I wouldn’t, although I find it easier to use endnotes than footnotes (ie, when you print them, they all appear at the end of the document rather than on the relevant page).
In the text, use an <abbr title=“Content of footnote” class=“note”>1</abbr> element for the footnote reference. You can number it manually, or if you want to be clever (and not worry about legacy browsers), you could probably get it to display the number with a counter. This will give the contents of the footnote as a tooltip on :hover. Then add a <dl> list at the end of the document defining each number with the appropriate footnote as the definition.
Other options would be to display the content of the footnotes in a frame/iframe at the bottom of the page or as a lightbox if you want it to look fancy.
Copy & paste the code below to embed this comment.
Stephen Down
What’s so hard about semantically marking up a film script? No, you won’t get the page break every 60 seconds, but I’d be surprised if that works out at a hard-and-fast rule anyway, and maybe that’s the one thing that has to give.
<h3>Reverse angle – over their shoulders</h3>
Slowly, without any fuss, and without a pattern of sorts, that would be pretty if the impact wasn’t so frightening… slowly, all the runway lights are going out.
McClane
Jesus…
<h3>Int. Virginia Church – same time</h3>
As Stuart’s tech throws some more switches –
and so on.
Use a counter on the <h3> element (I’ve assumed that a couple of higher levels will be needed but use whatever level is appropriate). Or use an <ol> if you’re worried about legacy browsers.
Use margin-left to indent the different classes of paragraph to the appropriate point.
Use text-transform:uppercase; on <h3> and <strong>.
There’s absolutely no reason why it needs to have a fixed-pitch font, if you’re lining the edges up with proper margins, but you can choose to use one if you prefer the feel of it.
Obviously with HTML5, you can use the <dialog> element to improve the markup even more.
What, in all that, (a) is not perfectly semantic, or (b) able to replicate the rendering used in the original script?
Copy & paste the code below to embed this comment.
Stephen Down
OK, so the Textile preview bears no relation to what is actually output…
I used normal angle brackets around h3, which it is left alone, strong, which it has converted to square brackets, and p, which it has stripped altogether.
The screenplay should have read:
[p class=“direction”]Slowly, without any fuss, and without a pattern of sorts, that would be pretty if the impact wasn’t so frightening”¦ slowly, all the runway lights are going out.[/p]
[p class=“speaker”]McClane[/p]
[p class=“speech”]Jesus”¦[/p]
In the author bio it has the phrase, “His ongoing missions”. This smacks of bringing religion to the ignorant natives to me. Here’s a piece of dogma about using “presentational HTML and inline styles” delivered without benefit of any technical justification, “These are, of course, outmoded development methods.” Just what is your objection to John August’s desire to have his RSS feeds look the way his readers, screenwriters, expect it to look? And what is it that makes them outmoded if the politically correct methods don’t work in RSS?
So I guess I missed the sermon on why everything on the Internet should be marked up systematicly. What I’ve never seen explained, in detail, is what the benefit is to this and why it is promoted with such zeal. Give me some examples of where this has been done and what benefit was derived.
The first step in justifying any technology is to clearly explain the goal of that technology. So let’s talk about the screenplay on the Internet. First, who is it there for and what are they doing with it? In the case of John August’s blog, they are there to be read by the people that read his blog and, I think, no other reason. They are there for education … to illustrate the point he’s making in his post.
Now I know that a script that is going to be produced does serve some additional purposes in the pre-production phase of film making. It is called the script breakdown. This is usually a manual step where the various film making departments comb through the script to find all the characters, props, locations, sounds and visuals that will be needed during production. There are script software systems which allow the author (or someone else) to “markup” the script to identify these elements and automate the breakdown. These are WYSIWYG tools and not HTML like at all.
So let’s talk about marking up a script in HTML:
“Would class names really suffice here—that is, H2 class=“slugline” versus H2 class=“charactername”? Really, the answer is no. Script headings and HTML headings are two different things.” They are? Who says? Why? Again, no technical justification … just dogma. You might as well say, “Well, everyone knows that’s just wrong.”
Each of these IS a heading that introduces and pertains to what follows until another heading is reached. That, to me, is the essence of what a heading is. Do class names not add to the systematic metadata about that entry? Here’s my shot at marking it up:
<DIV class=“manuscript”>
… stuff left out here …
<DIV class=“slugline”>
<H1>THECAB</H1>
<DIV class=“pagenumber”>32</div>
<DIV class=“continuetopofpage”>CONTINUE -</div>
<DIV class=“description”>But Barns doesn’t reply … just tries – and fails – to point out the window. Everybody turns.</DIV>
</DIV>
<DIV class=“slugline”>
<H1>REVERSEANGLE – OVERTHEIRSHOULDERS<H1>
<DIV class=“description”>Slowly without any fuss, and with a pattern of sorts that would pretty if the impact weren’t so frightening … slowly <SPAN class=“breakdownVisual”>ALLTHERUNWAYLIGHTS GO OUT.</SPAN>
<DIV class=“charactername”>MCCLANE
<DIV class=“dialog”>Jesus …</DIV>
</DIV>
… stuff left out
</DVI>
(I hope this comment preview is accurate.)
