Error in Characters Report

[frunjer]

The Character Report shows (CONT'D) extensions as separate character.  I'm assuming this is an error but possibly not (I'm new to screenwriting).  Is there a reason for this approach?  I'm assuming it results in a miscalculation of the character dialogue statistics.  Am I missing something or is there some value to knowing details on dialogue continuation?

[Anil Gulecha]

That would be a bug, yes. Needs to be fixed.

~Anil

[frunjer]

Just wondered.  I really like Trelby.  I was trying demos of Final Draft and Movie Outline.  Trelby works just as well as both.  Don't want to shell out 200 bucks on those if not necessary.  I'll make a big donation after I sell my first screenplay.  Thanks for a great tool.

[Osku Salerma]

So there's a few things to discuss here:

1. When the same dialogue section is broken over multiple pages, Trelby automatically adds a "XXX (cont'd)", for display only. This does not show up in any of the reports, so is correct.
2. If the writer either manually adds whatever variations after character names (CONT'D, VO, OS, ...), or importing from another file format brings such things over, then yes, they show up as different characters. https://github.com/oskusalerma/trelby/issues/42 talks a bit about this.

It is sometimes helpful to have voice-overs broken down into their own sections in reports, so if somebody wanted to change this, it would definitely have to be an option.

CONT'D doesn't really make sense to be a different character, but it's not trivial to automate combining that. People could spell that in different ways, or use non-english terms, or whatever. Realistically what would be most flexible would be a new script-specific config thing where it would show you a list of all characters in the script and you could group them together and say "treat these characters as the same character for report-generating purposes". Same logic as Script/Locations has (which has a horrible GUI).

On Thu, Jul 12, 2012 at 2:30 PM, frunjer <tdhil...@gmail.com> wrote:

The Character Report shows (CONT'D) extensions as separate character.  I'm assuming this is an error but possibly not (I'm new to screenwriting).  Is there a reason for this approach?  I'm assuming it results in a miscalculation of the character dialogue statistics.  Am I missing something or is there some value to knowing details on dialogue continuation?

--
Osku Salerma

[Thom Hill]

It really doesn’t make much difference to me and doesn’t impact how I plan to use the report.  I don’t manually add any continuation mark-up so it may have been a pagination issue or something.  Not a big deal and no need to fix anything if it “ain’t broke”.  Very nice software.  Actually the things I first thought were quirky and awkward, mainly text selection, ends up being a blessing in disguise.  Great job.

[Anil Gulecha]

On Fri, Jul 13, 2012 at 2:44 AM, Osku Salerma <osku.s...@gmail.com> wrote:
> So there's a few things to discuss here:
>
> When the same dialogue section is broken over multiple pages, Trelby
> automatically adds a "XXX (cont'd)", for display only. This does not show up
> in any of the reports, so is correct.
> If the writer either manually adds whatever variations after character names
> (CONT'D, VO, OS, ...), or importing from another file format brings such
> things over, then yes, they show up as different characters.
> https://github.com/oskusalerma/trelby/issues/42 talks a bit about this.
>
> It is sometimes helpful to have voice-overs broken down into their own
> sections in reports, so if somebody wanted to change this, it would
> definitely have to be an option.
>

I updated the pull request to make the current behavior optional (the
default being new behavior - treat them all as the character).

There's new function to get the name in util.py. It always removes any
extension starting with '(CON', but anything else can be ignored or
kept depending on the optional setting.

> CONT'D doesn't really make sense to be a different character, but it's not
> trivial to automate combining that. People could spell that in different
> ways, or use non-english terms, or whatever.

Sure.. but we're targetting english (for now), and my solution will
serve almost w=every screenplay out there. For the writer who chooses
to use KONT'D, well, he can choose to see the report with all
extensions removed (the default).

~Anil

[Osku Salerma]

We're not targeting english, we're targeting Latin1, because that's what PDF supports. I don't agree with "almost every screenplay" being in english either; france/germany/italy alone produce countless movies a year, let alone all the other languages latin1 supports.

I don't want to add anything english-language only.

--
Osku Salerma

[Anil Gulecha]

On Fri, Jul 13, 2012 at 7:54 PM, Osku Salerma <osku.s...@gmail.com> wrote:
> We're not targeting english, we're targeting Latin1, because that's what PDF
> supports. I don't agree with "almost every screenplay" being in english
> either; france/germany/italy alone produce countless movies a year, let
> alone all the other languages latin1 supports.
>
> I don't want to add anything english-language only.

Well , it serves our users with the addition, and as and when
additional requests come, we can always add to the list of CONT'D
equivalents. Right now, this patch makes Trelby better than it is.

We'll be adding an english-only implementation for now, but we're not
saying nothing more.. it is easily extensible.

