Name:
The Sculptor’s Chisel, The Editor’s Tool: Introducing Typefi Orion
Description:
The Sculptor’s Chisel, The Editor’s Tool: Introducing Typefi Orion
Thumbnail URL:
https://cadmoremediastorage.blob.core.windows.net/ab1d1ba9-bd48-44d3-8bc0-17cd232f08a0/videoscrubberimages/Scrubber_1.jpg
Duration:
T00H28M11S
Embed URL:
https://stream.cadmore.media/player/ab1d1ba9-bd48-44d3-8bc0-17cd232f08a0
Content URL:
https://cadmoreoriginalmedia.blob.core.windows.net/ab1d1ba9-bd48-44d3-8bc0-17cd232f08a0/SSP2025 5-28 1200 - Industry Breakout - Typefi.mp4?sv=2019-02-02&sr=c&sig=kAPMMf%2FDmPcESmzngVUJvGQBioBWPTQ92Yuj75VfMds%3D&st=2025-08-03T00%3A27%3A55Z&se=2025-08-03T02%3A32%3A55Z&sp=r
Upload Date:
2025-06-09T00:00:00.0000000
Transcript:
Language: EN.
Segment:0 .
Hi I'm Caleb closet. I'm with typify. And I'm excited to be here. This is the first time actually, the second time I've been to SSP. It's been many years since I've been back, but to start. So for 18 years, typify has partnered with NRA. And we've picked up where anara leaves off to transform jats bits and STS XML into beautifully composed publications.
But with textiles being sunset next year, we're stepping up and launching project or typify Orion. It's a purpose built drop in replacement for textiles. So the title of my presentation, the sculptor's chisel, the editors mark, might seem a little unusual for software, but I think there's a really interesting connection in art, language, and editorial processes that I'm going to explore. So I just celebrated my 20th year with typify and crafting solutions for publishers around the world.
I've really developed an appreciation for just the complexity in representation and meaning, and our publishing systems don't contain actual knowledge. They contain structured representations of knowledge. This distinction is at the heart of this piece, where Renee Magritte challenges us with the seemingly simple contradiction an image of a pipe, with the text declaring, this is not a pipe. Of course he's correct.
It's not a pipe, it's a painting of a pipe. This beautiful imprecision of language, with its gaps between representation and reality, is what enables us to convey depth and meaning beyond simple fact. The way we frame something shapes how it's understood. Banksy the provocateur responds to Magritte with an actual physical pipe mounted, framed and labeled this is a pipe.
He even mimicked Magritte's cursive script. His juxtaposition is humorous and profound, reminding us that context and presentation fundamentally affect meaning. As editors and publishers. Aren't we engaged in a similar work. We shape understanding through prose and presentation. We decide how knowledge is structured, clarified, and delivered.
The tools we use for this work are not neutral. They influence how we think about and process information. From the conceptual playfulness of Magritte and Banksy. Shifting to the halls of the Galleria in Florence, where Michelangelo's sculptures provide a profound meditation on the nature of representation. Here, his prisoners struggle to emerge from their marble confines, caught in various stages of Revelation, while the perfection of David looms over us in polished splendor.
The coarse and fine chisel marks in the prisoner Atlas are evidence of Michelangelo's process and echoes of his dialogue with the marble. According to Gombrich and the story of art, Michelangelo saw his role not as a creator but liberator. His envisioned forms were already present within the marble. His role was to chip away the excess, to reveal the figure lying dormant within the stone, to reveal what was always present but hidden from view.
The chisel isn't a tool of invention, but revelation, gradually exposing the ideal form imprisoned in the stone. Moving from the intentionally unfinished prisoners to the perfection of David, Michelangelo didn't just reveal an ideal form, but thoughtfully edited it for our benefit. The head and the hands of David are optically exaggerated, because the statue was originally supposed to be placed high up on the Cathedral of Florence as roofline.
The calculated adjustments demonstrate Michelangelo as a sophisticated editor of visual experience. His chiseled didn't just free figures from marble, he shaped how we would encounter and understand them, anticipating the relationship between the viewer and the sculpture across space. This insight from John Culkin explaining Marshall McLuhan's media theories captures something profound about our relationship with creative tools.