I see no reason this markup, which is systematic to me, can’t be styled using CSS and I know that I could write a parser to automate the breakdown of a script formatted this way.
So I guess I’ve missed your whole point, being one of the ignorant natives.
“but what you should feel is cheated” and don’t tell me how I should feel either. I don’t feel cheated when I buy those screenplay books. If you feel cheated then own that.
After writing this I guess my objection is to the tone of the article. I guess you may not like the tone of my reply either. Well the Internet is a big place.
I agree with most of what RobShaver said. While I think I understand Joe Clark’s point, I agree that functions like movie script writing and formatting need a “Movie Script Editor” program. If there is adequate demand, it would probably provide a web formatted output, maybe even like Joe suggests.
As far as “outmoded development methods”… there are so many responses to that statement that I can’t even begin to enumerate them here. Try using modern methods with HTML email. There is also the implication that people who still use “outmoded development methods” are Wrong. I know of people who built their sites with Front Page 4 and are still maintaining them with it. Would I recommend Front Page for anything? Hell, no… but I refuse to say these people are Wrong for doing something that works for them. Fortunately, the web browser manufacturers believe they need to support the existing web including those “outdated methods”.
Speaking of browsers, I think the browser writers are going to trump all of the supposed standards and keep trying to make software that works. HTML5? Yeh, we’ll support that… and HTML 1, 2, 3.2, 4, XHTML, javascript, ECMAscript, DOM, whatever it takes to make it work. ‘deprecated’? Not in this browser!
If W3C can’t make things that Improve the situation, maybe Adobe will. The web page of the future:
Copy & paste the code below to embed this comment.
Richard Fink
Thanks to Rob Shaver for going to the trouble of creating a demo for what I had in mind when I said:
“Using class names may be hacky but at a later date, at least the document can be machine-parsed and transformed into something better.”
There might be a smart way to get rel, rev, and title attributes in on the act, too. BTW – is this not the basic technique used in Microformats?
If Rob’s example were to be codified – spec’d to exactly when and where these combinations of tags and classnames are to be applied, it certainly would provide a “schema” of sorts.
It’s one way to approach the problem, surely.
Surely we could accomplish any of this with a simple <pre> tag? Thus giving the author pretty much the same freedom of the typewriter. A little clever javascript to deal with interface differences like tabbing (or any other typerwriter oddities) shouldn’t be too hard either. The argument for semantics is voided by the counter-argument that the printed version is automatically semantic by reader interpretation – the same would apply to pre-formatted text… My pencil doesn’t need to know whether it’s writing a paragraph or title, and nor would an online screenplay. It’s fair to say that typical web semantics don’t apply, and that the screenplay format could very easily be replicated.
Also the notion that an HTML document is automatically “Pageless” is daft, if you want to represent “1 minute” intervals what’s wrong with denoting every 50th line or so?
PS – apologies for my rudeness, I think that your technical argument is overshadowed by your obvious sentimentality for the typewritten form. If you’d presented your article with more romanticism and less “fact” I would have enjoyed and appreciated the article as I imagine you meant!
“Michael Newton”:http://www.alistapart.com/comments/unwebbable//#9 if the document is XML, it has to be usable by other applications, doesn’t it?
No it will not. Any XML application is opaque to any other XML application unless you have prior knowledge. The claim that an XML processor somehow magically knows what an XML application is about is a myth based on “mismarketing”:http://my.opera.com/jax/blog/html5-xml-stealth XML, and probably part of the reason for the perceived backlash against XML. XML solves problems, just not the problems it often is claimed to solve.
Back to the article, there is no reason why the internal format shouldn’t be a task-specific XML format (or any other format for that matter), and XML has the advantage that it can be transformed into HTML fairly easily. However I don’t think the particular example of film scripts was that well-chosen, as they can be encoded in HTML with no loss of information. The example in the comments with musical annotations might be a better one, as there is no adequate support for that in HTML.
Copy & paste the code below to embed this comment.
louBurnard
I’m surprised to see no reference to the Text Encoding Initiative (http://www.tei-c.org), which has been saying more or less the same thing as this article since the mid 90s. And which is now more or less the de facto xml vocabulary of choice for marking up the meaning structure of texts rather than their accidental appearance.
42 Reader Comments
Back to the ArticleStephen Down
I wouldn’t, although I find it easier to use endnotes than footnotes (ie, when you print them, they all appear at the end of the document rather than on the relevant page).
