10:53 <@tuo2>  ... so when I need a webserver, I just
10:53 <@tuo2> $ cat /dev/urandom | perl
10:53 <@tuo2> It usually works in 3 tries or less.

It turns out that publication bias was rampant. Pfizer and Lundbeck, the two companies running the studies, didn’t publish a lot of their data, especially the data showing no effect and unfortunate side effects

via Guest Blog: The antidepressant reboxetine: A headdesk moment in science.

If something as simple as selectively reporting results gets drugs past FDA screening, something is horribly broken in pharmacology. I know the patent cliff is putting a lot of pressure on the larger pharma companies, but this kind of outright lying by omission is surely not legal.

There’s an assumption by many that cloud computing will be the savior of the IT industry. Beset on all sides - by shrinking budgets and increasing demands - IT feels pressured, and the cloud looks like a miraculous escape [*]. By eliminating CapEx and the cost of massive integration projects for proprietary technologies, cloud solutions tempt us with the promise of simple, elastic, transparent OpEx models.

And it’s true. The cloud does offer a radically simpler version of computing. Taking a page from Apple’s book, these services target the median use case, tear out unused complexity and configurability, and radically improve the experience of what remains. Suddenly, IT no longer has to endure the long lecture about “why can’t we do x”; instead simply blaming the cloud provider.

But this leaves many, particularly those in governance roles, in a difficult position. What if that uncommon, underused functionality that was torn out in the name of containing costs is essential in a given industry? Or worse, demanded by a regulator over whom you have no bargaining power.

In this case, one is torn between two difficult positions: either wait on the sidelines with an increasingly uncompetitive cost structure and ageing technology, or risk a regulators wrath.

[*] Worse, SaaS solutions provide the business with an opportunity to sideline the IT department entirely, by self-sourcing both critical and emerging applications.

Anecdotally, non-users struggle to understand why they might adopt Twitter and the iPad, but medium-term users seem to derive immense value from them. I suspect this is because they are platforms - providing a framework for other parties to deliver value upon - rather than being applications. Perhaps applications, with narrow use-cases, find quick adoption because of their obvious value propositions, while platforms need a while for the network effects to become obvious.

iPad’s attraction is deeply bound up in apps, and the idea of an instant-on, always-connected general-purpose computer. But the initial use, and apparently for some the only use, is as a browser with a new UI. It’s only really through acquiring a selection of applications that deliver the user value that the utility of the platform becomes obvious.

Likewise for Twitter. When you first log in, it invites you to contribute, but gives you no obvious benefit upfront. While twitter has done a great job of helping people start following others immediately, the real value I get from Twitter is the ability to get a needle-thin narrowcast of news that is super-relevant to my interests and entertainment that I love. Getting to that place has been an exercise in curation, of carefully discovering people that tell me things I’m interested in, and weeding out the red herrings. The value of Twitter has grown in proportion to the effort I’ve spent tuning in to those sources. But that process is emergent and ad hoc; nobody tells you how to undertake that process, but most people seem to discover it on their own.

Newsweek rips into Microsoft; it’s not pretty, but much of the logic is sound.

What journalists like Lashinsky are saying, by implication, is that they’re not afraid of having their access cut off, because they no longer believe that access is worth anything, because these days it’s hard, perhaps impossible, to imagine that any top executive at Microsoft would have anything to say other than the same old rubbish and spin.

via Drumbeats: The Tech Press Turns on Microsofts Ballmer - Newsweek.

I was at the chemist today, begging for more Codral. To my surprise (and Jenn’s), the (very sympathetic) pharmacist told us outright that the phenylephrine (PE) versions don’t do anything, that there was no good research showing effectiveness, that the Pharmacist’s Guild recommended against it, and that we should insist on the real thing with pseudoephedrine (PSE).

To say I was shocked doesn’t cover it.

Interesting disclosure of search warrants and evidence receipts. But most interesting of all, the person typing the evidence inventory seems to be oblivious as to the existence of either the shift key or spell check.

But for all that noise about citizen journalism, is a blogger legally a journalist, and entitled to those protections?

Police Seize Jason Chen’s Computers - Iphone 4 leak - Gizmodo.

MIT Sydney, through my association with the University of Ballarat (site of the ICSL) asked me to be the occasional speaker at their Graduation Celebration today. I was terrified, but agreed. Here’s what I said.

This is truly amazing; using makeup patterns to defeat facial recognition systems.

What would camouflage look like if it let you hide from a camera?

By reverse engineering the algorithms behind face detection, I generated a preliminary series of images that could be the building blocks of anti-surveillance makeup.

via AH Projects - Thesis; hat tip William Gibson

I’ve been reevaluating the way I manage my non-work life, particularly with regard to the tools I use. Hiveminder’s served me very well, but I’m trying new things.

As a result, I completed the (delightfully named) Here Be Dragons - Atlassian Integration Guide, ending up with a complete Crowd+Jira+Confluence+Fisheye+Crucible+Bamboo stack on my formerly underworked Linode. I’ve now got a couple of projects in Jira which are slowly filling up with my open actions; the misc bucket has the project code “LIFE”.

I clicked through the Dashboard to get some info about some of the miscellaneous actions I’ve got outstanding only to get this message:

Thanks, Jira; good to know the resolution of my life will be empty. :)