32/250 What about when it breaks?

The internet isn't Life.

The light kept flashing.

It had been flashing for the best part of half an hour.

Even when it stopped flashing, nothing connected. Huffing and sighing all over the place, I grabbed my phone and reluctantly turned on the hotspot, knowing it would chew data like a maniac, and pretending I didn’t care.

Just as the hotspot established itself, the mac rebooted.

And rebooted.

And rebooted.

And then calculated how long it would take to update (and that took forever).

Then it told me it would be 33 minutes.

I stared dumbly at the screen; simmering, that’s what I was doing.

Eventually, I got my shiiit together, pulled out my laptop and turned it on. But even though it turned on, the trackpad didn’t work for some reason, the mouse buttons didn’t respond, and while it connected to the network, there was “no internet”.

At that point, I ought to have known better than to persist, but I did. Usually, it’s a sign to accept whatever is going on, put on some sneakers, and go outside for an hour or so for a walk. But, being the very definition of Stubborn when the mood takes me (just ask my parents), I didn’t do this. Finally, getting into the groove and almost at the point of delivering a job to someone, I pulled up Flickr to get an album link for a client, only to see that Flickr was down.

I might have sworn copiously at this point, using words that would make a sailor blush. Flickr is never down. That attempt was about six hours ago; and—racing off to check—it is still down.

You and I have lost something, in racing to move our businesses to the cloud, in prioritising movable convenience over sensibility.

In the downtime that I had today, it struck me that having your entire business locked up in the cloud is the shittest risk mitigation factor ever, in a country with unreliable internet. I was, at the time, rather het up about the whole thing; that’s why that initial assessment is wrong.

Cloud is great, for loads of reasons that speak against on-premise servers. But what you give up is control over actual files. I’ll give you an example: When you use Google Suite, the only way to keep your storage costs down is to use Google’s proprietary format. The problem is that it’s proprietary and you need to be online in order to read those files, unless you have them set to be viewed offline.

The risk of accidentally removing your own access to these files is very real; in fact, I did it by accident not two years ago, in a failed attempt to keep a backup. What I’d backed up were links which broke when the files were moved. That’s a story for never.

When I began mapping the current situation of the platforms to which the business is beholden, I began to see the problem: A business built on Google is effectively locked into it unless you decide to lock yourself into someone else’s (more bloated) system; namely Microsoft.

Unless you roll with Markdown files (as we do, for client delivery); maintain git repositories; and use open source files for the remainder. This is the most sensible result (perfectly blended cloud/client, access forever), but you sacrifice ease of sharing and mobile access.

There is no clear-cut solution to this, in my opinion.

The solution for us will be to move 100% into a docs-as-code framework, and to maintain as much as we can in Git-tracked, Markdown respositories. It’s already proved to be a winner for our client work, so the next step will be scaling it out to the rest of the business, moving and converting existing file storage, and moving forwards.

The beauty of working in Git repositories isn’t just the impeccable audit trail.

It’s being able to work offline, wherever you are, without sacrificing collaboration.

While that would have been beautiful for today, my friend, there is one thing better that came out the other side: The lesson.