Tag: tools

Revisiting Mind Mapping

I was reminded recently of the power of mind maps.

My kids are just starting to get writing homework where they need to collect ideas and make structured plans before they write their paragraph or page, and I realised that I had not talked to them about mind mapping. My older boy had already been introduced the idea and so wandered off to read his book, but the younger one found it fascinating.

I took as an example a writing project that I needed to get a structured jump on – presenting about NaNoWriMo at my day job for a Friday lunch in a few weeks.

This reminded me that mind maps are, in fact, awesome. I already knew that, of course, but sometimes you need to be reminded of these things – I’m not doing much at the moment either at the day job or in my writing which really needs a ground up mind map. I’m still mapping meeting notes sometimes, especially for training where I need to retain the information, but there is something wonderful about seeing a page fill with ideas.

After I’d gone through drawing that example mind map I showed my boy the Tony Buzan book I have (with its pictures of highly decorated – nay, illuminated – mind maps) and some examples of the mind maps I have drawn for projects in the past. I think the fact that I have so many of them made an impression.

And now, it’s back to writing the presentation I drew that example mind map for.

One Response

Back Up Your Data

Yesterday my laptop died.

It was my day job laptop, a MacBook I’d had since I started there two and a half years ago. It worked fine on the bus in the morning – I got a bit of outlining done, in fact – but when I plugged in at my desk it started to fritz out on me: random blocks and lines appearing on the screen before it reset.

And reset. And reset. Power cycling did not help.

The IT guy took it away and reset the P-RAM which got the computer to boot again, but about five or ten minutes into starting my working day the screen fuzzing came back and… well, that was about it.

The going theory is that it’s GPU-related since I had a few system crashes over the summer which had no specific trigger but left stack traces which pointed to something breaking in the GPU driver. I hadn’t seen that in a little while, but screen breakup is consistent with such a root cause. Not fixable internally, anyway.

Goodbye, upney.

And so I got to find out whether my backups worked.

I am delighted to say that they did: I have Time Machine setup on my Macs, and that last few minutes of clean runtime were enough to complete a backup set and sync Dropbox files up to the cloud, so my files were restored onto a replacement computer.

It was not a quick process, though: more than four hours from a local USB disc, with occasional in-process estimates that it would not be done for the entire day. I was still connected via my phone and I had things to work on in the meantime, but it was not the day that I had planned.

The moral, then: back up your data.

And be ready for some inconvenience when you need to restore.

Welcome, plaistow.

Leave a Comment

Keeping Time

Narrative fiction is essentially a linear medium: we read one word at a time and it is only by reading them quickly that they make sentences and layer image upon image to construct scenes and stories and the illusion of things happening at different times.

But even in this linear medium there are occasions when the events and knowledge of those events needs to be tracked: could the antagonist really get from the tube station to the top of the skyscraper in time to launch the drone? Who saw the monkey take the antique pen from Aunt Jemima’s desk? And when did Mr Kinnear learn that the monkey was actually a cunningly concealed alien spaceship?

Tracking this kind of information is enormously confusing and error-prone. Scrivener has custom metadata which can be applied to a scene and used to track exactly when that scene takes place, but that only works at the level of the scene: if there are multiple events which occur in the scene, or where there are disjoint sets of observers of the event, then metadata at the scene level might be too coarse.

Other options include constructing spreadsheets to track the event details, annotating outlines at a finer level of detail than the scene (see previous posts about detailed outlining, although where the time would be recorded is unclear – import the outline into a spreadsheet?), or the use of physical tools like index cards. The sequence diagram from UML is also of use here.

For NaNoWriMo this year, I trialled a tool called Aeon Timeline which supports detailed breakdown of events and their observers using a zoomable chart view to record and connect events.

My use of it was to help in figuring out the story events and how long things were really taking in the story: what the characters locations were at particular times, who could see certain occurrences, and so on.

