Do you have a clear idea of what that would be? Without a clear vision, I suspect its one of those things that can never be perfect.
Quite clear. Not complete.
Philosophically:
- Enable content from much more diverse sources than academia currently does.
- Automate what is automateable.
- Pay attention to what engages and what disengages. Act accordingly.
I'm personally a fan of going as far as possible with something, and when I find I've reached a block then re-think it all and start again. Although it seems a burden to re-start everything, discarding it all is also quite liberating - it allows you to hone in on whatever you thought was most interesting. So, I continuously tear it all up and start again. You still have the old as a reference. So, I don't mind a simple text editor.
I'm with you on that: iterate your way to a workable solution (be agile, fail fast and all that). This is easier if there is no commercial goal of course.
But obviously you have your own ideas. I'd be interested to hear what you have in mind - probs the people around here would be interested too given they manage research too. It would be great to have the main themes, best research, etc, naturally 'bubble up' and have a 'contents page' functionality automatically available, with little effort or continuous sifting of data, nor a big commitment from the reader. It would be great to get the idea at a glance and then delve deeper if you are interested. Perhaps this is what you thinking of? Which sort of lends itself to a tree structure, or graph of associations... But that's quite a hard (slightly scary!) task. Amazing if you got it right, of course.
It's too early to claim it's finished. I'm iterating... There may be several levels of perfect and each level may change with technological advances and as the viewership's understanding develops. Here's what I've got so far. Some principles first, so we're sure we're looking at the same terrain.
I'm assuming a three-part publishing model:
- Content production: (research, writing, finding/preparing imagery and video, explanatory tools (eg, maps, diagrams))
- Editing: too complex to summarise here so summarised a bit below.
- Publishing: rendering the final content to a consumable form
Unpacking each of those parts:
1. Content production: content should be unique where possible but is therefore more time-consuming and work-heavy. Its challenges call for human creativity. The 'unique' requirement sometimes interferes with quality and scheduling. So, quality and scheduling challenges call on human flexibility for solutions.
2. Editing: This is a complex one but I suspect a lot of editing could be automated with, for example, online tools. Not all though. So editing can be only be semi-automated.
3. Publishing: must be automated. A reliable, computer-handled rendering of content-assemblies into consumable, enjoyable, shareable content.
These three parts are very different from each other (and within 'Content Production', text production and editing is very different to image/video production and editing). Taken together though, this model invites the standard approach for implementing automation:
- Use human creativity and flexibility in the unpredictable part: content production.
- Automate the predictable parts of editing; retain human intervention for the unpredictable parts.
- Automate all publishing.
Editing is a tricky one to discuss in any concise way. So let's try. Some parts of editing are easy-ish for computers: spelling, punctuation, possibly grammar and syntax. Translation is semi-automatable so why not also publish in the most widely-spoken languages?
But 'editing' also encompasses tone, 'look' and the use of juxtapositions to create implications, irony, humour, surprise and delight. The
gibbet post uses a number of juxtapositions; the
decomposing hand post uses two examples, one obvious, the other fractionally less obvious. These are much harder to automate. Maybe silicon AI can do it now. I don't know.
So that's the terrain.
This IHASFEMR thread - and its earlier posts in the Cannibalism thread - moved from text-heavy posts to mixed text- and video-heavy posts. One reason was that videos are more engaging than images, which are more engaging than text. Sure, history geeks like history text. But for anyone else, moving images and static images grab more attention quicker, and are more likely to be shared.
This holds true no matter how witty or erudite your text.
Any publishing model has to build this in if it is to compete for attention. In practice it means: turn content into small, preferably visually impact-full chunks that users choose to interact with or not.
I began to prototype a content-production and publishing system to test these ideas. It doesn't fully implement all the above, but it does implement what I could create with my strengths and with my weaknesses and in the time I've put into it so far.
Once it had enough content to be presentable, I showed it to a couple of younger people. I wanted to see what they cared about that was missing and what they didn't care about that was present.
Predictably, they had little interest in reams of text (though history students that are safely protected from academia might be interested in irreverant, long-form text). They enjoy short video clips like this one:
Operetta offended John Byng's German sensibilities. Source: In The Night Garden - Daisy's Big Loud Sing Song
They enjoy irreverance, preferably iconoclastic irreverance - delivered as video and images. This is useful feedback about what constitutes engaging content that viewers are happy to share among themselves (ie, make viral).
