ICTlogy

Working with Information Society / Digital Divide indicators is a tricky thing to do, as definitions (along with technology) change in short periods of time. Some months ago, Tim Kelly asked me what did I consider “broadband”, as it was one of the hottest issues that researchers, in general, and the ITU, specially, had to deal with. Let’s see an example.

Broadband is defined by the International Telecommunication Union (ITU) Telecommunication Standardization Sector (ITU-T), in their recommendation I.113, as transmission capacity that is faster than primary rate Integrated Services Digital Network (ISDN) at 1.5 or 2.0 Megabits per second (Mbits). On the other hand, the OECD gives its own definition of broadband stating that for a service to be considered broadband, [the threshold] in respect to downstream access [should be up] to 256 Kbps. The fact is that, as the OECD itself admits, Network operators widely advertise DSL and cable modem services to users starting at 256 Kbps as being
‘broadband’. Actually, the Core ICT Indicators, promoted by the Partnership on Measuring ICT for Development — partnered by the ITU — also defines broadband as technologies that provide speeds of at least 256 kbit/s, where this speed is the combined upstream and downstream capacity.

Summarizing, all of these are technical definitions, based on the fact of transmitting more than one data stream in the same wire by using different frequencies or channels. But for the not-technical user, broadband is strictly tied to “effective” speed, or, in other words, “subjective” speed: if your 1 Mbps is the slowest in town, it is no more broadband. This was Tim Kelly’s point last time we met.

Thus said — and leaving technical issues behind to focus in this “subjective” broadband perception — my proposal is to build a basket of tasks the way economists use to calculate changes in inflation based on a basket of products. Of course, this basket of tasks is also likely to evolve with time, but what is crystal clear is that the technical definition of broadband (the one about channels) is no more useful, and the decision to state that i.e. 256 Kbps is broadband should lean on objective basis more than “network operators advertisments”.

Proposal of a basket of tasks for a broadband definition

• Work in online, synchronous collaborative environments with rich media: VoIP, videoconference, screencasting, presentations/drawings…
• Work intensively/exclusively with online, asynchronous desktop/office applications: word processors, spreadsheets, math/scientific calculators…
• Usually access online applications with richest graphical content: SIGs and mapping tools, 2D and 3D simulators and environments
• Have online environments as primary communication and information channel: e-mail, instant messaging, browser and desktop widgets. It includes software downloads and updates.
• Manage a website: upload files, install applications, change configuration/setup. It does not include writing on a weblog/wiki and other low-tech “webmastering”
• Work with remote computers or in grid computing, including intensive use of P2P networks

This basket of tasks and the minimum speed required to perform them correctly/comfortably should help in setting the threshold of what we could call broadband. As those tasks will evolve dynamically along time, same will happen with the broadband threshold. As an example, some years ago you needed a then-so-called-broadband to check the Perry-Castañeda Library Map Collection when looking for geographical information, as most maps are some hundreds of Kb weight, being the heaviest up to some Mb. Nowadays, you would browse Google Maps, for which a today-so-called-broadband is required, maybe more than the “official” 256 Kbps to browse at ease.

Proposals, corrections, comments gratefully welcome.

Further reading

• Birth of Broadband – Frequently Asked Questions

Previous post: Workshop. Fostering Innovation in eGovernment (part IV)

Next post: e-Stas 2007 and the experts group on ICTs for nonprofits