Hacker News new | past | comments | ask | show | jobs | submit login
One man's obsession with rediscovering the Doves typeface (bbc.co.uk)
80 points by tankenmate on July 26, 2015 | hide | past | web | favorite | 32 comments

The Doves Type font can be seen at this site http://www.typespec.co.uk/doves-type/, which als tells the story how it was recovered.

Why doesn't BBC have even a single link to the project page? Do they assume nobody actually cares about the news they read on their website? The entire point of the WWW is to make a network of information with hyperlinks connecting them.

Perhaps because you can buy from the page? They might have an editorial style-guide blocking links that might suggested some kind of inferred commercial endorsement.

I find a lot of online news publications still write as though they're dealing with print.

NYT has made huge strides with this, greatly increasing the amount of visualization and interactivity in their articles.

Some bloomberg articles are downright gorgeous to look at, this brilliant minimalist gradient design.

It does seem ironic that Cobden-Sanderson threw the original type in the river in order to avoid Walker selling it, yet here we are with typespec selling licenses of the digital version for 40 quid.

Last year I found just the typography in a PDF (http://www.sinnebild.se/font/large/dove.pdf) with the typeface embedded. It's a beautiful font that works incredibly well for a number of projects.

Thanks for the link. On the typespec site is a PDF preview of the font and on page 6, there is a more detailed and also concise summary of the history of the type from Robert Green himself. It's worth a read.

He finishes with this..

'In reading about the Doves Press on the internet there is a danger of consuming the story as fiction. I could not have undertaken this digital recreation or the task of recovering the real type without hard facts. I found these in Marianne Tidcombe's book on the Doves Press. If you have any further interest in the story, please seek it out. ―Robert Green, 6th March 2015'

I'm assuming he means 'The Doves Bindery' by Marianne Tidcombe which is currently selling on Amazon for $287. Almost cheaper to hire a diver and find some more type yourself (I joke)


No I mean the Doves Press which Marianne Tidcombe wrote after her work on the Doves Bindery. I'll admit it is expensive at around £70, but I've seen copies for £40 http://www.amazon.co.uk/The-Doves-Press-Marianne-Tidcombe/dp...

Why is the dot on the i so small and off center when the diacritics are all large, centered diamonds?

I wonder, how was the "i" cut in lead with the dot being off-center? Wouldn't this cause the letter to be unusually wide? Or was it cut as some kind of ligature, which I find hard to imagine.

Addendum: there is a picture of an "i" at http://www.typespec.co.uk/recovering-the-doves-type/. The "i" is cut with the dot at the right edge of the metal plate that holds it

Huh, interesting bit from that site, about purchasing the font:

> It is also available in web font formats for self hosted websites (up to 10K monthly pageviews on average).

Now I can totally understand this feeling. If your art is deployed on a large and popular website with millions of monthly pageviews, you'd like a different (higher) compensation than if it was used on a relatively small time website with less than 10K monthly views.

I'm also aware of the fact that licensing for fonts works rather differently than most other sorts of intellectual property. In IP law, fonts are not copyrighted like graphics/art/design is, but more like a computer program would be[0]. The font file is seen as the computer program (that generates the outlines), rendering text with the font is seen as running the computer program[1], and therefore the creator can request a license for every "user" of their program. Which, in the case of web fonts, are the visitors of the website. This is a legal thing.

A graphics artist cannot license their designs depending on the number of viewers, a least not in the same way. This is because whoever buys the design buys a license to "publish" and "make public" the design (which can of course be negotiated depending on the expected exposure). From there it is (for the most part) out of the hands of the original creator. However the feeling I described in the first paragraph above, is (IMO) just as reasonable for a graphics artist as it would be for a font designer.

Does this make sense? At least for the case of graphic design, the way the law operates (mostly) makes sense in our current digital age.

And websites also contain other code, like Javascript, and they are merely licensed to whoever built the site. Even though the users of the website execute the code in much the same way.

I always find it weird how fonts are such a special case in IP law. I get that creating a font takes a lot of work, but so can large graphics design projects.

[0] Please note that this has nothing to do with the fact that some parts of many font file formats are Turing-complete. It's a purely legal distinction, which does not care about the computer science / theory.

[1] And running a computer program is protected by copyrights because it involves copying the program from storage to working memory and such. At least that's what I learned about Dutch IP law in university. It's a little bit ridiculous, but this is the reason why software licenses can exist.

