Within any start-up there are always differing opinions on the level of reliability required. My first real job in IT was with the Dutch division of the Chase Mannhattan bank, they made it pretty clear upon hiring me that they did not mind hiring me into a responsible position as young as I was but that my first fuck-up would also be my last. I'm not sure if they would actually follow through on that promise but it certainly made me a bit more wary and more responsible. The fact that I ended up leaving of my own accord means that their threat probably had its intended effect.
When you're in the payment chain, whether as an IPSP, a bank or something else that directly affects peoples income and ability to function in society you have no excuse other than 'meteor fragments took out all three of our datacenters'.
Stuff is up, check-totals are at '0', transactions are never lost. 24 hours, 365 (of 366) days per year.
If you can't do that don't bother playing.
The bank I'm with (Rabobank in nl) lost a lot of respect from me (after building that up for 2 decades) when they blamed a bunch of down time on 'hackers'. What bugged me even more than the fact that they were down was the fact that they wanted to blame another party for being down.
The one thing I'll leave a little bit of room for is capacity issues during the holiday shopping spree, but even that is pretty bad form and indicative of very bad planning.
Judging by the post the error lies with Simple's partner, unfortunately for Simple customers will likely not care about that at all. To them their simple card doesn't work when they expect it to work.
I hope for simple they'll be able to fix this very quickly.
It feels a bit like when the general public got used to see computer fails, and "the computer is never wrong" mental image became a "it crashes from times to times, you just have to reboot" one.
I didn't say that those things don't happen. I said that they are not acceptable. The fact that others fuck up does not give you an excuse to fuck up.
These are exactly the kind of discussions that I was hinting at above.
The fact that they're being proactive about their status updates, and transparently honest, helps a lot, too.
A relationship with a bank is a lot like a relationship with a doctor: bedside manner is sometimes the most important element to patient (customer) satisfaction - sometimes even surpassing outcomes, as long as it's not too serious.
With major banks like BofA, Citibank, and others, I've had…
• credit cards often disabled for fraud-alert false alarms
• check deposits sometimes subject to mysterious fund-availability delays
• individual ATM clusters or whole ATM networks down for short periods
• account information and ATM withdrawals sometimes unavailable during 'system maintenance' hours (often early Sunday AM)
• a check erroneously bounced by a major bank (BofA) when both available-by-rule funds and enrolled 'overdraft protection' would have each been individually able to cover the check amount
• my ATM card unusable while traveling to the exotic third-world locale of Ottawa, Canada due to some sort of more-than-one-day network security issue
Stuff happens, to banks also. Cut Simple some slack.
Stuff happens to banks and then people switch to other banks, especially when you're a new player.
If my bank did this to me I'd be gone instantly, I almost did just over their inability to keep telebanking working through a denial of service attack, and then blaming it on the hackers. All my cards (credit, debit across several companies and private accounts) link back to the same bank. If that bank does not do its job I can't do mine.
ATM and pos worked fine during that period, if it had not I'd be gone for sure.
If banking is that important to you, then perhaps you ought to consider redundant providers? I mean, that's what you do with connectivity. you don't rely on one provider with a good sla; no, you get two providers.
They deal I have with my bank is simple: they do their work, they get to keep transaction fees, screw me on the interbank interest rates and delay foreign transactions just long enough that they can go play with my cash. In return I expect them to deliver one thing: utter reliability.
If it gets to the point that there is no single bank that gives me the feeling that they can do just that I will indeed have to split assets across several banks. (there are other good reasons to do that, such as the maximum insured balance per account holder).
But the costs of doing so are what is holding me from that.
I wonder if it's related to the leap second bug discovered in July. Apparently, a some popular NTP servers incorrectly sent out leap seconds at midnight. Even if it's not that specific issue, I bet the root cause is a time-keeping bug.
EDIT: "The Bancorp Bank", the bank Simple farms out the actual banking to, has been around since 2000. Simple _might_ have an excuse for screwing up like this if this was all their own system, but an established bank doesn't.
Edit: it's exactly like trying to reinvent Twitter by consuming their APIs, and we've all seen how that goes
Bancorp is a closer analog to EC2; they let Simple get off the ground quickly without having to build and harden their own banking systems and transaction networks. Also like EC2, Simple gets to take the heat equally if Bancorp screws up.
That said, this is pretty bad, but if Simple had built their own network and left a glaring security hole open, they'd be utter toast.
No doubt you already see where I'm going with this.
They also don't deserve any special 'slack' just because they use other companies services. You wouldn't accept that excuse from Bank of America.
