house. it is a limestone building that served as the original jail for Simpson county. Now it is filled with huge, old, mahogany bookcases that reach the high ceilings. Those bookcases are packed with hundreds ... perhaps more than just hundreds of old ledger books, where old historic county records of all types were manually entered as far back as a couple hundred years ago. These books are not what we see today. They are about 30 inches tall x 24 inches wide! All of the entries are by quill pens and other implements of that time. many are faded and require special treatments to bring some contrast back to the page, just so it can be read by the volunteer. it is a long, arduous project that was started many, many years ago, and still continues today. Franklin, KYT is not the ONLY town doing this, by the way.

The task of digitizing ALL of these public records is nothing less than monumental. However, we have a small army of volunteers who patiently and painstakingly go through each book, crisp, aged brown page by crisp, aged brown page, to manually enter the DATA into a computer. Property records, marriage records, birth records, death records, still birth records, crime records, hangings, elections, celebrations, floods, catastrophes, etc. etc. are ALL, e-v-e-r s-o s-l-o-w-l-y being digitized for future use, by this small army of volunteers.

Yes, there are a LOT of fonts, but there are also tens of millions of people in America who are savvy with computers. If a central database was established, and FONTS were gathered and catalogued as they came in, and there was a simple downloadable utility that EVERY PC and MAC owner could use to manually scan every character of every font, it wouldn't take long at all to digitize the fingerprints of every conceivable character of every FONT.

Such a project could also weed out the duplicates, where three, or ten or 20 different fonts BY NAME are all actually the SAME FONT.

In CommieFornia, they can sell a BILLION LOTTO tickets in a week. Each ticket has at least ONE series of six or seven numbers selected, and most tickets have more than one game on them. Each ticket has a long individual serial number.
Each ticket is recorded by SCANNING the hand drawn black marks on the tickets. When the numbers are drawn, the LOTTO Commission computers know IN A FLASH, exactly how many tickets "hit" either 3 numbers, or four, or five, or all six, or four + the bonus, etc. etc. They know exactly WHERE the tickets were sold. They know EXACTLY how much money was taken in, and exactly what must be paid out. INSTANTLY.

Please do not tell me that an army of TENS OF MILLIONS of computer geeks around the world cannot convert all of the existing FONTS into individual fingerprint records using their home computers, iPhones, iPads, and other devices. I refuse to accept that this is not possible. I am GUESSING that it could be done in less than a MONTH.

A database of KNOWN FONTS would be listed. Those that HAVE been digitized into a fingerprint file would be GREEN, for example. Those that have NOT been digitized would be RED.

So you log into The FONT Project. You look at the list of FONTS that need to be digitized, and you randomly pick one called Eddie's Scrawl. You download eddiesscrawl.ttf and open it in the utility on your computer. It displays the individual characters of that FONT. The utility has a crosshair marquis, and you begin ....

Instruction: "Draw a marquis around the Upper Case A and press RETURN..."

You do it, and a few seconds later, it prompts you to draw a marquis around the Upper Case B, etc. etc. Rinse and repeat.

You get through all of the upper and lower case letters and numbers, the punctuations and anything else the FONT may contain.

"Congratulations! The FONT eddiesscrawl.ttf has its own fingerprint!"

325667. eddiesnightmare.ttf - Conversion Needed
325668. eddiescreamer.ttf - Conversion Needed
325669. eddiesscrawl.ttf - by John Doe - Anytown, USA
325670. eddiemunster.ttf - Conversion Needed

I know what you are going to say now ... "Who PAYS ME for this service?!"

What's in it for YOU? God forbid you use some free time to help humanity.

"It is amazing what man can accomplish, when no one cares who gets the credit." - Ronald Reagan

Joe



.

 

[The Hobbyist]

What you're asking for is not realistic at all.

https://www.fedex.com/en-us/about/history.html

There are multiple variables why this font project effort can't succeed. First, the sample images posted in this forum vary widely in quality. Some would work in automated tools like What the Font. But other images are so poor in quality that it depends on the human mind to make abstract judgments that are impossible for computers.

I agree that CURRENTLY, computers have trouble with raster images, and deciding what is "black" and what is "white" when trying to see the edge of a fuzzy letter. Better CODING would resolve that issue.

Ultimately this font project thing would depend on all typefaces being coded into some kind of grand database.

Right. It would be called The FONT Project.

There are many commercial vendors of type. I think MyFonts is the biggest online store of commercial type, but it doesn't cover everything. That can lead to tools like What the Font not bringing up a correct match for certain uploaded type samples. Then there is a bigger problem of all the many thousands of free fonts being produced and hosted at sites like DaFont. None of those fonts are covered by tools like What the Font, which only search commercially sold typefaces available on their web site. There is absolutely no way anyone could expect all people who design typefaces, free or commercial, to submit their designs to some ultimate database of fonts. It's just not going to happen.