~Anil

[Osku Salerma]

I don't want to keep adding to some arbitrary list. CONT'D is nothing special anyway, there's lots of other extensions people might want to combine as well. My outlined solution would cover a) all languages b) all usecases, so that's what we should do.

--
Osku Salerma

[Anil Gulecha]

On Fri, Jul 13, 2012 at 8:09 PM, Osku Salerma <osku.s...@gmail.com> wrote:
> I don't want to keep adding to some arbitrary list. CONT'D is nothing
> special anyway, there's lots of other extensions people might want to
> combine as well. My outlined solution would cover a) all languages b) all
> usecases, so that's what we should do.
>

It would be completely accurate but also extremely cumbersome: to make
the user select which entities combine to form a unique element for
reports. My solution will plugin, and let the computer, and not the
user, deal with this tedious task. A little inaccurately perhaps, but
that is fixed by us when a different extension becomes known to us.

Do you perhaps recall a single extension apart from CONT'D was used in
a way which would defeat my implementation? If none, perhaps that is
good enough to use?

[Anil Gulecha]

Also if this is not agreeable, how about I remove the CONT'D
intelligence, and simply add an option that allows any tensions to be
removed and reported? That way whenever any intelligence needs to be
be added in the future, we can wait for an implementation that we can
agree on.

Then it would be a simple feature addition agreeable to me as well.

~Anil

[Osku Salerma]

VO and OS are two extensions that immediately spring to mind. Some people, for some scripts, would want to combine those; some people, for some scripts, would not.

--
Osku Salerma

[Osku Salerma]

Why would it be "extremely cumbersome", exactly? If the GUI for combining characters is done well it should take a few seconds.

--
Osku Salerma

[Anil Gulecha]

On Fri, Jul 13, 2012 at 8:45 PM, Osku Salerma <osku.s...@gmail.com> wrote:
> VO and OS are two extensions that immediately spring to mind. Some people,
> for some scripts, would want to combine those; some people, for some
> scripts, would not.
>

Hmm.. these are already taken care of.. have you tried the patch?

Let's say the script has the following
JIM
JASON
JASON (CONT'D)
JASON (V.O.)
JASON (O.S.)

Then with my implementation, depedning on the boolean selection in
script settings, the reported characters would be

either:
JIM
JASON

or
JIM
JASON
JASON (V.O.)
JASON (O.S.)


I can remove the CONT'D intelligence, so then the options would be:
either
JIM
JASON

or
JIM
JASON
JASON (CONT'D)
JASON (V.O.)
JASON (O.S.)

---

~Anil

[Anil Gulecha]

On Fri, Jul 13, 2012 at 8:45 PM, Osku Salerma <osku.s...@gmail.com> wrote:
> Why would it be "extremely cumbersome", exactly? If the GUI for combining
> characters is done well it should take a few seconds.
>

'Done well' part would need the effort, and testing. To me, getting
out of the user's way even for those few seconds makes the experience
of using the program that much better. As you mentioned, something
like the locations dialog is not what we'd want.

~Anil

[Osku Salerma]

I don't want to keep adding half-baked features into Trelby because we can't be bothered to do things the right way. That way does not lead to success.

There is no shame in not doing something until you can do it right.

--
Osku Salerma

[Thom Hill]

Sorry Guys. Didn't mean for the issue to become that debated. Maybe look
at how Final Draft, Movie Outline, or Movie Magic address it. Also every
writer may have a different preference. I'm new to screen writing so you
may want to ask around. Maybe have a user preference to either treat the
extension as combined or separate. If separate then treat as already done
and if combined just strip off the extension. Thanks for the time and
effort, and mostly the desire to get it exactly right.


Thom Hill
281-840-7094
tdh...@aeqquus.com
http://www.linkedin.com/in/thomasdewaynehill

-----Original Message--
From: tre...@googlegroups.com [mailto:tre...@googlegroups.com] On Behalf Of
Anil Gulecha
Sent: Friday, July 13, 2012 9:47 AM
To: tre...@googlegroups.com
Subject: Re: Error in Characters Report

[dspu...@fostercaretech.com]

Love your comments in this thread. 

I've been using Trelby for years - love it (If I ever make it big I'm mentioning Trelby in my acceptance speech!)

Just getting into character reporting and it would be nice to be able to combine  CHARACTER + CHARACTER (V.O.)  + CHARACTER (O.S.) as one character -- esp. in # of lines.  Agreed that it would have to be an option.  Also, and this is more 'essoteric' what about where two characters speak CHARACTER1 & CHARACTER2  -- would be great if system could add these to single character totals

Maybe allow users to declare what character strings will be used for voice overs, off screen, and character combinations

;P