Like Churchill's earlier observation about architecture. We shape our buildings and afterwards our buildings shape us. It recognizes the reciprocal influence between creator and creation. Michelangelo's chisel revealed figures from Marshall marble, but the properties of the stone itself, its grain and character, also shaped his artistic choices. He didn't simply impose his will on passive material.
He engaged in a dialogue with it, adjusting proportions and forms to work with rather than against its nature. This reciprocity exists in all creative tools. We design them for specific purposes, but once created, they begin to influence how we think and what we consider possible. The best tools acknowledge this relationship. They extend our capabilities while respecting our intentions.
This reciprocity between tools and thinking takes on a fascinating new dimension with AI, a rich topic that deserves its own discussion on another day. Turning from art to the written word, we find the same reciprocal relationship between tools and thinking here in this second or third century papyrus of the gospel according to John, we see some of the earliest forms of what we might call punctuation.
These mid dots that are circled originated with Aristophanes in Byzantium, a second century Greek scholar and librarian in Alexandria. His system of low, mid, and high dots marked short, intermediate, and long breaths. These early punctuation marks weren't passive notations. They actively shaped how the text was performed and understood.
From breath marks to inked representations, punctuation underwent a profound transformation with the advent of movable type. What began as prompts for orators are now literally cast in lead. Ancient texts with minimal punctuation and absent word spaces demanded scholarly attention. Readers had to immerse themselves deeply in the material to invest that time and intellect to understand it.
Gutenberg's innovation changed this relationship. His standardized marks accelerated this long evolution towards making text more accessible. This isn't just about the mechanization of writing, but the process of designing for efficiency, of understanding, optimizing content for quicker consumption rather than prolonged contemplation to accommodate Gutenberg's aesthetic standards and the strict, fully justified columns of text.
These thin, upward angled parallel lines that I've highlighted signal a hyphenated word. And then each punctuation convention the semicolon. The colon has served as a scaffold, meaning signaling specific logical relationships between ideas, not just pauses of different links. What's fascinating here, in this dense, Black letter text, that the text itself doesn't just affect how punctuation looks, but also dictated how they would work effectively.
So instead of commas, which might get lost in the forest of these thick vertical strokes. Gutenberg employed mid dots again as a way to separate list items. The medium and the message become inseparable. The tools we use to determine not only how we express ideas, but what expressions we find possible. This principle of adapting to technological constraints extends to other editorial marks.
These horizontal lines. They're conceptually similar to the way that we use a contraction today. And while a contraction is primarily about linguistic efficiency, Gutenberg's emissions were typographic solutions. Strategic compressions again to enable the strict columns of justified text. Again, here's an interesting case where we have the shorthand substitution.
So the stroke 7 at the top there this is a tironian et from the first century BCE. And this is a predecessor to our modern ampersand. And this is where Gutenberg again, is trying to maintain the visual measure of the content by replacing a two letter word with a one letter substitution. Like Michelangelo manipulating David's proportions for the benefit of the observer.
Gutenberg crafted these editorial marks not merely to represent language, but to guide the reader's eye, shaping how they encounter and understand the text, from the sculptor's chisel to the editor's mark, whether revealing figures from marble or meaning from text. We work within that beautiful imprecision between representation and reality. We don't merely represent, but we guide understanding, shaping how ideas are encountered.
Which brings us to this essential question why do we do what we do. I'm guessing that you didn't become an editor primarily to wrestle with markup syntax and chasing down formatting inconsistencies. The work we really care about is helping ideas find their audience with clarity and impact. Which brings me to the core philosophy behind why I'm here today.
That tools should free us to focus on meaning and not trap us in TDM. This philosophy has guided typekit for more than two decades. Many of you might know us through our InDesign InDesign server based solutions, but consistently we've found ourselves helping organizations transform not just their workflows, but their entire approach to content and publishing.
This role has taken on a new urgency and represents a unique opportunity for typify. Our long partnership with anera has given us deep insights into how you use extyles and what you'll need in its replacement, and with an arrow's blessing, we're drawing on this experience to launch Orion, a purpose built drop in replacement that maintains the workflows you rely on while expanding what's possible.
So today I'm going to preview an early development build of Orion that sort of punctuates our progress towards a 1.0 launch by the end of this calendar year, well ahead of exile's August 2026 sunset date. Like the constellation it's named after, Orion is recognizable and familiar a drop in replacement. Our goal is that you can be working in textiles on Friday and continue with the same document in Orion on Monday.