They are much happier with symbols and images for navigation than text. So they appreciate a clickable image carousel much more than my attempts at wryly humourous headlines. I hadn't anticipated that.
The implication is that image creation, selection and linking is way more important than mastery of English and headline-writing. Very contrary to traditional 'serious' publishing.
I've only just started asking for feedback but it's already clear that text should be used to hold pieces of content together rather than to dominate the content. Text as glue rather than as content.
Now, tools...
For the presentation side - the bit viewers see:
They remark with surprise about the speed of it (on their smartphones) over a data link (ie outside where there is no high bandwidth WiFi). How was speed achieved? By minimising javascript in rendering and instead using CSS wherever possible. This goes against the current default practice, which is unquestioned use of JS frameworks to cut effort, costs and risks.
For the content production side:
Diverse content production - eg from non-academics and non-historians and non-English-speakers - requires distributed content assembly and production. For the text part of my experiment, that means using git or similar. Writers don't have to learn much to use git to contribute. Git is great for any diverse, distributed group working on text-as-content and text-as-glue. But today 'diversity' includes a new participant: computers. Computers get git naturally because git was written for them. So the publishing model I'm looking at needs git because with git, computers can automate version control (and other technical aspects of production and publishing).
You must know where I am going next: markdown. Content production and editing in markdown reduces inconsistencies and enables computer mediation of text (both of text-as-content and text-as-glue). Computers can read markdown, which means markdown enables automated and semi-automated editing in a way that binary content formats such as .docx, etc, can't.
However, not everyone is happy with a text editor. The prototype I'm working on uses
Obsidian as the content editor. But I use vim with the content too. You can use whatever editor you want - as long as it can handle markdown. Obsidian and vim are interchangeable for me but for most people Obsidian would be more fun to use. Especially for reviewing written work and links.
I think I outlined to feralimal in a private message some months ago how
RoHT is auto-published by a 'git push'. The prototype I'm currently working with uses a variant of the same. It's a bit technical so I won't go into it here but feralimal can share it if it's useful. I will say that captions for images, videos, maps and diagrams are a publishing standard. Sometimes italicised. But markdown does not support captions. I coded caption support into my system. It took about two hours to write and test. So, although markdown doesn't support captions, the consistency of markdown made it easy to add caption support elsewhere in an automated publishing process.
I'm currently adding mapping (
Leaflet) and diagramming. Mapping checks the 'visual not textual' box and enables location-based user interactivity.
As
the discussion about Ewaranon shows, video is where information publishing is at. Graphics aren't my skillset so I clip videos with ffmpeg - as often demoed in this thread. But the reality is: animation and video creation skills are now a must-have.
If I could create animations, I would have said all the above in a short cartoon.
Returning the thread to IHASFEMR... and the association of crosses and churches with the Devil...
As in post-114599 above, we find another example of how reaching a certain number - this time 13 - makes the Devil appear at a church. From
Akenham, Suffolk:
Walking anticlockwise around the church of St Mary thirteen times is said to be a sure way to make the Devil appear.
1
1 The Paranormal Database
Why did 17th century Quakers oppose the (re)building of churches?
From
The Diary of Abraham De La Pryme, the Yorkshire Antiquary, published 1870, allegedly written around 1697, second footnote on p140:
Ye quakers are ... mighty backward to pay anything of dues to ye churches.
...awhile ago I was with ye pious and learned Mr. Tho. Place, Winterton, who told me, that when he began at first to build and repair that church, that there met him suddenly in the street a grave old long-bearded quaker, who accosted Mr. Place thus : ' Thou Place, (says he) I have a message to thee from God, who commanded me to tell thee that thou must desist in going out this work of the devil, ye repairing of ye steeple-house of this town ! ' And then ye quaker stamped at him, and denounced several woes against him if he did go on.
Why does so much East Anglian folklore support this Quaker's point of view? At
this Hidden East Anglia page, we see folklore frequently associates the Devil with churches, most often claiming the Devil had a lot of say in where churches were built.
As if the Devil were the lord of the manor.