The UI is easy follow, with simple keyboard shortcuts to add events and a very clear presentation of which actors know about which things.

What I am not clear on is whether this is actually any better than just laying out the events in a spreadsheet or sequence diagram. It is prettier, but it is also still a separate tool which you need to manually enter the data into. In those terms, annotating detailed outline text with event timing is likely to be more efficient – for me, at any rate, since my narrative is presented largely in the order it occurs.

Still, Aeon Timeline is a cool tool, and if your timelines are more involved than mine or if your storytelling is significantly more out-of-order than my stories then it might be worth your time investigating.

Leave a Comment

Evernote and I Part Ways

I’ve been an enthusiast for Evernote for some time. I started using it in April 2010, and I found it very useful: it was a way to share data across all of my devices, from phone to tablet to multiple computers. It was also a good audio note taker, although I don’t do that much any more.

How Things Went Wrong

Things started to break down only a few weeks ago when there was a database change in the tool which broke my ability to use Evernote to edit notes on my tablet, a first generation Kindle Fire. I wasn’t too concerned about it because in truth this was one of several applications which the Fire could not deal with (eg by not having a modern web browser), although having to fix the numbering on my long list of scene notes for my NaNovel was, shall we say, frustrating.

Still, I was already expecting to have to install a more stock Android build on the Fire so that I could use it more generally – I figured I would probably be back to using Evernote on there by the New Year.

Then the other day there was a version bump for Evernote on my day job laptop: a jump to version 6, with a new UI and another database change.

I don’t especially care for the UI*, as it happens, but that was not the thing that now has me throwing Evernote away: the database change nearly lost me some critical data.

Here’s the sequence of events:

  1. make changes on yig (personal laptop) to novel notes.
  2. Evernote on upney (day job laptop) is updated, including database change.
  3. make changes on upney to the same notes.
  4. return to yig for more writing, updating the novel notes again. Note at this point that there is no indication anywhere of accessing data from an incompatible database format.
  5. view novel notes on upney. They are gone. GONE. There is an empty note called “Untitled” where the novel notes used to be. Blood runs cold, stomach flips, etc.

Now, as it happens I could see these notes via the Evernote client on my phone so I knew they were not completely destroyed, but when I returned home I needed to take urgent action to make sure my notes were not lost – I disconnected the Internet by turning off the router so that yig’s Evernote instance would not sync to the cloud, then copied the novel notes into a text file** and also cloned the note within Evernote.

Then I reconnected the Internet and synced Evernote, and the novel notes document was destroyed just as I had feared it would be.

Another Orphan

There is no new version of Evernote for yig – upney is running Mac OS 10.9 (Mavericks) while yig is on 10.6 (Snow Leopard), and there is no version 6 build of Evernote for OS X 10.6.

Is that reasonable? Sure – I’m not going to say that Evernote or any company should be required to support an OS version that old. And it’s not as if OS upgrades are necessarily that onerous really, these days: Apple don’t even charge for them any more.

But they are disruptive. Both yig and upney have gone through OS upgrades, and it completely destroyed my dev environment in both cases. I lost hours of time while the OS upgrade was going on, but days of productivity while I rebuilt compilation systems and language environments.

In yig’s case I am not even convinced that an OS upgrade is a good idea at this point. It’s an old computer – I’ve had it almost six years now – and it’s more than a little creaky. The App Store claims it will run Yosemite (10.10) but will it really? What are the chances my computer gets completely screwed in the process?

Regardless of the viability of upgrading yig so I can run Evernote on it, I’m not doing it now. I’m in the middle of writing a novel: I can’t afford to have my primary writing tool out of commission for any amount of time, let alone a couple of days.

Why I Won’t Use Evernote Again

The real problem for me with Evernote is now one of trust: I no longer trust the tool, and I don’t feel like I can afford to use something which is only ever one bad database upgrade away from destroying my most important notes.