Allowed: scheduled downtime announced ahead of time for reporting, unscheduled downtime with localized effect due to 'acts of god'. Not allowed: customers can't access their funds when they should be able to.
When you disrupt the banking industry you especially should be doing at least as good as the established parties do in the reliability department.
Let's hope they have it fixed very quickly or it may well kill Simple.
Imagine being in a bar somewhere right now and having just bought a round finding out that you can't pay for it. You'll hear about that one for years.
Easy to say but I think it is unreasonable to expect a company that is couple years old to have as much reliability as ones who have had decades to figure it out.
I think the Simple customers have every right be absolutely pissed. For some, they may even realize that the downside of going with Simple isn't justified by the upside. Hopefully, for the majority of Simple users the upside will make the rare downtime bearable.
My view would change if this becomes a regular occurrence with Simple. But that does not seem to be the case.
Payment systems have an impact that is unlike any other system.
It's very well possible that lives depend on your system working. You may never know about it but you should definitely build as though they do.
You can't build financial software the same way that you build some CRUD site or the latest social fad.
No one is arguing that you should be able to do that. And it is kind of insulting to imply that that is how the folks at Simple look at development without some source.
Systems break for many reasons, and you can't simply assume the cause to be a social fad attitude of the dev team.
If Simple is used the way they intend you to use them those scenarios are playing out right now. You really can't shift the blame for that onto those that believed the Simple proposition and acted as though simple was going to deliver.
Ordinary everyday people do not do single-point-of-failure analysis on their wallet, they simply expect it to work.
I don't see any marketing claim on the Simple site that they'll have fewer outages than the Rabobanks/BofAs of the world. (Where do you see that claim, or actual delivered level of service, from any actual existing bank?)
'get ready to leave your bank'
'unmatched security and support'
'replace your bank'
'use the simple card for all your purchases'
Those set the stage for some pretty high expectations, they make it plain simple expects to be your sole provider.
Rabobank fucked up once in 20 years, and then only on a peripheral part of their service (ATM and point-of-sale still worked). If they had dropped the ball on either I'd be out of there for sure.
Almost four hours isn't brief. And I've never had a point-of-sale or ATM outage like that, ever. (other than the power blackout a few years back in the North-East, but not being able to pay for stuff was the least of your problems back then. By the way, our gas station was the only one running for 75 miles around because I had actually thought through how we'd operate in a situation like that).
It looks as if we hold our payment service providers to different standards. I expect mine to be there when I need them, you're ok with whatever your definition of 'brief outages' is.
It makes me wonder what it would take for you to start wondering if you've got your money parked with the right institution.
For me not being able to get at it when I need it is the one thing that would do it.
Anyway, great for 'simple' that they have it resolved, this will likely slow their adoption for a bit (though sync is on the ball in this thread and turning lemons into lemonade) and it will hopefully cause some serious introspection inside simple to make sure stuff like this will never happen again.
And maybe I'm wrong and old fashioned in my attitude towards payment services and we should just cut all those banks some slack and accept periodic outages. But my mid 80's bosses would fire my ass very quickly if I displayed that attitude on the job and I think it is too far ingrained in me to let go of.
Doesn't worry me in the least.
For example, in surrounding tweets @Skroob jokes about paying with Bitcoin, mentions his dining party has other options, and then despite some disappointment concludes with sympathy for @Simplify. He further says that Simple's "company policy of ignoring the s* out of HN, especially during outages" is something that "just makes me love you guys more".
So: while non-customer HNers with crazy-high standards whinge, real customers can laugh about the situation (and even laugh about HN's negativity-bias).
My current setup is a cash-back credit card, with the bills for it paid from a high-yield savings account. I net close to 2% cash back on the card, and around .8% APR on the savings account, and pay exactly $0 in fees. Based on the language on Simple's site, I'm guessing the APR on their savings is closer to .25% or so, and they definitely don't offer cash back on the use of their card.
Sure, the tools they offer are nice, but my credit card provider also has pretty nice spending analysis tools (that don't require an iOS device), and lots of other providers are adding or improving spending tools as well.
What am I missing?
If my bank went down losing access to my money, I would be naked, I don't always have cash and money is our oxygen is this society, a absolute essential. Hopefully the downtime for Simple will be short, but if they had 100 million people using them as their main bank and this happened, you could expect mass exodus the next day.
The only time I have lost access to my money was a few years ago when VISA or MASTERCARD went down (or both), with them you don't have a choice and everyone lost access to their money, but if ever just one bank went down it wouldn't turn out good at all.