Hacker News Reader

Comment by maukdaddy | Hacker News Reader

Cute doesn't cut it when people are relying on your business. I sure hope no SaaS vendors take this approach.

The major downtime sucked, and the minor hiccups still continue to suck. But I think we also need to realise it's free. It doesn't excuse the lack of communication, but lofty expectations of uptime for a service that's offered freely seem asinine.

On a site like Hacker News, we judge Tumblr both from a consumer point of view, but also from a start-up point of view. And from the latter, this instability is a really big problem.

and beyond that, if you set the expectation that you're giving a pro service, regardless of the price, then you should probably do what you taught people to expect...especially with millions on funding.

Just because you're giving away a service for free doesn't mean you should take QoS lightly. Nobody would say that complaints about google search going down were asinine. Same for local broadcast tv service. Or twitter. Or youtube.

If your business is providing hosting for content or providing a service that is expected to be ubiquitously available then QoS is a significant part of your business whether your service is nominally free or not.

Edit: Whether your revenue comes from users or from advertisers the reliability and quality of your service matters. If your userbase decides to abandon you then you are screwed regardless of whether they are your customers or your product.

Downtime is ok, shit happens. to everyone. But availability issues which are going on for 6 weeks now (still major IMO), is it? I can't think of any technical problem that would take 6 weeks to solve. And being tumblr, attracting the best system admins, with all clouds resources available today..

It's like Tumblr are trying to prove some kind of point with the insane downtime.

Reply on news.ycombinator.com