They could have handled these database upgrades better, but Evernote’s value to me was always in allowing me to share and edit notes across multiple devices: that value was eroded with the loss of the Kindle Fire as a client, but it has been entirely lost with the orphaning of yig. If they’re going to change the shared database structures to be incompatible with older versions of the tool, then the tool is no longer useful.

So goodbye Evernote. It’s been a good four and a half years, but it’s time to go back to text files on a cloud drive. At least then I control the format, and I get versioned backups from Time Machine.

[*] the flatness I can deal with, up to a point, but I really dislike the lack of visual separation between areas. It’s a usability disaster, as far as I can tell.

[**] and, in fairness, the Evernote text copy does the right thing here with numbered lists and checkboxes.

Leave a Comment

Tech Goes Obsolete

The nice thing about pen and paper is that it doesn’t crash, and it doesn’t become obsolete.

I’ve been using and programming computers for a long time and they play a central role in my life, but I am not particularly an early adopter of new technologies. Part of it is nervousness over data loss, but part of it is cost – it took me a long time to get a smart phone, because they just cost too much. New technology costs more.

I looked at tablets for a long time, thinking they might be useful but cringing at the price. I also wasn’t keen on the size of Apple’s iPad: I could see the benefit of the 10″ screen, but the device felt too heavy for me to hold for a long time. It was therefore enthused by Amazon’s Kindle Fire: a 7″ tablet for a reasonable price.

And I have used my Fire ever since I got it three years ago – it’s a good, if now basic, device: the battery lasts well, and although the touch screen is not very accurate* it was good enough to type on. I would have used it more if it had had a camera or Bluetooth, but it did most of what I wanted.

That is becoming less and less true.

The problems now are triggered by the fact that the device is, as these things go, old. The Fire runs Android (although with Amazon’s own tweaks) but the version is relatively ancient now, so new software doesn’t run on it. However, Amazon have never supported a lot of the applications that I actually want to run – the installed web browser doesn’t support a lot of things that modern web sites use, and their app store has no other web browsers for download. I side-loaded a slightly better browser, but that’s been overtaken too.

I’ve stopped using the Fire for writing now, because there are two workflows that just don’t flow at all:

  1. TiddlyWiki doesn’t run – part of the the problem is that there’s no Dropbox client in the Amazon app store, and then I’d have to sideload Firefox as well.
  2. new versions of Evernote won’t install any more – I use Evernote for a lot of notes in my writing, but I can’t edit them on the Fire any more because the database format has changed, and the newest versions of Evernote don’t run on the old Android version the Fire uses.

At least I can still use it for reading and annotating manuscripts.

I’ve been through this cycle recently with both my phone and my laptop: the phone was four years old and wasn’t running what I needed it to run any more because of version issues, and the laptop is a MacBook which was running an unsupported version of OS X. The phone was replaced, the laptop at least has a newer version of Mac OS on it so I can continue to use it for a while longer.

But the Fire? I have a couple of options, both of which would take me away from Amazon’s half empty app store: replace with a Google tablet that would get OS updates, or install a more recent build of Android on the Fire.

More research needed, but the Fire is about to go out.

[*] the first generation Fires used resistive touch sensors rather than the more accurate, but presumably more expensive, capacitive.

2 Responses


I’ve been using TiddlyWiki for my wiki construction needs for a few weeks now. This post is about making a world book with it.

First Steps

Review my getting started instructions from the sharing wiki.

The first thing to do is always to get a blank TiddlyWiki file. I usually grab one from the TiddlyWiki site to be sure of having the latest, but backing up a local copy and cloning it would be reasonable also.

Next, move it to Dropbox or other storage location and rename appropriately. I’m going to using as an example a world of plant-based superpowers, so I called the file SuperPlants.html.

Open this file up in Firefox.


Let’s give the file an identity that matches its name.

