MisEntropy

"The opposite of a hero isn't a villain...

... it's a bystander."

- Matt Langdon, Hero Construction Company

Via The Psychology of Heroism

[Original pic by westpark]

How advertising can help improve machine translation of languages

Automatic translation of languages by machines has been a standard fixture of science fiction - and like quite a few other sci-fi standards, it has been painfully slow to cross over to reality.

A recent breakthrough called 'statistical machine translation' promises to pluck this fantasy out of its permanent future resident status. Here's a description of how the technique works (from an Economist article in June 2006):

"Statistical translation encompasses a range of techniques, but what they all have in common is the use of statistical analysis, rather than rigid rules, to convert text from one language into another. Most systems start with a large bilingual corpus of text. By analysing the frequency with which clusters of words appear in close proximity in the two languages, it is possible to work out which words correspond to each other in the two languages. This approach offers much greater flexibility than rule-based systems, since it translates languages based on how they are actually used, rather than relying on rigid grammatical rules which may not always be observed, and often have exceptions."

Not surpisingly, the company which is at the forefront of statistical machine translation is Google. Whenever you use Google Translate, this is what's happening behind the scenes (via Economist Feb 2010):

"For translation, the company was able to draw on its other services. Its search system had copies of European Commission documents, which are translated into around 20 languages. Its book-scanning project has thousands of titles that have been translated into many languages. All these translations are very good, done by experts to exacting standards. So instead of trying to teach its computers the rules of a language, Google turned them loose on the texts to make statistical inferences. Google Translate now covers more than 50 languages, according to Franz Och, one of the company’s engineers. The system identifies which word or phrase in one language is the most likely equivalent in a second language. If direct translations are not available (say, Hindi to Catalan), then English is used as a bridge."

But currently there are a few drawbacks with statistical machine translation - which have primarily got to do with the kind of readymade translated texts they rely on. From yet another recent Economist article:

"It is getting better, but it still struggles with colloquialisms and idioms. As Ethan Zuckerman, co-founder of Global Voices and a researcher at Harvard University, puts it: “If you sound like an EU parliamentarian, we can translate you quite well.”

What's foxing these gargantuan statistical crunching machines is the linguistic equivalent of the last mile problem. The everyday spoken language which is rarely captured and archived - let alone translated into a dozen languages before archiving.

For advertising and advertisers who believe in their existence serving a larger purpose and providing a public good, there might be an opportunity here.

Ads and commercials are routinely translated into different langauges - especially when they come from global multinationals. Because they aim to communicate with end consumers, these also contain the kind of colloquilaisms and idioms that EU and UN speeches lack.

What if advertisers could provide Google or a non-profit third party transcripts of these ads and commercials along with the translations that have been professsionally created through human experts. If a sufficiently large number of advertisers commit their future and past archives, it may end creating a formidable archive of spoken and eveyday lingo for the statistical inference bots to bite their silicon teeth into.

The drawback - as some skeptics will point out - is that the language of advertising may not be any less stitled and far removed than a EU speech. On the other hand, for advertising that seeks to leave behind a cultural impact, this could provide a platform to really make it come true.

If such a thing can be worked upon, the advertsing itself may have a limited run - but it's value could live on forever by providing us with better machine translation for ages to come.

[Original pic by Otie]

The possible negatives of clean energy

In November 2009, Scientific American - with an eye on then upcoming Copenhagen Summit - published a manifesto piece titled "A Path to Sustainable Energy by 2030."

It was a bold vision charting out how to get no less than all of the world's energy requirements through wind, water and solar power (WWS.) Eschewing even nuclear power, the authors had an uncompromising vision considering "only technologies that have near-zero emissions of greenhouse gases and air pollutants over the entire life cycle, including construction, operation and decommissioning."

The piece went into some detail about how to overcome the hurdles that will no doubt present themselves along the way. The authors went as far as to provide projections of the availability of metals that would be required in turbine gearboxes. Smartly they also anticipated and provided for the possibility that "the U.S. could be trading dependence on Middle Eastern oil for dependence on Far Eastern metals."

I am no climate skeptic - but even I found myself believing the plan a bit too far-fetched. Not because it wasn't desirable, but because it was unrealisable.

I also forgot all about it, until I found this letter by Paul Roetling of Grand Island, NY to the editors in this month's edition of Scientific American:

I found it surprising that in “A Path to Sustainable Energy by 2030,” Mark Z. Jacobson and Mark A. Delucchi do not mention the effects of the suggested energy sources on climate. The authors propose to absorb about six terawatts of energy from about 60 terawatts available in the wind, or about 10 percent of its total energy. Because the winds, at least near the U.S., usually flow around highs or lows, where the speed and related Coriolis force tend to maintain the pressure difference, I can easily envision that absorbing the energy will change the rate at which the pressure centers collapse. How this would change the weather, I do not know, but it must make a change to give us some of the energy. Possibly, the weather change would be an improvement, but as a believer in Murphy’s Law, I would be surprised. About 100 years ago dumping garbage into the ocean was justified because the oceans were infinite compared to the effect, so no one calculated how much was allowable. Let’s be smarter this time! Why not do the calculations before we cause more problems?

Paul's astute observations lends weight to my own belief that we have to reduce our energy consumption as a species rather than just pin our hopes on finding alternatives to all our energy needs.

All of our choices, no matter how carefully we consider and plan for them, will have externalities. Some of them we can surmise - most of them we cannot. Not until it's too late, ie.

[Original pic by Insight Imaging: John A Ryan Photography]

How to put together a corporate social media policy in 5 minutes

PolicyTool for Social Media is a policy generator that greatly simplifies the task of creating a working social media policy. Answer about a dozen and odd questions and before you can hop across to Facebook, your policy is ready for your to copy paste anywhere.

