Adventures in XML

XML. What the heck is it, why is everyone talking about it and what could it possibly do for me? Well, I have a stack of ideas of things to do with it, only the most basic of knowledge about it - mostly theoretical - and absolutely no practical experience beyond writing an XML file and then staring at it wonder what all the fuss is about. So it's time to move beyond that level of ineptitude, so tonight I made a start. What's motivating this new push is work. We're still using FrameMaker, unstructured, and have so far done little to investigate structured Frame, even though I'm fairly sure we can simplify to doc types in one go if we just put some time into. So this is me putting time into it.

Enter, stage right, the XML1.1 Bible.

As I said, I have had a look at XML before so have some knowledge of coding, so I started off with a bit of a refresher course at basic structure and formatting with CSS. As I can't stand working with data that isn't meaningful to me I've already gone off-piste working with my own data. One of the applications I have in mind is register documentation, so I made a basic file and structure, which I haven't yet figure out how to copy here without losing all the tags, and this ugly style sheet.


CHIP {display: block; font-size: 24pt; font-weight: bold; text-align: center}
REVISION {font-size: 24pt; font-weight: bold; text-align: left}
YEAR {display: block; font-size: 12pt; text-align: center}
MONTH, DAY {font-size: 12pt}
INTERFACE {display: block; font-size: 20pt; font-weight: bold}
ADDRESS {font-size: 16pt}
DESCRIPTION {display: block; font-size: 14pt}
REGISTER_FIELD {display: table-row; font-size: 16pt}
FIELD_NAME, FIELD_BIT, FIELD_ADDRESS, FIELD_DEFINITION, FIELD_RW, FIELD_RESET {border-width: 1px; border-style: solid; display: table-cell; font-size: 14pt}

Strangely, it works.

So there I have it. My first, ugly but functioning register doc. I'm sure it'll be refined as I progress thorough the rest of the book, but it's a good start. Next stop XSLT.

Cleaning up interview recordings

Tonight's mission is to rid a recent interview recording of an unbelievable amount of noise. I was using the Olympus TP-7 Telephone Pickup Hands-Free Cable, which seemed to work okay for my last phone interview, but maybe I was using the other DECT handset or something, because this particular recording was dreadful. So, enter Audacity. This is a free, open source sound editing package that has capabilities way beyond my needs but providing the functionality I need to clean up the MP3 file.

For the kind of interviews I'm working on, for research rather than broadcast purposes, the most useful tool is the Noise Removal effect.

To use the Noise Removal feature:
1. Find a section of your recording that is just noise.
2. Go to Effects and then Noise Removal.
3. Click the Get Noise Profile button, then select a section of noise from your recording. If you select a really small section (less than a few seconds long) it will look like nothing has happened. That's fine -- Audacity will have grabbed the noise, so you can go onto the next step. If you grabbed a longer section, you may have to wait a second or two while it processes.
4. Click Ctrl+A in your recording to select the entire track.
5. If the Noise Removal window has closed, go to Effects and select Noise Removal to reopen it.
6. Click Preview. Audacity will play a short section at the start of your recording with the noise removed.
7. Make any adjustments to the slider (click Preview to hear the difference) until you are happy with the result, and then click the Remove Noise button to go ahead with the process
8. Noice removal will begin and will take a few minutes, depending on the length of your recording.
9. Save your edited file to MP3 (or other format) to use in your transcription software.

And that's it. Easy peasy -- with the only requirement being some patience! Next load it into your transcription software and away you go. What? You don't have any transcription software? Then my next post is for you!

For more information about Audacity's noise cancellation feature and other ways to clean up an audio recording, visit the Audacity wiki.

SyncBack - free and fabulous

So, you've got a desktop or some big laptop beast that you use at home for just about everything, but when you go into town, to a meeting, to the library - wherever - you take your altogether more portable netbook gizmo. You're suspicious of online storage (no thank you, Norton) but want to be able to work on both machines without driving yourself crazy. So how do you sync your precious files?

Well, the simple answer is SyncBack by 2BrightSparks. There are various flavours of the SyncBack application - from a free home app for home use through to a "pro" version of a more robust nature with a few more bells and whistles. I'm using the free one, but am so impressed with it I'm tempted to buy the upgrade.

Anyway, it's all so simple. With SyncBack installed on the machines you want to sync, you set about creating profiles, so I've got four in total: two on each machine, one syncing a named folder (could be My Documents) to my USB key and another back from the USB to the folder. Same set up on the other machine. Perfect-o. Put in the key, run the profile, remove the key.. and so on.

Now, a limitation is if you want to sync your whole My Documents folder, obviously you'll need a monster USB key, or whatever -- but I'm only interested in syncing a couple of project folders, so for me this is just the bees knees.

Word clouds

Word clouds, to me, are things of great beauty. There's something fascinating about them, and endless hours of fun to be had fiddling with color schemes, angles, typefaces, and - of course - the contents. Here are some of my favourite online apps for creating them:
http://www.wordle.net/ - allows visual formatting, but not much control over content.
http://tagcrowd.com/ - more control over content with this one (minimum frequency, group similar words, etc.) but much less satisfying visually.
http://www.tagxedo.com/ - I've not tried this one (content has to be online) but it looks like fun, forming tag clouds into images.
http://www.alphaworks.ibm.com/tech/wordcloud - This is a command line tool, useful for analysing data stored offline. Requires login to IBM developer site.

ScrapBook Plus

This is a free plugin for Firefox that enables you to download a web page and the contents of that page - so images, pdfs, etc. to a single folder on your hard disk. You can decide how many levels deep to go.

You can download it from here:
https://addons.mozilla.org/en-US/firefox/addon/8186/