User:Blythwood/Writing larger articles

Over the last few years, I've written several reasonably large articles on the topic of the history of printing completely from scratch, something I'd not done so much before. None are huge, but all took a lot of thinking to decide what to include and where to the point I was happy with them. They include:


 * Fat face
 * Vincent Figgins - didn't create this one, but totally rewrote it as the original content wasn't good.
 * Caslon Type Foundry
 * Wood type
 * Hendrik van den Keere (a bit smaller than the others)

Prose composition isn't something that comes naturally to me, so after seeing a question on the Reference Desk about how to structure big articles, I thought I'd set out what I've learned.

It's worth getting structure right
When you're writing an article, you've got two main lines of action: chronological (in order of what happened) and thematic (by topic). It's hard to get the balance right between them.

What came easiest to me was to split articles into general sections of time and within them approach the topic thematically, but you do have to let the content guide you a bit in deciding what those sections are going to be. I've found in these articles that writing a section titled "Influence" or "Legacy" was helpful to describe overall effect on society without being limited by chronological order.

Taking time away from articles can help with seeing what's important. With my wood type article I found that the first version I published just tailed off with no real ending or conclusion: I realised that adding an "after people stopped using it" section helped round things off and make it satisfying. A student editor managed to find additional sourcing which helped explain this from an angle I hadn't really considered.

It seems helpful to decide the "detail level" you're going to cover in the article: are we going to explain everything or are we just going to say "there were lots of companies in the industry" and link to sources for anyone who needs to know more? You can decide this based on what source materials are available, how much time you have to write, and to be honest how interesting the topic is.

Sourcing
Sfn sourcing is sooo helpful when you're citing many pages of the same book. It's really nice to be able to cite exactly where a specific piece of information came from so you can come back to it. Endnotes are also really helpful to fit in asides which are interesting, worth clarifying, or just plain funny but don't feel like they should get in the way of the main thread of the story. (This can be a good stopgap: often when I've had time to think about expanding an article I've found a way to move an endnote into the main body of the article once more sections are there.)

Sometimes there's something unusual. In the article on Vincent Figgins, there are really only a couple of primary sources with first-hand knowledge of his career. It made sense to put a list of these few primary sources at the front, otherwise I was shoe-horning in explanations of what they were in the most random places out of order. With only a few sources kicking off with a bullet point list of them felt structurally right.

I love reading big articles and strongly dislike having to click through several separate articles to understand a topic, but sometimes it makes sense to create or expand a separate article in order to clarify topics you're finding take too long to explain otherwise. With the Vincent Figgins article, I didn't really explain anything about the man he did his apprenticeship with, Joseph Jackson, a point the GA reviewer wisely picked up on. Writing a short article on him (it does need work) allowed me to move necessary information into a separate article for people who were interested. Writing an article can give you the idea for improving or writing others: my wood type article began when I thought "look, between the Vincent Figgins and Caslon Foundry articles and a few other things, I've got ingredients for the first draft of this article"...it did then sit around on my to do list for about six months before I decided to get it finished, though.

Good sourcing solves problems
Wikipedia doesn't allow original research, so you can't write things sources don't say, no matter how strongly you believe them to be true or important. So the more and better sources you read, the better. Reading contemporary reviews of publications can be incredibly helpful if they're by experts: it can throw up good quotes which cleanly explain topics and boil things down into a short, sweet quote to explain a topic to outsiders.

With my fat face types article, I suspected that some typefaces really weren't commonly used, as the authors of a book seemed to have to struggled to find examples: they only showed one, from a fairly obscure publication. But this wasn't clearly stated. It was really satisfying to find a review which backed this up and mentioned other research on the topic confirming this. Reading through the entire back archives of journals is like nothing else, you never know what's going to come up. Getting obscure sources the exposure they deserve is one of the most satisfying things about editing Wikipedia, I feel. Searching Twitter has also in the past thrown up unexpected things and directions for more research, even if what you find isn't itself a reliable source it can be a good direction finder. Luckily I've been able to visit some big libraries to do research but of course not everyone can do this (and of course Wikipedia's role should be to give everyone something as close to that as possible).

Minimum viable product is fine
It's hard to find time to do all this. What's the next best option? A quick, short article that isn't very long but cites the best sources clearly so people know what they are and where to find them.