  1. click on the “control panel” link on the Getting Started tiddler, or click on the settings cog in the right hand panel. You should see a tiddler called “$:/ControlPanel”
  2. select the Info tab and Basics sub-tab.
  3. enter a title for the wiki. In my example, I called it “~SuperPlants” – the tilde character prefix prevents the wiki-link name from being rendered as a link.
  4. enter a sub-title. In my example, I wrote “a vegetative setting for superpowered adventures”
  5. in the “Default Tiddlers” box, type the names of pages you would like to be displayed when you first load or reload the page. In my example, I put in the following two lines:
    [[Plant World]]
    [[Plant Powers]]
    Note the double square brackets around these titles. Usual wiki syntax would omit the space call these pages “PlantWorld” and “PlantPowers”, but I like my pages to be real compound terms and those square brackets force linking to happen anyway.

You can also change the colour scheme, but the main thing is to click on the red save icon (the arrow pointing down into an open box) which will persist your changes.

Your Starting Pages

This world book will ultimately form a web, but it will start as a tree.

For each of the pages you wrote into the “Default Tiddlers” box, do the following:

  1. click on the “+” button
  2. type in the name of the page to be created. For example, Plant World. Note that you do not include the [[square brackets]] here
  3. type in the text for starting the page. Usually, this will include more links to other pages you want to write.
  4. click the tick to the right of the page title to save the page content.

The text for Plant World that I typed in was:

In the greenhouse world of the future, plants have mutated and grown at the expense of animal species - including humanity. Hubristic notions of the anthropocene have been washed away in a tide of chrophyll: the true extinction event was the scouring of the land and seas by new plants that thrived in the warm, carbon dioxide-rich atmosphere.

There are pockets of humanity still around:

* [[Mountain Men]] - hardy souls that climbed the highest peaks because there is still a treeline.
* [[Troglodytes]] - they took to the caves, because where there is no light plants cannot survive.
* [[Technocrats]] - sealed in domes and steel cities they fight to maintain the technology that defines humanity.
* [[Orbiteers]] - survivors of space colonies, living meagre lives in the airless wastes outside the warmed atmosphere.

Note that the different human survivor pockets are again enclosed in double square brackets.

Follow the Links, Make the World

Filling in the world book is as simple and as complicated as clicking on each link to create the page linked to, then typing out more text with links to new ideas.


  • don’t try to do it all at once. If you have a large page to write, you may be better off writing a bit at a time and adding more ideas later. You can always come back and edit the page later by clicking on the pencil icon.
  • use page naming conventions you will remember. For example, either use wiki-link syntax (no spaces, embedded capitals to mark word boundaries) or normal title format for page titles. Don’t forget to enclose normal format titles in double square brackets.
  • use the “Recent” tab on the right to see which pages you’ve written lately to remind yourself of the names.
  • use the search box to find relevant pages. This does search the tiddler text as well as the titles.
  • look at the help on wiki text. There’s a link to the help with each edit form under the “wiki text” llink.

How I Am Using This Right Now

The SuperPlants example was just that: an example, incompletely worked through, but I am using exactly this kind of world book structure in the following projects at the moment:

  • Songs of Atlantis – the dungeon crawl setting for my boys
  • Song – building a world book for Song to capture the current state of it so I can update the manuscript to match.

… but I’ll be using it for lots of things.

Leave a Comment

Oh, ‘Ello

There’s been a lot of noise about Ello in the last few days, triggered specifically by Facebook’s missteps on requiring real names but powered by a deep undercurrent of dissatisfaction with the social behemoth.

One Tool To Bind Them All

Facebook wants* to be not only the primary social network for the entire world but the primary method of accessing the Internet: if Amazon wants to be the Everything Store, Facebook wants to be the Everywhere Site. This motive makes a lot of sense, because if everyone uses Facebook then Facebook can deliver the attention of everyone to their customers, the advertisers**.

