Archive for the 'Uncategorized' Category

Mar 18 2011

Getting going

Published by under Uncategorized

I am a broken record. First I love Macs, then I love Linux, then I love Macs, the I love Linux, and so on. Even I’m fed up of hearing it so I would like to set the record straight here:

I use Linux. I will, at some point in the future, have the situation where I think that the Mac is more interoperable with other people, and I will want to swap over. I will then find that it really doesn’t wortk any better with others and I will switch back. Given that I can see this all coming I’ve switched back for good this time, using Linux Mint Debian edition – and that’s the end of it. Some days I will find things I can’t do, but there will be a solution somewhere so I won’t panic and I will fix it and carry on my open source way.

Now, on with business.

I have finished my current project with macmillan Cancer Support, introducing Tridion as a new CMS for managing our print and digital content workflow. It all wnet well, all things considered, and was good fun, but I am ready to move on. I have been offered a role as Digital Publishing Director with Galore Park publishing, which starts full-time on April 4th. I am doing two days a week with them until then, but I am left with the luxury of a few spare days to catch up with things. I have some ends to tie up from my old business and I will finally be getting on with fixing the outstanding bugs and issues on my TasksApp application. I have some JQuery stuff to go in to pep up the usability, and also need to build in a proper config section so other users can manage their settings properly without having to access the database natively (archiving, customising the clients and contexts they use and so on). I also need to build a proper ‘install’ routine so that people can get and install the app on a local server too, which has not even been started yet. I am excited to get on with it and will keep this blog updated as I go.

No responses yet

Feb 09 2011

Linux Mint

Published by under Uncategorized

Early 2011 has been something of a landmark for me. One of my resolutions has been to stick with one distro for the year, and I am so far managing this well with Linux Mint. I have, up to this point, been a bit of a distro hopper, and sticking with the same one has been, for me, an achievement. I’ve taken the time to get the setup just right for me, and I’m happy that I have a solid system that works well for my home life and plays well with work networks and printers. Nice.

All that said, Linux Mint was not my first choice. I put quite a bit of thought into this, based on the latest testing I had done last year, and had a shortlist of possibles to try. My selection was, in rough order of appeal (and with reasons for dismissal):

  1. Arch Linux. I love this distro, but decided that the overhead of managing it was just too much. I keep a version running in a VM for experimenting with as its a great place to learn about the workings of Linux.
  2. Crunchbang. Another excellent distro, but one that I can’t quite get on board with as I find the standard Openbox window manager means that keeping things working well at work is too much time. I have been using Gnome for years and have a machine that doesn’t suffer from the extra resource needed to power it, so that’s my choice.
  3. Debian. I like this one a lot, and this would really be my top choice for a Gnome-based distro, but not enough works out-of-the-box for me at work. For instance, I’ve always had issues getting Apache up and running for local web development with Debian (not sure why), and have never really cracked accessing network shares or printers in the office (pretty sure its a Samba thing, but not sure what).
  4. Ubuntu. This one just works out of the box, but I’m not really a fan of where it taking Gnome, or how little of its interface development it seems to be pushing back upstream. I realise that any company in Canonical’s position is going to attract criticism for its actions, but I don’t think feel they deal with it well, and I’m not sure I want to be attached long-term to where they look to be taking the distro.

None of these seemed to have quite the right mix of features and setup, which led to a surprise entry at 5 – Linux Mint. Based on Ubuntu (so it all works out of the box) but without the overarching spectre of Canonical in direct control and with some cool UI tricks that make it a joy to use. I may make a small tweak and switch to the Debian-based version, but I’m not counting that as a distro-hop as they’re both Mint at heart.

One issue that is still thorny is the freedom of my current system. One of the plus points for Mint is that it has great support for multimedia codecs and plugins and I have enabled everything I want to use (I won’t say “need”) – Flash, mp3, VirtualBox, Spotify among others. I would rather keep a clean and idealogically free system, but I also need to be a pragmatist and realise that I work in a world where I need to proof-read Flash content, where every document I am sent (and expected to create) is in MSOffice format, and where all the best videos of cats playing keyboards just don’t render in HTML5. I try and be a good global citizen; I use and support free software where I can do so without pissing off the people who pay me.