So let's get over into. Microsoft Word. So this is a file that was graciously shared with us, shared with us by the World Health Organization to do some demos with and again, to start here. What were.
We should do the trick. Thank you. OK, we're back in Microsoft Word. So this is a file that we can get rid of that. There we go. This is a file that the World Health Organization shared with us for doing these demos. And what we want you to focus on in this is the similarities and how we are picking up key aspects of textiles so that you can use that muscle memory that you have from the past.
So starting off with our document, we want to make sure that we have a way to capture and enter and manage the metadata. And so the document information dialog here should look very familiar if you've used textiles x-files before, one of the things that we are doing in Orion is there's another sort of principle that I hear very strongly to is that I want things to be self-service.
I don't want you to be beholden to me, to make changes that you are able to make and edit and manage this system yourself. And so this entire dialogue is actually driven through a web interface that is available as part of Orion. And so we can go in and as a managing editor or administrator, you can define all the different pieces of metadata that you want to capture at any stage within your workflow.
You can set the different types of metadata, whether it's a text box or checkbox or combo box, what's required. You can even provide sample text or boilerplate content within here so that it's pre-filled, and then they can modify it. This could be instructions to your editor, or it could be actual data.
So back into my file. The next thing that should be extremely similar. If you've ever used before, is going through and doing some of this mechanical cleanup. Now there's internal debates about how much of this stuff should actually move into the smart replace versus staying in this dialogue itself. But this is almost a pixel for pixel recreation of the way that textiles displayed this information.
So you can clean up and remove excess spaces, normalize tables, and clean up the structure of the content before you even start looking at the content itself for semantic tagging and whatnot. Another thing that we want to focus on is when you're working within textiles, a portion of your time is going through and tagging semantically that content. And many of the customers we've talked to have spent a considerable amount of time in setting up and creating this paragraph styles panel with different palettes and how they organize these styles.
This is something that is defined as a preference within x styles, and we can literally import that entire preference. So there's no recreation. We just take it and transplant it over into Orion. And the same way we also offer a. A web based.
It's not even showing now there we go a web based interface where you can again, manage this stuff. And so instead of having to deploy customized builds of the software to each of your users, you can centrally manage it, set this up, and deploy it across your entire group and for each publication individually. So we can move through my content and we can see various things, and you can jump through like into the references.
If I want to tag this, go into my back tab. It's already tagged as a reference, so it's not going to look any different. But the whole process is here. Your keyboard shortcuts carry through. Everything is about transplanting the knowledge, this institutional knowledge you have into just a new tool auto replace, smart replace, auto redact different words for the same thing.
Again, this is something that we are building a solution for you so that all of the rules and customizations that you have already built and developed over for some people, 20 plus years. We can import those into Orion so that those same rules automatically apply. So we can go through and this triggers that process to go and check for various in this case, we're looking for some UK English spelling and convert it into us English.
One of the things that we want to also point out from the way that we're approaching this, is that when extyles, one of the common complaints we heard about extyles is the time that it takes to run certain tasks, especially if you have many tables in that content. Orion is built on a modern layer of working natively within the Office. Open office. Office open XML, the doc format.
So we don't have to actually take the word file, kind of flatten it into an ETF, send it to the omnimark engine to process it, bring it back, and then convert it from TTF back into Microsoft Word and do diff compares to figure out what actually changed. We can actually support native track changes, so you don't have to automatically accept them, and all of your content comes through and is restored.
You can see the track changes are present in this document where we've converting some of that spelling. Another thing that we are doing that's a little different in this sort of preview session, is that another aspect of extyles, it's very important is making sure that your reference section is accurate and correct. So I'm going to jump down to the references, and I'm going to be mean and I'm going to remove some people and change some letters and some numbers.
And so we're going to invalidate this. I know this is wrong. So within Orion I can run this through. And Orion has a long standing working relationship in supporting the fix solution from an arrow, which is the exact same reference checking, correction and linking tool from extyles, now available through a web interface. And so I can select.
Something like this. OK thank you for that. We'll correct. We'll check. We'll see. We'll run it through here. So in extyles you might have to choose multiple steps like I want to validate and go things through PubMed.
