Robbie MacKay

I’ve been working my way in to the crisis mapping and humanitarian ICT community for a while. Firstly with some minor contributions to Sahana Eden, then when earthquakes hit Christchurch I did what I could to help with eq.org.nz. Since then I’ve been helping with Standby Task Force tech team, and lurking in skype chats.

Last week I flew to Geneva for the 3rd International Conference of Crisis Mappers (ICCM). It was a 2 day event (free of charge!): day 1 is keynote speeches and 24 short 5 min presentations; day 2 is 4 timeslots of self organised discussion sessions, up to 9 sessions at once (barcamp style). All interspersed with lots of networking, talking in the hall ways, plans being hatched, and jet lagged attendees trying to stay awake.

The short summary is that it was a brilliant event, I felt over my head a bit – not being part of the humanitarian part of the community – but made a lot of connections and got a good feel for what is going on in the community as a whole, and where the biggest issues are right now.

The conference themes, as outlined in keynotes were: Validation, Security, Partnerships, Scaling and Mainstreaming (of crisis mapping). And as Sanjana highlighted during his key note: We are trying to save lives. That is what we must be driven by. That is what we are working for.

… and that proved to be well on target really. Especially in the self organised sessions, we kept coming back to security issues, other sessions worked on validation, and the other themes showed up everywhere.

I’ve got stacks of notes, some more informative than others. I’ve posted some to the main list already and I’ll try to write up some more and post them on here soon.

Back to reality for now..

For anyone who hasn’t played with an XO, the laptops made by the One Laptop Per Child project, you really should. They’re quite brilliant: simple, easy to use, and very unlike any laptop I’ve used before.

There are testing groups in both Auckland and Wellington, New Zealand, that spend their weekends messing with XO’s, installing software, testing it, tracking bugs and reporting the results. There’s a simple joy to it. Learning how to work these laptops. Trying out all the software. Showing young kids how to use them. Discovering the kids can use them better than you can (!).

Find your local group. Join in.

OLPC testing. The joy of

Off the top of my head here are some things that are hard and shouldn’t be:

• Code review
• Switching version control systems (not all projects will use the same tool)
• Collaborating closely but not breaking the staging site
• Version control for non-geeks (i.e. designer changing CSS, manager messing with the text or something in a template)
• Deployment (sometimes) – especially updates and especially PHP applications
• Trivial HTML rearranging in CMS output (Drupal in particular)

These are just pulled from my daily work, or other things I’ve done recently.

When I say these are hard I mean that that they take significant effort, and significant thinking. I say that they shouldn’t be hard, because these are all ‘admin’ processes, parts of the development that aren’t actually Development, but need to happen to allow development to go smoothly.

In some of these cases, I probably haven’t found the right tool yet. In some they are a system that’s ‘broken’, not fit for purpose. But in other cases, the most interesting ones, these represent a gap waiting to be filled. An opportunity for a product, or package, or simply a formula (just a blog post maybe?) which solves the problem and turns it into a near automatic process.

I’ve known about Silverstripe for a while but haven’t had much of a chance to try it. For past projects I’ve opted for JojoCMS instead, because I know one of the lead developers personally and the way Jojo works makes sense to me (little details like being able to set up my database structure by creating a DB table first and building the admin from that – it’s the other way round in Silverstripe)

However, I recently gained various motivations to try Silverstripe out.

So I started by grabbing the latest version (2.2.2) and a stash of modules of their download page (Gallery, Blog, Flickr and a few others).

The initial setup was very fast and easy, I was up and going in 1 minute.

However when I started looking at the blog module I hit some trouble – given I’m thinking of moving my blog into Silverstripe and using it to build a personal portfolio site – this wasn’t a good sign.

After a couple of reinstalls, lots of trail and error, and some much appreciated help from Hamish Campbell (all.or.nothing.net.nz), I managed to trace the bug to an issue with the BB Code parsing library being broken by the PEAR library which it was originally based on. It’s an issue thats been found by others before, but hasn’t been narrowed down since it depends on the environment and gives no error message.