In the text, use an <abbr title=“Content of footnote” class=“note”>1</abbr> element for the footnote reference. You can number it manually, or if you want to be clever (and not worry about legacy browsers), you could probably get it to display the number with a counter. This will give the contents of the footnote as a tooltip on :hover. Then add a <dl> list at the end of the document defining each number with the appropriate footnote as the definition.
Other options would be to display the content of the footnotes in a frame/iframe at the bottom of the page or as a lightbox if you want it to look fancy.
Stephen Down
What’s so hard about semantically marking up a film script? No, you won’t get the page break every 60 seconds, but I’d be surprised if that works out at a hard-and-fast rule anyway, and maybe that’s the one thing that has to give.
<h3>Reverse angle – over their shoulders</h3>
Slowly, without any fuss, and without a pattern of sorts, that would be pretty if the impact wasn’t so frightening… slowly, all the runway lights are going out.
McClane
Jesus…
<h3>Int. Virginia Church – same time</h3>
As Stuart’s tech throws some more switches –
and so on.
Use a counter on the <h3> element (I’ve assumed that a couple of higher levels will be needed but use whatever level is appropriate). Or use an <ol> if you’re worried about legacy browsers.
Use margin-left to indent the different classes of paragraph to the appropriate point.
Use text-transform:uppercase; on <h3> and <strong>.
There’s absolutely no reason why it needs to have a fixed-pitch font, if you’re lining the edges up with proper margins, but you can choose to use one if you prefer the feel of it.
Obviously with HTML5, you can use the <dialog> element to improve the markup even more.
What, in all that, (a) is not perfectly semantic, or (b) able to replicate the rendering used in the original script?
Stephen Down
OK, so the Textile preview bears no relation to what is actually output…
I used normal angle brackets around h3, which it is left alone, strong, which it has converted to square brackets, and p, which it has stripped altogether.
The screenplay should have read:
[p class=“direction”]Slowly, without any fuss, and without a pattern of sorts, that would be pretty if the impact wasn’t so frightening”¦ slowly, all the runway lights are going out.[/p]
[p class=“speaker”]McClane[/p]
[p class=“speech”]Jesus”¦[/p]
RobShaver
In the author bio it has the phrase, “His ongoing missions”. This smacks of bringing religion to the ignorant natives to me. Here’s a piece of dogma about using “presentational HTML and inline styles” delivered without benefit of any technical justification, “These are, of course, outmoded development methods.” Just what is your objection to John August’s desire to have his RSS feeds look the way his readers, screenwriters, expect it to look? And what is it that makes them outmoded if the politically correct methods don’t work in RSS?
So I guess I missed the sermon on why everything on the Internet should be marked up systematicly. What I’ve never seen explained, in detail, is what the benefit is to this and why it is promoted with such zeal. Give me some examples of where this has been done and what benefit was derived.
The first step in justifying any technology is to clearly explain the goal of that technology. So let’s talk about the screenplay on the Internet. First, who is it there for and what are they doing with it? In the case of John August’s blog, they are there to be read by the people that read his blog and, I think, no other reason. They are there for education … to illustrate the point he’s making in his post.
Now I know that a script that is going to be produced does serve some additional purposes in the pre-production phase of film making. It is called the script breakdown. This is usually a manual step where the various film making departments comb through the script to find all the characters, props, locations, sounds and visuals that will be needed during production. There are script software systems which allow the author (or someone else) to “markup” the script to identify these elements and automate the breakdown. These are WYSIWYG tools and not HTML like at all.
So let’s talk about marking up a script in HTML:
“Would class names really suffice here—that is, H2 class=“slugline” versus H2 class=“charactername”? Really, the answer is no. Script headings and HTML headings are two different things.” They are? Who says? Why? Again, no technical justification … just dogma. You might as well say, “Well, everyone knows that’s just wrong.”
Each of these IS a heading that introduces and pertains to what follows until another heading is reached. That, to me, is the essence of what a heading is. Do class names not add to the systematic metadata about that entry? Here’s my shot at marking it up:
<DIV class=“manuscript”>
… stuff left out here …
<DIV class=“slugline”>
<H1>THE CAB</H1>
<DIV class=“pagenumber”>32</div>
<DIV class=“continuetopofpage”>CONTINUE -</div>
<DIV class=“description”>But Barns doesn’t reply … just tries – and fails – to point out the window. Everybody turns.</DIV>
</DIV>
<DIV class=“slugline”>
<H1>REVERSE ANGLE – OVER THEIR SHOULDERS<H1>
<DIV class=“description”>Slowly without any fuss, and with a pattern of sorts that would pretty if the impact weren’t so frightening … slowly <SPAN class=“breakdownVisual”>ALL THE RUNWAY LIGHTS GO OUT.</SPAN>
<DIV class=“charactername”>MCCLANE
<DIV class=“dialog”>Jesus …</DIV>
</DIV>
… stuff left out
</DVI>
(I hope this comment preview is accurate.)