If you enjoyed reading about the quirky history of different typefaces, like this one, I can reccomend the book 'Just My Type' by Simon Garfield. Each chapter introduces a bit quirky of history of a type face designer along with details of the type they cremated and what it's good for.


Thanks for the recommendation!

A related book would be 'Type Matters!' by Jim Williams - it's actually suggested for me at the bottom of the 'Just my Type' page. It's a fantastic book on typography, layout, font detail and suggested practices for legibility. I own it and would highly recommend it, especially at the price on Amazon (it's over $60 in Australia).


And when you've finished that, you'll be ready for "Type: the Secret History of Letters" by Simon Loxley.


I remember there being an article about this font a few years back, perhaps 2010. I, like most would, fantasised about what it would be like if someone actually retrieved the type. Crazy idea that perhaps this Mr Green also read that article and actually got a mudlarkers permit to (literally) get to the bottom of it.

The article doesn't mention if the 150 retrieved pieces together actually constitute a useable font or if there's any essential letters missing or damage from being underwater for almost a hundred years.

http://www.typespec.co.uk/recovering-the-doves-type/ better pictures of recovered type here.

The Economist had a good article about this in their 2013 Christmas special:


Why couldn't the font be reproduced from printed sources? Seems much easier than to try and fish old, rusty typefaces out of a river and then try to reconstruct them.

That is how he started, but it does not give completely satisfactory results. The article in the Economist addresses this:

> That sounds simple—yet the uneven printing that letterpress-lovers cherish made tracing the type impossible. Once ink hits paper, no single letter is reproduced identically. Guessing the shape of the metal that made the marks takes time and patience. Guess wrong, and the error is imperceptible at first; but lined up in text the letter looks awkward, the typeface distracting.


It's not clear that this font, as a replication of a work out of copyright, is copyrightable in the US. See Bridgeman vs. Corel.

The font files that have been created are copyrightable. Fonts have been classified as computer programmes and as such are afforded copyright protection.

However, the font itself is long out of copyright, so that you or I could create our own version of the font from printed samples by digitising the font.

This happens all the time.

The other issue is that sometimes trademarks can get involved. In that the name of the font could be trademarked and thus the font offered some small protection in that it's hard to then find a duplicate font because it has to be called something different.

True, though if the hinting was done manually or the outlines adjusted for the screen, that's plausibly a new creative work.

The outlines were redrawn manually because there were no drawn outlines as such in existence. It is a new creative work, in that it's my interpretation of the Doves type which only ever existed in metal. It was never redrawn for filmsetting or digital by its original creators because they disposed of it before these technologies came into existence. So the final typeface was never drawn as such, as would have been necessary for adapting it (as so many other types have been) for new reproduction techniques.

The original drawings for the type, by Percy Tiffin, are more like sketches or visuals – guidelines to give the punchcutter a direction rather than precise instructions. They bear little relation to the end product, so they are not the typeface in & of themselves. (I initially tried to use these as reference & after a couple of hours realised that they were pretty useless, containing no uniform measures, proportions or elements). The final 'outlines' for the type were created in the metal itself by the punchcutter, Edward Prince. He is responsible for drawing, or rather carving the form of the type straight onto the steel punches from which the final type moulds (matrices) were cast.

The name Doves Type® is copyrighted to protect my drawings. But if someone wants to go ahead and do what I did and recreate the type from the original printed sources there's nothing to stop them, as long as they do not use my font data as a basis for their font. It took me 5 years, on-&-off, to reach the stage where I was satisfied that I had captured the overall essence of the original. Though one can never recreate the patina of a letterpress type – the appearance of each glyph varies from word to word, line to line, page to page. That's why I prefer to call my digital type a facsimile.

Hi Robert, thanks for your great work on Doves, it's really a treasure.

From what are you protecting your drawings/recreation?

Anyone taking repackaging and selling as theirs. It's my research & work. This is what I do for a living.

Sorry for the off topic but.... F*CK autoplay videos! Couldn't they at least use the page visibility API (or even just requestAnimationFrame) so it doesn't start playing after I've opened a dozen tabs?

Is there a Firefox add-on that enables audio/video and flash only when you see the page?

flash: enable "ask to activate" in the extension settings

<video>: set media.autoplay.enabled = false

you could have googled that instead of whining.

Or it could have been set as a sensible default, since an overwhelming majority of users despise autoplaying videos on sites other than Youtube?

That's not the same as "auto play when page is visible". I don't want to manually enable every video. Also I ask because I'll make such add-on myself if it doesn't already exist.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact