Too Big to Scale

"We’re looking at The Cloud from both sides now."


While it is not likely that the heterodox economist E.F. "Fritz" Schumacher was the first to use the term “appropriate technology” — he preferred “intermediate technology” — he certainly had a big role in defining it. In Small is Beautiful he described it as the “middle way,” which dovetailed nicely with his elucidation of Buddhist economics, or what Mohandas Gandhi called "Economy of Permanence." 

According to Schumacher, a technology is appropriate to preserve, adopt and adapt if it is truly village scale, lying in that mid-range between individualistic technology (toothbrush, smartphone, coffee cup) and industrial-scale (pharmaceutical laboratory, steel mill, railroad).

Examples of village scale are the old bakery, perhaps a large stone or brick oven where families bring their doughs to become breads; the bicycle repair shop; or a family-run tofu shop (as in the 10,000 or more in any large Japanese city) because handcrafted tofu is much to be preferred in taste, texture and nutrition over machine-produced.

James Earl Jones as Locust-Man

As early as the 1960s Schumacher, as president of the UK Soil Association, was correctly diagnosing what was wrong with the atom as an energy source. In 1977 he published A Guide for the Perplexed as a critique of materialist scientism. It was also a foray into the nature and organization of knowledge. He championed the style of Ivan Illich's conviviality: user-friendly and ecologically suitable; applicable to the scale of the human and natural community.

Born in the late 1940s, we were witness to Moore’s Law from its birth. We watched electric typewriters replace manual portables, then IBM Selectrics arrive with their changeable font-balls and auto-erase tape. We were there when punch cards and tape readers began to type form letters like a player piano. From the days of our youth, hand calculators kept getting smarter than we were. 

In the late 70s we automated our Plenty Office and the Book Publishing Company with arrays of linked, part home-brew, part off-the-shelf, CPU-and-dumb-terminal minicomputers. Soon came inexpensive personal computers that put desktop publishing and spreadsheets into the hands of the masses and made small fortunes for Apple, Atari, Dell and Texas Instruments.

Office networks of linked hard-drives using first ethernet and then wireless LANs and WANs were middle scale appropriate technology as long as you could service the devices or maybe even build them yourselves within the village. All was well on this good earth. Desktop computers were like tractors or teams of oxen, shortening the time it took you to furrow your inbox or do your taxes.  

Then came The Cloud upon the land. Cut to the scene in The Good Earth where the Chinese farmers look to the sky as their faces darken — the locusts are here! That was about 10 years ago, or 5 generations in factor-four Silicon Time.

Boston-based research outfit Forrester calls cloud computing—that’s public cloud computing—a “hyper-growth” market. In a recent report, it predicts the market for cloud services will grow to $191 billion by 2020, a 20 percent leap from what it predicted just a few years ago. “The adoption of cloud among enterprises, which is really where the money is, has really picked up steam,” Forrester analyst John Rymer recently told us. “It’s a big shift. The cloud has arrived. It’s inevitable.”

- Cade Metz, Wired 12-22-15

Getting back into our annual workshop schedule here at The Farm, we find ourselves stuck without a middle way, with no “village scale” with regard to either email or accounting. We have always suffered the digital divide by electing to live in a rural area in a country without Net Neutrality, but we take clean air and birdsong more seriously than ones and zeros.

What passes for broadband in rural Tennessee would be laughable in Romania or Thailand. We live beyond the profitable reach of the cable companies, or even DSL from the quasi-federal phone monopoly. Getting a dumbphone mobile connection here can be challenging, never mind G3 or G4. We pay far too much for far too little connectivity, but then, welcome to the unpaved precincts. Have you seen the stars at night?

But now they only block the sun
They rain and snow on everyone
So many things I would have done
But clouds got in my way

— Joni Mitchell, Both Sides Now (May, 1969)

We’re looking at The Cloud from both sides now. Many, if not all, of the email and accounting packages that have the capabilities we need have discontinued stand-alone functionality and hard drive data storage on your personal device in favor of wireless subscription plans. An unbeckoned choice is being thrust upon us. Either we late-migrate to the Cloud and trust in her all-knowing beneficence (and suffer indignities whenever there is no connection) or we put up with rapidly-shrinking features and capabilities. 
 
For code-writers keeping legacy software working may be somewhat easier. But most code-writers are Cloud addicts, not old school.

We use Photoshop but seldom have need for the other Adobe apps packaged into their (formerly $3650) Master Suite. To us it was worth several hundred dollars plunked down every few years to have that one app. We’ve tried GIMP and other freeware but they are no substitute for Photoshop. Now a subscription to Adobe’s Creative Cloud would cost us about $2,400 — assuming the price doesn’t go up. And that is just one subscription, from one cloud provider.

Microsoft rolled out Office 365 in 2011 but still plans to sell packaged software for a while, which makes sense given how much of the world has weak to nil internet connectivity. “Unlike Adobe, we think people’s shift from packaged software to subscription services will take time,” Microsoft told Wired.

The largest cloud storage provider, Amazon Web Services, reported $2.41 billion in revenue for the fourth quarter last year, or more than $9.6 billion in annualized sales—and that’s after the $10-billion-dollar company Dropbox ported off Amazon to build its own server farms in Q3.

Dropbox calls each of its storage machines a Diskotech. “The thing we care about the most is the disk,” its chief engineer told Wired. “That’s where all the bytes are.” 

Measuring only one-and-half-feet by three-and-half-feet by six inches, each Diskotech box holds as much as a petabyte of data, or a million gigabytes. Fifty of these machines could store everything human beings have ever written. Maybe even all the cute kitten videos on YouTube (“Maru gets into a box” - “大きな箱とねこ” - 8.1 million views).

At one point in 2015, when it was moving from Amazon to its own 40 acres and a mule, Dropbox was installing forty to fifty racks of hardware a day, each rack holding about eight individual machines. That installation rate continued for nearly six months. They surpassed Peak Kitten in the first month.

We have had the trauma of a terabyte data fail. It is not pretty. It means we now have to have 2 or 3 terabyte safety redundancies. If you go to DVD you can become dependent on legacy hardware (DVD readers and burners), calling up recollections of floppies, cassettes, optical readers, etc. we may still have in the attic but prefer not to think about. 

A flash drive is ephemeral - how many years will it hold its charge without any degradation or chance encounters with moisture, temperature change or magnetic fields?

We want to be able to access 20-year-old data using only the power of a Biolite Stove and no cloud. We can do that right now with an iPad and a portable HD. Can we do it still in 2017?

There may come a time when we just have to go our own way and de-cloud. At the moment we are struggling to remain amphibian, with a webbed foot in each world. Thanks for all the fish, but for now we intend to keep our paper-based bookkeeping and a sharpenable pencil.

Many years ago Amory Lovins’ Brittle Power described how lack of prudence and foresight had allowed city and regional planners to erect a monumental infrastructure of energy supply that keeps the lights on at night across North America but can be taken down by a tree branch falling on wires in a blizzard, or a pipe bomb in a pipeline.

The same kind of blind spot infects the planners of the Cyberverse. Mighty as they be, they are not Gods. To get to be in their club, you have to take the blue pill — to believe the separate reality the Google-vets believe; the one with Space X missions to Mars and fusion-powered Teslas.

This represents an attitude that began with Google and has gradually spread across Silicon Valley. Google was so successful not just because it built a pretty good Internet search engine, but because it built the underlying technology needed to run that search engine—and so many other services—at an enormous scale. Facebook, which recruited countless ex-Googlers, did much the same. And so did Twitter and its ex-Googlers. And, now, so has Dropbox. To become a giant, you may have to stand on the shoulders of others. But once you become your own giant, you start to feel like you need to build a home that’s just right for you.

— Cade Metz, Wired 3-14-16

The problem, as we see it, is that the parallel reality field is eating away the brains of its wizards. Wormhole-brained, they keep edging farther out onto the limb of a system that is just one fallen-tree-branch or cyberattack away from ruin. Worse, they are forcing the rest of us to follow along and add our weight to that same weak limb.