Based on my answers (and relative lack of paranoia) here's what a policy for my own company (if and when I start one) would look like:

MisEntropy Social Media Policy

This policy governs the publication of and commentary on social media by employees of MisEntropy and its related companies ("MisEntropy"). For the purposes of this policy, social media means any facility for online publication and commentary, including without limitation blogs, wiki's, social networking sites such as Facebook, LinkedIn, Twitter, Flickr, and YouTube. This policy is in addition to and complements any existing or future policies regarding the use of technology, computers, e-mail and the internet.

MisEntropy employees are free to publish or comment via social media in accordance with this policy. MisEntropy employees are subject to this policy to the extent they identify themselves as a MisEntropy employee (other than as an incidental mention of place of employment in a personal blog on topics unrelated to MisEntropy).

Publication and commentary on social media carries similar obligations to any other kind of publication or commentary.

All uses of social media must follow the same ethical standards that MisEntropy employees must otherwise follow.

What fiction writers can teach us about writing creative briefs

Recently, Guardian invited prominent authors to list out their 10 rules for writing fiction. It became a popular feature, collating the wisdom of several authors and even overflowed into a part two.

Reading it, I wondered if a few of these rules can be picked out and applied to something closer home - like writing a creative brief. It does seem like a stretch as the two activities are possibly diametrically opposite, if nowhere else but in our own minds.

Still, armed with my own thoughts on writing creative briefs, I set out to trawl the entire list and came up with this - 10 rules for writing creative briefs from fiction writers.

I have copied each rule and tweaked it - minimally - to suit our need. My own edits in the rule are italicised. The author's name and the corresponding rule number (from their list) are mentioned in brackets for reference.

1. Keep your exclamation points ­under control. (Elmore Leonard, 5)
2. Try to leave out the part that readers tend to skip. (Elmore Leonard, 10)
3. Do change your mind. Good ideas are often murdered by better ones. (Roddy Doyle, 8)
4. Learn poems by heart. (Helen Dunmore, 5)
5. Have regrets. They are fuel. (Geoff Dyer, 6)
6. Beware of clichés. Not just the ­clichés that Martin Amis is at war with. There are clichés of response as well as expression. There are clichés of observation and of thought – even of conception. Most creative briefs, even quite a few adequately written ones, are ­clichés of form which conform to clichés of expectation. (Geoff Dyer, 8)
7. When information becomes free and universally accessible, voluminous research for a creative brief is devalued along with it. (Jonathan Franzen, 5)
8. Style is the art of getting yourself out of the way, not putting yourself in it. (David Hare, 3)
9. Ignore all proferred rules and create your own, suitable for what you want to say. (Michael Moorcock, 10)
10. If it reads like a creative brief, rewrite it. (Elmore Leonard, summation)

Of course, these rules are my own version of what I think these fiction writers would say to us. I would love to hear what other rules you would pick out.

Virtual Black Boxes for Aircraft

Leafing through an old copy of New Scientist I came upon this wonderful idea in the letters to the editor column. It's so simple and elegant in hindsight - I found myself wishing that I had thought of it.

To give it its due, here's the entire letter sent by Gerry Nolan of Sydney, Australia to New Scientist. (Printed in 16th September 2009 issue.)

"Questions on the safety of modern aircraft raised in your report on the loss of an airliner in the mid-Atlantic on 1 June may never be answered if the plane's "black box" flight recorders cannot be found (13 June, p 19). The drawback of the current system of on-board black boxes is that they may often not be found, so valuable information as to the causes of accidents is lost.

Other reports of the incident mentioned that before the crash some of the plane's flight data was sent by satellite. This made me wonder why all flight data can't be sent by satellite for every flight. This data could then be recovered from a "virtual" black box, regardless of the physical inaccessibility of the crash site.

The Aircraft Communications Addressing and Reporting System (ACARS) that was used to send data from the doomed aircraft allows only limited data to be transmitted. But why not combine satellite communication with the internet to record a much greater range of data? Once the transmission has been received at an internet node, of which there are thousands, it immediately becomes available worldwide.

The technology for this already exists. There are well-established standards for digitised transmission, and encrypted transmission is used for many internet communications. If the data were continuously transmitted and recorded on the ground, the crash investigators would have instant access to it, and would no longer have to wait for the physical black box to be found - if indeed it ever is.

Automatic warnings and alarms could also be added at both ends. What's more, the exact location of the aircraft would be known, allowing rescue missions to get under way without delay. For the overwhelming majority of flights that are completed safely, the recording could be automatically wiped when the plane lands."

A couple of points I'd like to add here. The use of a virtual black box shouldn't always preclude the continued use of on-board black boxes. In the case of loss of contact preceding a crash, the on-board black box, if found, can reveal important additional clues as to what went wrong. It already does that now, but what Gerry's brilliant idea suggests is that the black box also be continuously backed up using the internet.

Secondly, while Gerry suggests that flight data of successful flights be wiped after the plane lands, I find myself believing that storing the data for a few more flights - possibly with a window of a few weeks - should instead be stipulated. In the case an aircraft is lost, one can analyse not just data about its current flight but also see if anything suspicious cropped up earlier - and if no action was taken by the airline/operator. (Current black boxes record data for anywhere between 30 minutes and 2 hours, overwriting earlier data as they go along.)

Simple and feasible as the idea is, I doubt though that the airlines would be in a hurry to embrace something like this. As an economist would put it, there are misaligned incentives at work here. The airline business isn't flying high at the moment and any expense adding to the possible safety of flying is just that - an expense with no one to pass on the bill to.

They find it worth their while to try and sell us on-flight internet access so we can update our facebook pages instead.

[Original pic by Telstar Logistics]