It was a normal Tuesday morning, and things stopped working. People were stood panicked at ATMs, unable to take out the cash they needed to live their lives; others reported being stuck in foreign countries with no way to pay for their public transport; some said they had been stranded in the middle of some of the most important moments of their lives, unable to make the initial mortgage transfer that would allow them to move into their house.
It would be tempting to say that it was an otherwise normal Tuesday morning. But, in fact, when Revolut and other important services went down because of a Google Cloud outage, it felt dramatic but unsurprising. Our important systems feel more and more like they are buckling – if they haven't broken yet, then it is only a matter of time.
But that same sense of resignation might be exactly the worst thing for us to feel. We have grown accustomed to living through a period of unprecedented online precarity – unprecedented both in our reliance on those systems and our feeling that they could fall apart at any time – and that means that it can sometimes feel like a disastrous global outage, cyber attack or other shock is greeted with a simple: oh, of course.
Revolut was far from the only organisation whose systems stopped working properly. The same morning, outages were reported at everything from delivery company DPD to Rightmove. But Revolut bore much of the attention in part because the outage affected the very fabric of daily life.
If online outages or cyber attacks of this kind are met with much concern, it tends to be of a ponderous, global nature. This goes to show that we rely on these systems too much, people say, which is so obviously true that you can't do an awful lot with it. What are we supposed to do in response? Live like cavemen so that we don't rely on anything much at all? We only suffer during these (thankfully limited and temporary) collapses because these things provide us such value. Of course we rely on them.
A more concrete and slightly more useful response is that we rely on specific systems too much. Tuesday's outage was the result of one particular location run by one particular cloud provider going down. It does feel bizarre that so much of our apparently various online life is actually dependent on very small number of companies, which tends only to become clear when one of them suffers issues. But, again, what are we supposed to do – have every company build its own internet infrastructure by itself? It's not just wasteful but a direct challenge to the inter-related, co-operative spirit that makes the internet work in the first place.
There are some clues for more helpful responses in that one, however. It's certainly true that many companies have not done enough to provision for such outages: redundancies, backups and safety generally are expensive and unexciting, and only justify themselves when things go wrong, which mean that they tend to be neglected. But that's precisely the kind of thing we should demand from the companies who look after so much of our lives. Specific requirements from specific people at specific companies, in relation to specific issues.
Robustness and resilience are boring and tiring. But we are living through a period of unprecedented threat, and that's before we have even really faced proper AI-powered cyber attacks and the disruption and outages they could bring. We might never have lived through a time on the web that has felt more perilous – but, at the same time, we are probably less imperilled than we might ever be in the future. There has never been a better time to prepare; practically, promptly, particularly.
0 comentários:
Postar um comentário