Proposal to stop using "==" for strikethrough

[Martin Budden]

As you may know, I have done quite a bit of work on improving the
formatters for version 2.1. One of the things Jeremy and I have talked
about is supporting different wiki formats, both by extending the
existing formatter and also by providing replacements. (These could be
made available as plug-ins, or alternatively as custom builds of
TiddlyWiki.)

Now, although TiddlyWiki format is different from many other Wikis, its
format rarely conflicts with those of other Wikis. (See
http://www.wikimatrix.org/ )

One notable exception is strikethrough "==". TiddlyWiki is the only
wiki of the 57 currently listed at wikimatrix that uses "==" for
strikethrough. And "stealing" it in this way means it is not readily
available for other uses (heading is the most common usage of "=...").

Strikethrough is a fairly rarely used format, so Jeremy and I agree
that the impact of changing strikethrough to stop using "==" is
probably be small, but we thought we should ask the community.

So what are your views?

Martin

[Daniel Baird]

+1 .. I'd be happy to have that change -- I can't remember ever using that markup.  Is there a preferred replacement?  Double hyphens?

;D

(free, effortless TiddlyWiki hosting)
http://danielbaird.com (TiddlyW;nks! :: Whiteboard Koala :: Blog :: Things That Suck)

[Luke Blanshard]

I use strike-through all the time to mark off things I've done from
lists I've made. I'm not against changing the notation, but having an
easy way to do it is essential.

Luke

[Eric Shulman]

> I use strike-through all the time to mark off things I've done from
> lists I've made. I'm not against changing the notation, but having an
> easy way to do it is essential.

To make the transition to a new syntax less painful, I suggest moving
the "==" formatter to a plugin. That way, folks such as yourself can
use the plugin to support the old syntax while gradually migrating the
strikethrough content to use the new syntax (whatever that will be...)

Of course, that runs the risk of perpetuating the syntax usage, since
it can be installed as a plugin. We should state as emphatically as
possible that "==" has been deprecated and that the plugin formatter is
just a transitional tool. The plugin could check the TW version
numbers and only work in version 2.1.x, so that when 2.2 is released,
the formatter stops being added to the parser. That way, people have
some time to fix their content, but the syntax (hopefully) won't
persist once the transitional period has expired.

-e
Eric Shulman
TiddlyTools / ELS Design Studios

[Martin Budden]

Eric,

I think this is a good plan. Is there a way to ensure that users who
upgrade from 2.0 to 2.1 get the plugin by default (rather than have to
install it manually)?

Martin

[Eric Shulman]

> upgrade from 2.0 to 2.1 get the plugin by default (rather than have to
> install it manually)?

Eventually, we WANT to break the "==" formatting, so that authors are
forced to re-write their content to the new strikethrough syntax.
However, if it is painless to keep using the old syntax (by providing
support via a "default plugin"), there is no impetus for authors to
update their content.

Manually installing a plugin requires the author to make a deliberate
choice between deferring a re-write of their content vs the short-term
effort needed to get the plugin and add it to their document. The
plugin would just provide a transitional bridge so that authors with a
lot of content to re-write can take the time they need to get it all
done.

We don't want them to use that bridge forever... this is why I propose
having a version check built into the plugin code. That way, the
transitional period can be automatically ended with the TW2.2 release.
Of course, someone could always recode the plugin to remove the version
check and continue to use the old strikethrough syntax in 2.2 and
beyond, but then their content would be considered non-standard, and
wouldn't migrate well to other people's documents, unless those
documents also have the *modified* bridge plugin installed.

The hope is that this strategy would encourage authors to migrate to
the new syntax sooner rather than later, since there is a small, but
non-trivial cost involved in deciding to use the bridge plugin for
longer abosolutely necessary.

In addition to the above 'social engineering' argument for manual
installation, 'default plugins' might be a bit of a technical problem:

Plugins, by definition, live in the storeArea tiddler data. However,
the standard upgrade process (as described on www.TiddlyWiki.com)
doesn't replace the tiddler data, just the core code. Of course, the
core does contain shadow tiddlers, stored as hard-coded data
structures. It would be possible to put a 'plugin' into a shadow, but
that would just move the code from one part of the core to another,
with no net change in functionality, so why bother...

thoughts?

[Jules]

As long as there is a replacement, this is fine.

I do use strikeout sometimes but I would be happy to change the markup.

--
Jules
www.knightnet.org.uk
http://knighjm.googlepages.com/knightnet-default-tw.html