It Couldn’t Be Done​

BY EDGAR ALBERT GUEST
Somebody said that it couldn’t be done
But he with a chuckle replied
That “maybe it couldn’t,” but he would be one
Who wouldn’t say so till he’d tried.
So he buckled right in with the trace of a grin
On his face. If he worried he hid it.
He started to sing as he tackled the thing
That couldn’t be done, and he did it!

Somebody scoffed: “Oh, you’ll never do that;
At least no one ever has done it;”
But he took off his coat and he took off his hat
And the first thing we knew he’d begun it.
With a lift of his chin and a bit of a grin,
Without any doubting or quiddit,
He started to sing as he tackled the thing
That couldn’t be done, and he did it.

There are thousands to tell you it cannot be done,
There are thousands to prophesy failure,
There are thousands to point out to you one by one,
The dangers that wait to assail you.
But just buckle in with a bit of a grin,
Just take off your coat and go to it;
Just start in to sing as you tackle the thing
That “cannot be done,” and you’ll do it.

 

[The Hobbyist]

Just to bring you up to date ... In 1988, with the cooperation of most of the type foundries, my company brought a program to market named FontFinder for the purpose you suggest. It worked just fine and we advertised it at trade shows and in various trade magazines. Sign people in general shrugged and said "so what". The only folks who recognized its value were the professional document examiners and law enforcement. We sold licenses to the FBI, the Secret Service and lots of police departments. But sign people, printers and graphic artists sat by and watched.


FontFinder was eventually discontinued and a couple of others have found their way to market. The current program of choice is Find My Font.
As the owner and manager of Signs 101 for 14 years, I received many complaints from members about the lack of a need to have font identification at Signs 101. I disagreed and to this day you can upload a type sample and get it Identified in short order.


There is no need to develop a new way to identify fonts.

Hi Fred, We have discussed this before, elsewhere. I disagree that there is no room for improvement. I do not know how your project determines what character belongs to what font, but I go by the old adage, "There is more than ONE WAY to skin a cat!"

I am amused by your royal proclamation that there is no need for something better than what is currently available. That is simply ridiculous, because what is currently available (as good as it may be) has a LOT of room for improvement. Ah, but then this probably bruises egos, and those things are so delicate.

Joe



.

 

[Humble PM]

Two birds, possibly three.

Remember those tricksy things on websites, where you have to guess what the letters are, before you can get to the next level? Well... replace those silly old hand written texts with first draw a box around a letter A, then a letter a, now a letter B and a letter b and so on.

Sell this project to, I don't know, perhaps a large interweb firm. call it something snappy like Project GuBattenburg.

Bonus points for identifying serif or lack of.
Even better bonus points for getting all involved to do the work pro-bono.

I used to quite like cats, until they started to remind me of politicians.
I'm dexter, and cant code for toffee.

 

[Fred Weiss]

Well Joe, the fact is that for $49 you can license a copy of the Find My Font program. With it you get a program that simplifies and automates the identification of any type sample and a comprehensive database that is regularly added to and updated. You can also add in your own fonts to insure that they're included in any ID search. The program will also sort out the various ways a type sample may have been modified ... slanting, condensing etc. They do offer a 30 day free trial so my suggestion to you is to download and use the free version and see just how well it handles the task of identifying fonts.

I suppose that just makes me a nasty capitalist but Find My Font has already created a solution to what you are seeking without involving humanity in seeking the same. Putting a price tag on a product is how these things get done.

 

[The Hobbyist]

Bonus points for identifying serif or lack of.
Even better bonus points for getting all involved to do the work pro-bono.

My thought is that serif or sans serif will not matter one bit. Each individual shape will have its own unique fingerprint.

We are told that NO TWO HUMAN FINGERPRINTS are ever the same. Throughout history! Tens of billions of people who are alive NOW, and those who have passed, and yet, no two fingerprints are exactly the same.

We are told that NO TWO SNOWFLAKES are exactly the same. Untold sextillions of snowflakes have fallen on the Earth over the past few thousand years, and yet no two snowflakes are ever exactly the same.

But coming up with a digital fingerprint for a few hundred million, or a few BILLION font characters is completely out of the question.

So sad, that we live in a society where no one is willing do do anything that might benefit humanity at large unless they get paid for it.

Joe



.

 

[The Hobbyist]

Well Joe, the fact is that for $49 you can license a copy of the Find My Font program. With it you get a program that simplifies and automates the identification of any type sample and a comprehensive database that is regularly added to and updated. You can also add in your own fonts to insure that they're included in any ID search.

I did try it. It succeeded about 50% of the time.

The program will also sort out the various ways a type sample may have been modified ... slanting, condensing etc. They do offer a 30 day free trial so my suggestion to you is to download and use the free version and see just how well it handles the task of identifying fonts.

I suppose that just makes me a nasty capitalist but Find My Font has already created a solution to what you are seeking without involving humanity in seeking the same. Putting a price tag on a product is how these things get done.

If your program has SOLVED the Font identification problem, why are community forums littered with "Help me identify this font!" posts? It would seem that this problem should no longer exist.

Joe



.

 

[Notarealsignguy]