I see no reason this markup, which is systematic to me, can’t be styled using CSS and I know that I could write a parser to automate the breakdown of a script formatted this way.
So I guess I’ve missed your whole point, being one of the ignorant natives.
“but what you should feel is cheated” and don’t tell me how I should feel either. I don’t feel cheated when I buy those screenplay books. If you feel cheated then own that.
After writing this I guess my objection is to the tone of the article. I guess you may not like the tone of my reply either. Well the Internet is a big place.
Peace,
Rob:-]
Dave Baldwin
I agree with most of what RobShaver said. While I think I understand Joe Clark’s point, I agree that functions like movie script writing and formatting need a “Movie Script Editor” program. If there is adequate demand, it would probably provide a web formatted output, maybe even like Joe suggests.
As far as “outmoded development methods”… there are so many responses to that statement that I can’t even begin to enumerate them here. Try using modern methods with HTML email. There is also the implication that people who still use “outmoded development methods” are Wrong. I know of people who built their sites with Front Page 4 and are still maintaining them with it. Would I recommend Front Page for anything? Hell, no… but I refuse to say these people are Wrong for doing something that works for them. Fortunately, the web browser manufacturers believe they need to support the existing web including those “outdated methods”.
Speaking of browsers, I think the browser writers are going to trump all of the supposed standards and keep trying to make software that works. HTML5? Yeh, we’ll support that… and HTML 1, 2, 3.2, 4, XHTML, javascript, ECMAscript, DOM, whatever it takes to make it work. ‘deprecated’? Not in this browser!
If W3C can’t make things that Improve the situation, maybe Adobe will. The web page of the future:
<html>
<head>
<title>Untitled</title>
</head>
<body>
<runFlash src=“MyFlashPage”>
</body>
</html>
Richard Fink
Thanks to Rob Shaver for going to the trouble of creating a demo for what I had in mind when I said:
“Using class names may be hacky but at a later date, at least the document can be machine-parsed and transformed into something better.”
There might be a smart way to get rel, rev, and title attributes in on the act, too.
BTW – is this not the basic technique used in Microformats?
If Rob’s example were to be codified – spec’d to exactly when and where these combinations of tags and classnames are to be applied, it certainly would provide a “schema” of sorts.
It’s one way to approach the problem, surely.
David King
Surely we could accomplish any of this with a simple <pre> tag? Thus giving the author pretty much the same freedom of the typewriter. A little clever javascript to deal with interface differences like tabbing (or any other typerwriter oddities) shouldn’t be too hard either. The argument for semantics is voided by the counter-argument that the printed version is automatically semantic by reader interpretation – the same would apply to pre-formatted text… My pencil doesn’t need to know whether it’s writing a paragraph or title, and nor would an online screenplay. It’s fair to say that typical web semantics don’t apply, and that the screenplay format could very easily be replicated.
Also the notion that an HTML document is automatically “Pageless” is daft, if you want to represent “1 minute” intervals what’s wrong with denoting every 50th line or so?
PS – apologies for my rudeness, I think that your technical argument is overshadowed by your obvious sentimentality for the typewritten form. If you’d presented your article with more romanticism and less “fact” I would have enjoyed and appreciated the article as I imagine you meant!
David King
That should have read < PRE >, I would’ve given an example if it wasn’t for the terror that is Textile formatting!!
Joe Clark
Rob Shaver, for “missions” read “goals.” At least I have some.
jax
No it will not. Any XML application is opaque to any other XML application unless you have prior knowledge. The claim that an XML processor somehow magically knows what an XML application is about is a myth based on “mismarketing”:http://my.opera.com/jax/blog/html5-xml-stealth XML, and probably part of the reason for the perceived backlash against XML. XML solves problems, just not the problems it often is claimed to solve.
Back to the article, there is no reason why the internal format shouldn’t be a task-specific XML format (or any other format for that matter), and XML has the advantage that it can be transformed into HTML fairly easily. However I don’t think the particular example of film scripts was that well-chosen, as they can be encoded in HTML with no loss of information. The example in the comments with musical annotations might be a better one, as there is no adequate support for that in HTML.
Roguebfl
“Another question would be why go to the web with the scripts?”
@mattrossidesigns (Post #5)
To share them with community theatre and school drama groups, is my first thought.
louBurnard
I’m surprised to see no reference to the Text Encoding Initiative (http://www.tei-c.org), which has been saying more or less the same thing as this article since the mid 90s. And which is now more or less the de facto xml vocabulary of choice for marking up the meaning structure of texts rather than their accidental appearance.