No responses yet

Dec 23 2010

Music on console

Published by under Uncategorized

Running console apps appeals to me. I know we live in a world of vast computing power, we have great UI experts who create gorgeous user experiences around interacting with windows, but there’s something pure about the console. Its no-fuss, its resource-friendly, it feels like I’m using a tool to do a job, not ‘having an experience’.


I am using Mutt for mail, abook for contacts, identicurse for microblogging and emacs for all my text-based needs. All these things work as well as, or better, for me than gui-based equivalents. But I was still drawn to Rhythmbox for playing music. Until now.

Music on Console (Moc) is exactly what it says it is – a full-featured music player for the terminal. Everything is keyboard-driven, and is so much faster to work with once you get the keys sorted. Playlist support is superior to any other music app I’ve seen and navigating around and playing files is a breeze. The only thing I was missing were the global keyboard shortcuts that Rhythmbox has. But this was surprisingly easy to sort out.

Moc has a comprehensive list of arguments that you can pass to the program on the command line. Once the moc server is running and playing you can control everything with remote calls to the music server – this sounds much more involved than it is). Enter ‘mocp -G’ at a command prompt to toggle play/pause, ‘mocp -f’ to play the next track, and so on. But how to get these commands available as Gnome-wide shortcuts? Thanks to the wonder of the internet, the answer was not far away…

First, create three scripts to toggle play/pause, play next and play previous. Each script is a simple bash script file containing just the moc command for the action:

    #! /bin/bash
    mocp -G

for the toggle play/pause script, for example. Make sure that you make these scripts executable (run chmod +x on each of them to do this), and store them somewhere accessible by your $PATH.

Next, open up gconf-editor and select apps->metacity->keybinding_commands
There are 12 commands listed here that have empty values – I used command_1, command_2 and command_3 for mine. Click in the ‘value’ field so you can edit it and enter the path to the script for each function that you created above (e.g. command_1 value is /home/nigel/bin/pauseMoc for me). Next select the apps->metacity->globalkey_bindings node and enter the keyboard shortcut you want to use for each command in there. In my case I like to use ctrl-alt-win as a modifier and space, left and right for toggle, previous and next respectively. So to enter the correct shortcut for toggle, enter a value of ‘<Ctrl><Alt><Mod4>space’ as the value for run_command_1.

Close gconf-editor and you should be all set – works like a charm for me!

No responses yet

Oct 26 2010

Trapping information

Published by under Uncategorized

I have spent a lot of time thinking about how to capture notes and tasks. I have tried many systems and many approaches, and I have never really found one that works exactly as I need. The closest I have is my current setup, using my own task management app for tasks and plain text files for notes. Its still not exactly right though.

What I need is something that allows me to trap things, not capture them. Using a trap is about having the right things in place to make sure that, as things pass, you grab ‘em. Capture implies a chase-style activity, more about an activity that you have to do as an extra rather than something that flows out of what you are doing anyway.

I thought org-mode for Emacs would let me do this. I like the idea – minimally-structured text files that can capture text, todos, appointments easily, and present them to you when you need them. Trouble is, you have to invest a lot of time in learning emacs, and it feels like you’d have to move all your data in to get the benefit; not necessarily an issue, but tricky to get out of if it doesn’t work.

Something like DEVONthink or Tinderbox for the Mac would be nice, but again, its all locked in and you have to invest a fair bit to use it. Not to mention that I’m not really happy with the way the Mac platform is heading – another post perhaps?

What I want is:

- store text notes and tasks in Markdown format, with minimal extra syntax

- system-wide inbox to capture stuff seamlessly

- export to PDF, ODT or HTML in one click

- be able to pull out tasks by project, priority, date or context

I think the parts exist for this, and it would be nice to pull it all together into a Gedit-like interface. I’ve had mild success with adding pandoc scripts as external tools into Gedit, but maybe it would work to extend the Gedit interface to support a core set of functions natively? The joy of FLOSS is that I can grab the source-code and have a go. The downside is that I have no idea where to start…

One response so far

Sep 21 2010

Open Source and the avoidance of lock-in

Published by under Uncategorized

I’ve been playing with this post for a while. The motivation behind it is a conversation I had a while ago with a SharePoint developer, about why I thought Microsoft perhaps don’t have the user’s best interests at heart when they act. I want to try and clarify why I think this, and to try and work through why I made such an ineloquent argument for it at the time.