Where Facebook’s model breaks is that as they seek to deliver user attention to their customer’s advertising, they introduce barriers to communication between users. This reduces the value of Facebook to the users, making it more likely that they will look for other ways to keep in touch with their contacts.

What Facebook needed to do was to make themselves indispensable before they brought down the monetisation hammer, and they don’t quite seem to have managed that. They’re trying to balance the value their users get from the site against the inconvenience they need to inflict on them in order to deliver user attention, but even though so many people continue to use Facebook because of the social network it captures, that use is tainted by unhappiness that they feel forced to use something that doesn’t actually do what they want any more.

Poisoning the Well

What is Facebook doing wrong? With so many users, how could it fail?

Facebook is not the only manically agglomerative Internet corporation – Google is doing basically the same thing in terms of trying to deliver attention to the advertisers that actually pay the bills, but their business operates alongside the web rather than trying to replace it: Google provides services to its users that are still of value to those users. Many people are unhappy with Google in principle***, but Google’s services enable people to use the Internet more efficiently.

Facebook’s service is getting in the way.

The obvious comparison for me is with Twitter – Twitter is a firehose, if you choose it to be: an unmediated flood of information, which you filter by being selective in what you consume. Facebook mediates – it filters on your behalf, without your explicit input or approval – which means you don’t always see the things you would expect to see. Worse than that, it posts things into your timeline which you would never choose: sponsored links, invitations to play crummy games, likes and proxied posts. What you get in your stream is unpredictable, uncontrollable, and unusable.

The immediacy of posts is what Twitter does very well****, and it is exactly where Facebook fails.

Making Choices

Facebook is not the only social tool, though. There are a number of options for users, depending on what it is they want to achieve. All of them have their problems, but all of them have their place.

The following are personal observations of tools I have used extensively. Other social tools are available.

  • Facebook – good for keeping in touch with real-life friends. I have contacts there with friends from school I would not otherwise talk to, and with former colleagues and family members. But Facebook has really broken the front page stream: its curation makes it unusable, and the lack of any kind of content search means that finding a post you saw before (or even a reply you made to someone else’s post) is functionally impossible.
  • Twitter – good for keeping real-time tabs on subjects and people. Less personal than Facebook but with less mediation so you can make direct contact more easily than in FB. I’ve already alluded to Twitter’s most pressing problem: dealing with the flood of information. There is also a risk that Twitter will move to exactly the auto-curation and proxy post model that makes Facebook so disliked.
  • Google+ – part of the Google collective, so that is be a problem for some. G+ did some things exactly right out of the gate: the circles were brilliant and were only faintly echoed in Facebook, and the introduction of Twitter-esque hashtags helped a lot. But they missed a couple of tricks on search and filtering, and they could have recovered a lot of the ill will lost when Reader was killed by incorporating RSS feed ingest into their stream model. G+ isn’t the ghost town that many claim, but it’s about half as useful as it should actually be. Still, you can search for posts rather than just people and that counts for a lot.

There are a couple of collaborative tools I use in the day job which should really be measured as social network tools, but they are not really of general interest so I will omit them here.

Pinterest? Instagram? Snapchat? I don’t use those much, if at all, so I have few meaningful opinions on them.

Why Ello?

Ello came to public awareness in the wake of Facebook’s misguided “real names” policy, although I wanted to look at it because I thought the idea of a more focussed tool that was only about social interaction rather than posting more crappy memes and manky games was attractive. As of this writing it is spare to point of unusability, but the features it has are pretty solid: it’s alpha-level features with beta-level stability, as one acquaintance put it, which is almost exactly the other way around from most tools like this (yes, Jive, I am looking at you).

It is simple, and it really does mostly work. I like it.

Will it become the next Facebook? I doubt it – I expect Ello to become one of a constellation of social tools which interoperate rather than becoming a dominant player – but I also think Ello will be genuinely useful when Facebook is a pile of faintly glowing blue dust.

Why Facebook?

After all of that, I still use Facebook and I don’t especially expect to stop.


