James Tindall : Blog

1MB ~= 1 lump of coal?

Fri, 01 May 2009 11:19:00 +0100

It’s hard to imagine that our browsing activities on the internet require any energy at all. Somehow it’s easier to accept that watching TV requires energy and to grasp that the longer we watch the more energy we consume. The NET is different. The information is not transferred in a steady stream as with television but more erratically. From low bandwidth activities like email and IM to video streaming and Skype calls.

In this TEDtalk Jay Walker states that it takes the energy in 1 lump of coal to move 1 MB of information across the NET. Of course not all of the energy we use comes from burning coal. However, this is an important idea. Whether Walker’s calculations are accurate or not, the “1 lump of coal = 1MB” soundbite at least acts as a memorable reminder of the link between information consumption and energy consumption.

For those wanting more in depth calculations, this article analyses the link between information consumption and energy consumption in more detail calculating that it takes 323,000 KW·hs of energy to download 100 web pages.

This Dark Comedy

Tue, 24 Feb 2009 15:49:00 +0000

To know how good you are at something requires the same skills as it does to be good at that thing. Which means if you’re absolutely hopeless at something, you lack exactly the skills that you need to know that you’re absolutely hopeless.

And this is a profound discovery. That most people who have absolutely no idea what they’re doing have absolutely no idea that they have no idea what they’re doing. It explains a great deal of life.

This wonderful quote - taken from a talk on creativity given at the world creative forum by John Cleese - like much of the comedy he has written over the years, is both deeply profound and darkly comical. I laughed out loud the first time I heard this. The truth of it and the horror of it’s implications ringing through my head.

Standard Semantic Markup for Scripting Language Reference Pages

Mon, 19 Jan 2009 10:55:00 +0000

I frequently need to jump between various programming / scripting languages and have often wondered if there’s a site out there that helps with this process by mapping the equivalent functions in the various languages.

I’m yet to find such a site and have recently wondered about the feasibility of a microformat for marking up programming language reference pages. I first thought about this while viewing the reference page for processing:

http://processing.org/reference/index.html

Would it be possible to mark up similar reference pages for other scripting languages like php, ruby, javascript, actionscript and python using a standardised set of class names and markup structure for the methods and functions that are common throughout each language?

I put this question to the microformats google group and Ilya Radchenko made the point that the scale and depth of languages like php would make this extremely difficult. But such an effort would not need to cover the idiosyncrasies at the dark edges of each of these languages. To be useful it need only map the common core, most often used functions like those dealing with strings, text, arrays and the more common math methods.

I’m sure that a consistent, standard and semantic markup format for scripting language reference pages and the resulting consistency in presentation would be a huge help to developers. As the microformat motto goes, “designed for humans first, machines second”, I wonder what the benefits of having this data machine readable would be?

One language, multiple platforms?

Thu, 11 Dec 2008 13:27:00 +0000

Wouldn’t it be nice to be able to code interactive artworks in a single language and easily deploy them on multiple platforms? Recent developments leave me wondering if this day may come sooner than I’d anticipated.

With the release of Alchemy - a way to compile C and C++ code to run on the ActionScript Virtual Machine - it might just be possible to code in openFrameworks and deploy to both the web (as well as other flash-player enabled platforms) and, with a little additional code to cater for particular interface considerations like multi-touch, also to deploy to the iPhone.

Among those leading the way in terms of successfully porting interactive art originally written in C++ to the iPhone are Andreas Müller and Lee Byron. Müller with his beautiful All Seasons and Byron with a version of Golan Levin’s wonderful Yellow Tail.

I look forward to finding time between client work in 2009 to cut my iPhone App teeth while scaling this small tower of babel.

Personal and helpful customer support in the cloud?

Mon, 08 Dec 2008 12:24:00 +0000

Timely, helpful and personal responses to email requests for support have always been important but in the past would probably not have affected my choice of application. This is no longer the case. Now that many of my most used applications reside in the cloud rather than on my desktop (along with much of my most valued data) the responses I receive to support request emails will without doubt influence where in the cloud I choose to upload, aggregate and organise my data.

Over the past few weeks I’ve sent a number of support requests / feedback emails to friendfeed and every time I received a quick, polite, and most importantly personal and helpful response. I even saw a quick change to the output of the widget after complaining about having to click to show all the images in a media post!

Positive support exchanges with other companies like Tumblr and Claimid have also been key factors in my decision to stick with them rather than competing services.

In complete contrast, a couple of days ago I sent a support request to delicious. I first received an automated response from Yahoo! Customer Care that my support request had been received and would be dealt with shortly. Twelve hours later I received a further message from Yahoo! Customer Care that despite being signed “Florence - Delicious Customer Support” was every bit as generic, unhelpful, impersonal and probably automated as the initial, explicitly automated response.

My experience requesting support from flickr several months ago (who are also of course owned by Yahoo!) was even worse. But is this a problem of scale? Does personal, helpful support simply not scale? Is it logistically impossible for really large companies to provide truly effective support to their users? Perhaps services like Get Satisfaction are the answer?

What seems clear to me is that for services operating in the cloud, especially as the data we pour into them becomes increasingly portable, good customer support could well become the deciding factor between them and their competitors. Now then, what was my magnolia account login again…?