Quixotic Californian crusade to officially recognize the hellabyte and hellagram is going hella nowhere

The Reg speaks to Stanford boffin behind decade-long push for SI prefix

In 2010, Austin Sendek, then a physics student at UC Davis, created a petition seeking recognition for prefix "hella-" as an official International System of Units (SI) measurement representing 1027.

"Northern California is home to many influential research institutions, including the University of California, Davis, the University of California, Berkeley, Stanford University, and the Lawrence Livermore and Lawrence Berkeley National Laboratories," he argued.

"However, science isn't all that sets Northern California apart from the rest of the world. The area is also the only region in the world currently practicing widespread usage of the English slang 'hella,' which typically means 'very,' or can refer to a large quantity (e.g. 'there are hella stars out tonight')."

To this day, the SI describes prefixes for quantities for up to 1024. Those with that many bytes have a yottabyte. If you only have 1021 bytes, you have a zettabyte. There's also exabyte (1018), petabyte (1015), terabyte (1012), gigabyte(109), and so on.

Support for "hella-" would allow you to talk about hellabytes of data, he argues, pointing out that this would make the number of atoms in 12 kg of carbon-12 would be simplified from 600 yottaatoms to 0.6 hellaatoms. Similarly, the sun (mass of 2.2 hellatons) would release energy at 0.3 hellawatts, rather than 300 yottawatts.

Over the past decade, other terms have been suggested for designating 1027 bytes, including brontobytes and xanobytes. There's also geobyte for 1030 bytes, and googol for 10100 of something. These terms see occasional usage but they're not standard measurements.

There are also binary-multiple International Electrotechnical Commission (IEC) prefixes, which are based on powers of two, like one kibibit (1 Kibit = 210 bit = 1024 bit), not to be confused with one kilobit (1 kbit = 103 bit = 1000 bit). But decimal-based terms tend to be preferred over binary.

Friends in high places

In May 2010 Google weighed in to support "hella-" by customizing its search engine. If you enter a search like "bytes to hellabytes" in the Google Search box, the results page will present a unit conversion widget.

In a phone interview with The Register, Sendek, now a Stanford visiting scholar and CEO of a software company called Aionics, said he had a friend interning at Google at the time who managed to get the conversion widget implemented.

A year later, Wolfram Alpha added support for "hella-" calculations. Rival prefixes like bronto- have yet to be endorsed thus.


Aptare: Eight exabyte-juggler pimps its 'data centre MRI' product


Sendek said "hellabyte" initially started as a joke with some college friends but became a more genuine concern as he looked into how measurements get defined and as his proposal garnered support.

While there wasn't much need for measuring hella- quantities at the time, he believed it could be useful for astronomical measurements. "The diameter of the universe is about a hellameter and a half," he said. "The mass of the sun is about a hellagram."

Alas, it's doubtful the hella- prefix will ever make it out of the vernacular into the sanctioned language of science.

According to Richard JC Brown, a fellow in chemical metrology at the UK's National Physical Laboratory, both "hella-" and "bronto-" have issues. The current leading candidates for new measurement terminology appear to be "ronna-" and "quetta-", which Brown proposed in the journal Measurement in 2019.

"The proposals currently being considered for extending upwards the range of International System of Units (SI) prefixes are 'ronna,' symbol: R, for 1027; and 'quetta,' symbol: Q for 1030," said Brown in an email to The Register.

"Hella-" and "bronto-"," he said, "wouldn't work well within the SI because they would likely use symbols B and H, which are already in use for other units or prefixes within the SI, and it is best practice to avoid this in order to lessen the chances of confusion."

The current global datasphere is probably around 0.1 yottabytes and so it will be a few years yet before the global datasphere surpasses 1 yottabyte

Brown said it's not clear when terms for such large quantities will be needed, but he believes they should be settled on sooner rather than later.

"The current global datasphere is probably around 0.1 yottabytes and so it will be a few years yet before the global datasphere surpasses 1 yottabyte," he said.

"But the rate of increase of the global datasphere is increasing rapidly. Add into this the need to start considering and talking about larger data sizes for future planning and conceptual studies, plus the advent of new, disruptive technologies such as quantum computing that may radically change the quantity of data produced, and it makes sense to have these in place soon to be prepared for when they are needed."

The soonest that could happen, Brown said, is in November, 2022, at the quadrennial meeting of the International Bureau of Weights and Measures (BIPM)'s General Conference on Weight and Measures, where changes to the SI usually must be agreed upon.

Asked how he felt knowing that "hella-" isn't a likely contender as a future SI prefix, Sendek said, "I'd be lying if I said it didn't disappoint me a bit." He said one reason he backed "hella-" was as a reminder to scientists and mathematicians not to take ourselves too seriously.

"A little bit of humor is nice now and then," he said. ®

Biting the hand that feeds IT © 1998–2021