Because of the people.

That’s the killer feature that Facebook has: the people on the tool that aren’t anywhere else. Until and unless we have a federated social landscape, that will be Facebook’s advantage to lose.

Although they seem to be working pretty hard at doing that.

Find Me

My Facebook profile is friends-only because it’s a personal presence rather than a professional one.

But, I can be found on Twitter and Ello at @dunxiswriting.

Say ‘ello.

[*] Facebook is a corporation without actual desires or motives, but it has behaviour so as a shorthand here I am going to impute motives based on the corporation’s actions.

[**] you knew this, didn’t you? You maybe a user, but you are not the customer. If you are not paying for the use of a service, then you are the product.

[***] I don’t mention privacy in this post, but this is where I would.

[****] at least for now.

One Response

Sharing wiki, part 2

I wrote last week that I couldn’t share a TWiki instance between machines. In the comments, Brian Enigma pointed me at TiddlyWiki*.

This tool is about as close as I can get to what I want:

  • I can share the data via a cloud drive
  • I can setup a distinct file for each cluster of data
  • bonus: I can view the wiki contents on any device I can install Firefox and the cloud drive onto**

Setting Up

The setup instructions for TiddlyWiki are pretty simple:

  1. install Firefox
  2. install the TiddlyFox plugin
  3. load the wiki file into Firefox

If this is the first time you’ve used TiddlyWiki you can download a blank file from the site, or open up an existing file such as this one: world-book.

Once it’s loaded up, go to the Control Panel icon (the cog) and look around for things to change.

  • name
  • subtitle
  • colour scheme


The workflow for making a world book with TiddlyWiki is the same as for any wiki: make a root page and type in links to pages you want to create.

I will write more about using this particular tool for building up a world book another time. In the meantime, experiment with this file: world-book


This has been going pretty well for me so far after a few hours of use, but I’ve tripped over a few issues.

  • editing colour schemes is fiddly.
  • synchronisation of the saved file content is seamless from the computer, but pulling it down onto the phone. Forcing the page to reload is not as easy as I would like either.
  • Dropbox supports access to previous versions of the file, but the whole file is versioned: there’s no reverting an individual wiki entry
  • linking to existing pages requires remembering the names of those pages. There are tools in the UI which help (in fact, more than in TWiki…) but there’s still a lot of error potential

Still, this is a usable little tool with lighter requirements and simpler setup. It won’t do for a collaborative wiki, but for individual data where there’s only one author but multiple authoring locations it seems pretty likely.

At least I don’t feel like I need to take six months off writing to develop DataFrame any more.

[*] which is a name I love just because of the tiddlywinks reference.

[**] I can do this on my phone, but not on my Kindle Fire since there’s no Firefox available on there (at least not through the Amazon app store).

Leave a Comment

Wiki, the Unsharing

I was hoping that this would be a triumphant post where I explained how to share a wiki installation between multiple computers, but I couldn’t make it work.

So this post is a series of markers on what I tried and where it failed.

I was operating on a TWiki instance I had running on my day job Mac.

Making Wiki Work Again

In my instructions for installing TWiki on Mac OS X, I noted that they are for versions 10.6 and 10.7. My day job Mac is now running v10.9, and the wiki didn’t work any more.

First of all, Apple removed control of web sharing in v10.8. To activate it again, download this plugin.

Unzip it and move it somewhere sensible (I put it in /Applications/Utilities) then run it. You will be asked whether to install it for all users or just the logged in user – for a single-user computer it doesn’t really matter which.

Then click on the big slider switch, and you should be back in business.

Secondly, TWiki uses RCS for versioning of the pages, but Mac OS X v10.9 doesn’t include RCS any more.

Fortunately, it is available through package managers like MacPorts and Homebrew.

  1. install the package manager of your choice. I use Homebrew because it doesn’t require sudo.
  2. install RCS, eg brew install rcs
  3. configure TWiki to use the newly installed RCS rather than the one which is now missing.