I want to go through Crossref now and so forth. All that is in a single step in Orion. We're going to check against all the things. We're going to insert the PubMed IDs. We're going to insert your. This is already locked in to the style that the World Health Organization uses. So I can either select an individual set of references and only do those two.
Or if I have nothing selected, you'll see that when I choose bibliographic references, it grabs all the references at once, sends them up to the web, and starts that processing. This is the one area within Orion that I have no control at this point that we are using this edifice to process this. This takes about a minute to go through and do these 10 references.
And we'll see the corrections come back in to fix my errors that I've introduced into that second reference. So fix when it checks your references it validates against PubMed, it validates Crossref, it validates against the Retraction Watch and also the predatory journal database. And so if it sees anything as a retracted article or anything that's considered predatory, it will add an annotation within your document to warn you.
As editor, you might need to look into this a little bit further. The other thing about when the references come back from fix right now, you'll notice that it's just plain text. It's black. The World Health Organization has could say, garish color scheme that they use for doing the visual check of how things are tagged.
We will also fully support that, and you'll see the colors come in just a second. If you have your own sort of color palette that you use with the textiles, those templates again can be ported directly over into Orion and we will leverage those same things for you. Here we go. Beautiful, slightly garish colors. And we can see that all of my errors that I introduced into this entry have been corrected and resolved automatically.
Scroll down a few, and we have a few warnings from artifacts to notify you that. Here are some references to check in to further. You'll also notice that every change that we make as a software tool we're going to annotate and say this is from typify Orion. So you can differentiate between changes that you make while track changes are on versus changes that we make on your behalf.
Last and most importantly is I want to be able to export this into XML. So if you already use typify what's happening right now when I choose this export this is creating a job within our typify server. And so I can see that job starting right now. You can see the little cog spinning over here. The interesting thing about this is that this typify server that's being leveraged by Orion has a workflow engine as part of it.
And so it's very easy for you to extend those workflows and do other things and incorporate other techniques or other tools into the workflow. So if you want to run the content, produce the Jets, and then apply a schematron validation to check for certain things. In that workflow, that's just a matter of editing the workflow and getting sick.
I think as people are getting in here, the Wi-Fi is getting a little slower. So if I can edit this workflow. And we can see there are five stages to this current workflow. If I want to add a new action to it, it's just a matter of choosing from the menu of available actions, and that will now be part of your export to jads workflow.
If you want to produce multiple files in a single step. It's just a matter of adding those steps in here. Lastly, the output from this export to jats. Come on, it's just Wi-Fi. It's got to download the file now. What it's supposed to be doing is downloading my output XML and then just popping that open inside Microsoft Word.
But I will open it in oxygen instead. And so here is the end to end preview of where we are currently with Orion as a drop in replacement for textiles. Now, there are several things that we haven't demoed. I've wanted to get this in front of people as early as possible, so they could start to see the progress that we're making.
And here we go. Here's what word did. It opened the word file. So we have the Jazz XML just popped open in here for review. If you don't have an XML authoring tool on your platform or here we go with the full XML rendered out of typefi. So that's I got five or six minutes left for any questions.
I'd be happy to answer them here. Or if you want to visit us in the booth, we're just inside door number five across the jetway. Otherwise, thank you very much for your attention. And yeah, I'll hang out if you have any questions. I think that's it. Oh, we have a question. Hi this is for tall people.
I'm Monica. I'm for with Jama Network. So I have one question because obviously right now we use textiles and the server that you demoed. It doesn't work like textiles. See in that it's single file in, single file out. And there's a queue process. Can it do multiprocessing of multiple files at the same time. Oh yes.
Great thank you. Yeah this is a high performance. You can run multiple parallel processes all simultaneously. If I look back at the job queue here that you can see that the auto replace was also run on the server. So each of these is just an action. So if you think about in an X files install like on a regular system like you have Microsoft Ward, and then you have an on device omnimark processing engine.
We've just instead of having it on device, it's in the cloud. And because it's in the cloud, it also supports parallel flows. You don't have any of the constraints that the way that x is operated and especially x styles, C was effectively single threaded and you had to have a massive queue to do that. No, none of that's an issue at all with Orion all massively parallel. Any other questions.
Well, thank you very much for your time. Enjoy the rest of the show. And again, we'll be at booth 104 just inside door number five across the jetway or breezeway.