I hope to resolve this soon.

Figure 1: Ushahidi upon loading.

Figure 2: Ushahidi upon refresh.

Update to the Update: It’s a feature, not a bug.
Upon further investigation, it appears there is no issue at all.  The map appears to default to showing only reports that have occurred in the last month.  Clicking on “All Catagories” shows all reports over the entire timespan covered by the instance.

Update Part the Third: That would be too easy.
I suspect it is a bug and not a feature. It doesn’t appear to be posting some very recent reports. I’ve come back around to the idea that there is some sort of caching issue at play.

The folks at Crowd Favorite are becoming my favorite WordPress devs very quickly; they’re making a lot of very high quality extensions for the platform which make my life easier and make this blog function well.

I installed Snow Leopard on this laptop last night. The installation went smoothly, and everything seems a little shinier and faster now. It also freed up about 24 gigs worth of space on this machine. From what I’ve heard, that’s about two to three times what most people are averaging, leading me to believe that I had a lot of unnecessary junk installed.

For all its technical advances, the feature that I’ve found the most compelling and useful is an older one with a number of improvements. The ability to annotate .pdf files in Apple’s Preview application is incredibly useful as a graduate student. Basic annotations, such as highlighting and notes, where around in previous versions, but this version adds self-contained annotation toolbar to the application, as well as rendering improvements with speed up what was already an incredibly snappy application. If the rumors turn out to be true, and Apple drops a touch screen tablet PC on sometime in the near future, features like this would make it an incredibly compelling tool.

I’ve been playing with an interesting web-based application called Ushahidi recently.  Ushahidi is an open-source application designed to crowdsource crisis information.  It was developed in the aftermath of the contested 2008 elections in Kenya to map reports of violence, and allow near real-time reporting of incidents.  It has since been used to monitor violence in the Democratic Republic of Congo, elections in Mexico, and by Al-Jazeera to report on Israel’s war on Gaza.

Ushahidi allows users to report incidents via e-mail, SMS, twitter, and a web form and then maps them to a Google map.  These reports can be categorized and administrators can control which are published and verify them. Users can also comment on submissions and note whether they are credible or not.  Giving it even more potential is its compatibility with FrontlineSMS, a tool that allows anyone with a computer and a cellphone to turn the machine into a hub for receiving text messages.  In parts of the developing world, where internet access is spotty and limited, but cellphones are ubiquitous, this can come in handy.

Beyond monitoring crises or situations, I can think of a further applications for software like this.  Locals could report illegal logging or poaching, you could more closely monitor conditions in refugee camps, and of course, you could use it to prevent situations from growing violent by creating an infrastructure by which police or elders to respond to tense situations and defuse them before they got out of hand.  That’s not to say it’s a panacea.  Ushahidi acknowledges the need for verification by providing tools for communicating the level of trust and verification of a report within the administration panel, but developing good policies about how to use this tool is critical.  Administrators need to have practices in place for verifying information, less the map be filled up with rumors and false information.  Without a plan in place to verify information and develop trusted sources, over reliance on this as a preventative tool could also expose police to ambush, or more mundanely, skew situational awareness and make decision-making more difficult or leading to the misallocation of resources.

I’ve set up an instance playing around with it as a tool for monitoring large events (such as the upcoming G-20 summit here in Pittsburgh.)  Not everything is configured quite right, so things like e-mail submission and alerts aren’t working yet. All that said, it’s a great concept and a tool with some potential, and although it’s only in beta form, it is easy to install and start running, and is already boasting a large number of features.  Give it a look.

I think the events in DC last week also illustrate this point, though they also illustrate the potential for inaccuracy, since Gen. Mukasey hadn’t actually passed away, as initially reported, but simply collapsed. However, those initial reports were corrected pretty rapidly, and prior to any reporting by the traditional media. I think the lesson to learn is that due to the “BREAKING” nature of news, inaccuracies make their way into these reports, but I that the correction mechanism here is built into the model. Due to the distributed nature of citizen journalism, inaccuracies can be rooted out and corrected quickly.

I’ve been following the events out of India via the excellent BreakingNewsOn and they’re quicker than CNN & the BBC and (though I haven’t conducted a study) very accurate and diligent about correcting themselves. Interestingly enough, they have just as much global reach as CNN, because their sources are distributed across the globe, providing information on an “as it happens” basis.

Time for the traditional media to hop on the clue train if they want to keep up. I think Peter Shankman’s Help A Reporter Out (HARO) is a step in the right direction. Another good step is Rick Sanchez on CNN, he twitters as @ricksanchezcnn and often uses it to solicit the opinions of viewers, and reads them on air. They’re mostly using these people for quotes, rather than for any actual content or reporting, but, this is a step in the right direction, and a step in the direction of using citizen journalists to drive reporting.

What really interests me is that we, the twittering public, knew what was going on before the medics had arrived, thanks to @bamber who was apparently in attendance. It’s the police scanner for the modern age, and with a lower bar of entry, everyone’s invited.

Over the past few months, I’ve been separately working for two people who are part of the University of Massachusetts-Amherst Department of Political Science, as a graduate assistant for one and a freelance web advisor for other. For both I’ve had to get my hands on the space UMass provides for personal, course, and conference websites, and in both cases I’ve found myself frustrated by a dearth of resources.

At UMass-Amherst every account (student, professor, course) gets 100 Mbytes of hosted space to do with as they please. However, this space is limited to static html pages; no databases, no php, and no apache redirect (which would be nice if you wanted to get around these limitations by hosting externally.) This likely makes administration marginally easier, but it makes the creation of modern, easily updatable webpages rather difficult.

Both of my clients need websites that can be updated on a regular basis, and neither of my clients are particularly inclined towards coding in html, nor do either of them have the time to learn, and in my opinion, they shouldn’t have to. A far better solution than those I have available to me at the moment would be to put up a drupal or a wordpress install for them, skin it to their specifications, and teach them the basics of creating pages and adding content. This would save both money and time and it would make their lives a lot easier if they didn’t have to worry about editing code every time they wanted to change a link, or messing about with .htaccess if they need to password protect something.

UMass-Amherst would simply make a lot of lives a lot easier, gain a reputation as being technology friendly, and allow faculty to have functional and modern sites if they added a few basic features to this hosting space.

This is what happens:

The code:

#container {
display: block;
width: 800px;
margin: auto;
background: #cccccc;
padding: 25px;
border-top: 2px solid #888777;
}

#left {
float: left;
width: 266px;
font-size: .85em;
/*padding: 12px 0px 12px 12px;*/
}

#center {
float: left;
width: 266px;
font-size: .85em;
/*padding: 12px 12px 12px 12px;*/
}

#right {
float: left;
width: 266px;
font-size: .85em;
/*padding: 12px 0px 12px 12px;*/
}

#footer {
background: #cccccc;
width: 800px;
display: block;
font-family: Arial, sans-serif;
font-size: 1em;
border-top: 2px solid #888777;
padding: 25px;
}

#clear {
clear: both;
display: block;
}

I want to simply make part of the page 3 columns wide with fixed width columns. I want the columns to appear to be of equal length, and share the same background color (#cccccc), and I want the page to work in IE.

Can anyone help?