The Easy Bit: Making the Wiki Run on a Shared Drive

  1. copy the TWiki data – copied the entire twiki directory to the cloud drive. Important here to preserve permissions.
  2. pointed the web server directory at the cloud drive – made a symbolic link to point the existing TWiki install location at the cloud drive.
  3. modify cloud drive permissions – need to loosen permissions on the cloud drive to allow the web server to read it.

With this approach, you don’t even need to restart Apache. I was browsing and editing wiki pages on the original computer in short order.

The Hard Bit: Making the Wiki Run on a Different Computer

Again, I was operating on an existing TWiki installation on another Mac.

  1. wait for the cloud drive to sync – need the current TWiki data!
  2. repeat the same trick with pointing the web server directory at the cloud drive – again, used a symbolic link
  3. loosen the cloud drive permissions – again, need to allow the web server to access the cloud drive

This got me to the point where I could browse the wiki, but modifying it would not work. I ended with these classes of failures:

  • incompatible data – I couldn’t log in with the wiki user from the original machine.
  • permissions issues – Dropbox could not sync all the files because the permissions were too close for it to read them. This particularly applied to working files (session data, error messages). I could modify those permissions so that Dropbox would actually finish its sync, but any new files are created with the tighter permissions.

So, overall, an interesting experiment but not a good result.

What Next?

I still like wikis as tools for worldbuilding, but I need to share the wiki across multiple computers with inconstant network access which means I need to find a different wiki implementation than TWiki, which makes me a little sad.

Or I may actually have to write DataFrame.

2 Responses


Last year I mentioned a tool called DataFrame which I wanted for my worldbuilding on a story, but then my attempts to try to build the tool derailed the making of the story.

What was DataFrame? Why would I want it, and what do I use instead?

The point of DataFrame was to have a frictionless tool for creating linked text fragments: I wanted to be writing something and to be able to create a connection to a related note without having to jump through any hoops – indeed, I really wanted some terms to be linked automatically, or at least to be able to apply a link after the fact using a search tool.

I also imagined having multiple windows to navigate through the network of, as I termed them, frames – open a link in a new window, or have a secondary window which would display the connected text as the cursor passed through a linking term.

All of this sounds very similar to the idea of a hypertext authoring tool. I was familiar with the idea of hypertext from Douglas Adams’ documentary film Hyperland, but the web was barely in existence and I didn’t really know what I had or how to build it.

Unfortunately the hypertext authoring tools I’ve used have been anything but frictionless. The closest thing to DataFrame that I have found is a wiki, but:

  1. the editing and the viewing modes are distinct
  2. switching from one to the other is painful
  3. linking to existing pages requires that you remember the name you want to link to*
  4. changing the name of a linked to page is painful**

So, I promote wikis not because they are perfect, because they are the easiest route to something like DataFrame.

I tried writing DataFrame a couple of times: once as a text-mode editor on my Acorn Archimedes (I got as far as having a basic text editor), and once as a plugin for the Eclipse IDE (I got as far as labelling some things on the screen as links, as well as making up a small presentation to explain the idea to colleagues at the day job at the time***).

What I should probably do is work on an emacs plugin. emacs can basically do anything, and the fact that it can render a web page as either raw HTML text or as clickable text should be an indicator that DataFrame should be something that could be supported. And it’s a Lisp environment, which would be a good thing for me to work on.

But then I would end up spending what little creative time I have on programming, which is not the worst thing in the world but won’t get the book written.

I really can’t let that happen again.

Are there any tools you hanker for to smooth your writing process?

[*] there are a lot of things wrong with the business social tool Jive, but its UI for linking to other pages is pretty good.

[**] this is an example of where maintaining the use-mention distinction would have been a good idea.

[***] the presentation was to be given on the same day I found out I was made redundant, so it was not in fact presented.

Leave a Comment