Now - 2 days and a fair amount of hacking later – I’ve reduced my fix to a couple of lines and submitted a patch to Silverstripe.

And I haven’t even started writing my own code yet!

I’m looking at building an Google App Engine based μblogging tool – in other words a identi.ca / laconi.ca clone (which is a twitter clone).

I would implement the OpenMicroBlogging spec so it could talk to laconi.ca servers.

I’m looking at building it on top of the talk.org code (source is on github). Currently talk.org still needs a lot of work – it has no followers implementation to start with.

If anyone else is interested in joining in to develop this leave me a comment!
I’ll post updates as to how I got – for now I must get back to writing reports

HURIDOCS is contracting a software developer to improve its WinEvsys
database for documentation of human rights violations.

http://huridocs.org/tools/monitoring/openevsys

If anyone wants to send in a proposal and wants an extra developer – Ask me

So I came up with this a while ago but never got around to posting it – on looking at it now I think maybe it would be more appropriate to call them distributed applications not portable applications, however I’ll run with the name I started with.

I’ve been thinking about the trend lately to build things as web apps rather then desktop apps – or the create web apps which attempt to replace desktop apps.

I love alot of the new apps coming out – but I think people are missing part of the point.

The cool thing with the spead of the internet is the possibility of ‘portable’ applications not just web applications. What do I mean? Well I think there is an opportunity for applications that are portable (or distributed) – in that you can access your data anywhere from any computer. This also requires that applications are portable in that they run on multiple platforms. This class of ‘portable’ applications includes web applications but it also includes desktop applications which store data on the web.

Mozilla Firefox is heading this way with their Weave project – which lets you store your settings, bookmarks, etc in the cloud, Chandler (which I’ve blogged about) also fits this category.. Technologies such as Adobe Air show a lot of promise in making these kinds of desktop apps easy – especially desktop apps that are an extension of a web app.

Personally I’m a fan of both desktop applications which distribute your data – partly because I so often crash my web browser so having an application running seperately can be more useful. And also I find that applications which sit in the background  work better as desktop apps – such as time trackers, or twitter clients – they can sit in my windows status bar (or equivalent) rather then being a tab in the browser.

Hopefully we will see some great portable applications in the future
I’ve certainly got a few ideas I’m playing with. Let me know about any portable apps you’ve found or ideas for apps you’d love to see!

P.S. – I’d also welcome any comments on improving my writing, etc.

The folks over at the OSAF have finally released a 1.0 release candidate for Chandler Desktop (Check out their blog post). For those of you that don’t know about Chandler – likely all of you – here is the description from their vision:

Chandler is an open source Note-to-Self Organizer. It features calendaring, task and note management and consists of a desktop application, web application and a free sharing and back-up service called Chandler Hub.

I’ve been following the project for a while and considered joining in on the development at some point. It’s a pretty cool project – the main bit I love is that you can get to pretty much all your information from the Dashboard screen.

However for me I find its still not quite the right fit – I think the developers have built what they intended but not what I wanted they’ll get there eventually . My main complaint is that it doesn’t allow me to manage my email in the same interface. You can set up your email accounts – but chandler only displays email sent from Chandler or emails you copy in to special Chandler created IMAP folders.

I have yet to find a mail/calendar/task manager that can give me a work flow I really love – for now I’m settling for a combination of Gmail, Google Calendar, Thunderbird and Remember The Milk (aided by RTM for gmail). What I was hoping for from Chandler was a tool that could integrate all of these and enable the workflow where: email comes in -> create todo and/or calendar entry. Allowing me to find the original email from the todo events etc.

If someone wants to build this tool (or join me in doing so) let me know… however for now it appears I’ll still have to wait – maybe I can get close to what I want with plugins to Thunderbird and/or Firefox.

Note: Apologies – this came out as a bit of a rant even though that wasn’t intended. Oh well it is what it is.

Update: I think OSAF are aiming to build Chandler a tool that will do what I want, they’re just not there yet (See Mimi’s comment). I look forward to seeing future developments.

Update: Chandler 1.0rc2 released – http://blog.chandlerproject.org/2008/08/05/chandler-desktop-10-rc2-released/