My argument

We were talking about lock-in and, in particular, why I thought Microsoft and Adobe were particularly evil in this regard. When pressed, the example I chose was that Microsoft changed file formats with Office 2007, bringing in the dreaded .docx and .xslx formats. This move, I suggested, was to make interoperability with previous versions hard, forcing people onto an upgrade path as they started to receive more of these new files from other people, either within the organisation or, more likely, from outside. The SharePoint guy countered this with the argument that they provided a tool to open these new files in old versions of Office – and they did this ON THE SAME DAY. This took me by surprise a bit — they had indeed done this — and we parted ways with me wondering if I’d got them wrong and him thinking that I was another flaky Free Software guy who just ‘didn’t get it’.

This has bugged me for far longer than it should have, but a long bike journey home today has given me some insight.

The McDonalds way

McDonalds are well known for their ruthless approach to world domination. You don’t meet many people who won’t join in a sneery rant against them, and most people think they are not to be taken at face value (at least in my experience). McDonalds have always offered a ‘healthy’ option – salad, carrot sticks, juice instead of coke, etc – but not as a genuine choice, as a way to appease the minority (they have a term for this that they use but I can;t remember it off the top of my head). The way it works is that the whole family want to go to McDonalds, except for the teenage daughter (again, their approach, not my example). The salad is there to give her something to eat while the rest of the family have the food that they really want to sell. They don’t care in the slightest if the daughter wouldn’t go there on her own, they just want to make sure she doesn’t stop everyone else going.

And this is what Microsoft did with their conversion tool. They don;t care if the single users upgrade or not, or even if they use cracked versions of their software. The people they want using it are the big corporations, the high-volume licenses. If they can offer something that stops the minority holding up the majority moving over, they win. Their intention is not that an office of 800 installs the converter on every machine, there intention is that the office of 800 upgrades all their users. If you have to go round and install the converter, why not bite the bullet and buy the upgrade now – you will anyway, so save two rounds of installation. The market for the converter is the home user and freelance who need a way to read the new file format without increasing the workload of the majority. By saying that the converter exists, Microsoft take away the one barrier to upgrade – you can use the new file format and get all the benefits while the minority, the whinging luddites who refuse to upgrade, have a tool to let them read your files, at no cost or effort to you.

The release of the converter on the day of release of Office 2007 was intended to drive people to upgrade, not as a tool to let them choose not to.

On the other side of the coin, Microsoft is using the exact same tactic to try and block OpenOffice. Despite the open document format being an open and visible standard, Word does not support it out of the box. Because Word is proprietary it is very hard to write the converter so OpenOffice cannot appease the minority, meaning the hurdle to move to OpenOffice (that other users can’t open the files) is in place and high enough to act as a barrier. If Microsoft were genuinely trying to be open (with their new format), and helpful (by releasing a converter) they would also support other people’s file formats wouldn’t they?

And what now?

It’s good to get that off my chest. It’s been bugging me for a while that something which I held as a fundamental truth – Microsoft changed formats as a deliberate sales tactic – was both not universally thought true, and seemingly so easy to dismiss. I think my reasoning above is very close to the truth; and part of me thinks that, if it is, someone at Microsoft deserves a pay-rise for spinning it into something that was seen as a Good Thing to have done. The other, larger, part of me thinks that I am doing the right thing opting for free software, and that there is a good reason to think that the areas I work in – charity and education – should be looking to move away from proprietary tools like office.

The difficulty is in how to get this message across, and how to convince people to switch to open tools while Microsoft keep the barrier to switching high. A report I read said that UK schools could save 80% of their IT budget because they use proprietary tools to do tasks that free tools could do equally as well. There are areas where free tools don’t match up, but think what development could be done if schools saved 80% of their budget on software – even 20% of that reinvested could make a huge difference to a lot of free software projects, perhaps eroding the other 20% over time.

Anyone want to set up a social enterprise to take this forward?

No responses yet

« Prev - Next »