字号·· | 护眼
theregister

沮丧的M365客户是时候发起一场“分布式拒绝满意”行动了Time for frustrated M365 customers to stage a 'distributed denial of satisfaction' action

点「原文对照」整页切到原文,或双击某段只看那段的原文。

2000年代初,我曾短暂协助微软澳大利亚分公司的公关工作。当时IE 6和Windows漏洞频出,简直是给黑客铺好了红地毯,补丁如潮水般涌来,管理员根本应接不暇。微软对此回应称,安全是一项共同责任,涉及人员、流程和技术。四分之一世纪过去了,微软的商业惯例依然让客户不得不勉强分担其失误的成本——最近一次便是上周大规模的M365服务中断。

I briefly helped with Microsoft Australia’s PR efforts in the early 2000s, when IE 6 and Windows were so bug-prone they essentially represented a welcome mat for hackers and required a deluge of patches admins couldn’t handle. Microsoft responded to that mess with a message about security being a shared responsibility involving people, process, and technology. A quarter of a century later, Microsoft’s business practices still make its customers reluctantly share the cost of its failings – most recently last week’s massive M365 outage.

为抗议这种乱局,微软客户应在该软件巨头用于处理未达服务等级协议(SLA)赔偿申请的页面上,制造一场新的“云端崩溃”。您没看错:“申请”。微软发生服务中断时,客户必须通过申请才能获得应得的服务抵扣额,方法是找到与每起事故相关的“中断ID”。申请页面深藏在Microsoft 365管理中心深处,而中断ID则记录在别处。

To protest that mess, Microsoft customers should create a new cloudy meltdown – on the pages the software giant uses to process applications for compensation after it fails to meet service level agreements. Yes, you read that right: “applications.”When Microsoft experiences an outage, customers must apply to get the service credits they’re entitled by finding the “Outage ID” related to each incident. The application page is buried deep in the Microsoft 365 Admin Center and the Outage ID is recorded elsewhere.

微软似乎不提供批量上传功能,因此管理大量M365部署的渠道合作伙伴面临额外工作。这令人愤慨,因为云服务的隐含承诺是以更少精力换取更多价值。然而微软却要求用户为其错误买单,还得额外劳动寻求补偿。云服务也为投资者带来丰厚利润。这大概就是为什么微软——明明能定期可靠地向客户开账单,因而清楚知晓他们使用哪些服务——却迟迟不愿将正常运行时间监测与计费系统打通,以便在服务中断后自动向客户发放抵扣额。

Microsoft appears not to offer a facility for bulk uploads, so the channel partners who manage so many M365 implementations have extra work to do. That’s outrageous because the implicit promise of the cloud is that it delivers more value for less effort. Yet Microsoft requires users to do extra work to seek redress for its mistakes. Cloud also delivers fat profits for investors. Which is probably why Microsoft, which can clearly bill its customers regularly and reliably and therefore knows which services they use, hasn’t done the work to link its uptime meters and billing systems so it can automate crediting customers after outages.

我知道微软认为AI很擅长这类集成工作。或许其大幅缩减的员工队伍将需要AI来完成这项任务。

I understand Microsoft thinks AI is quite good at that sort of integration. Maybe its enormously reduced workforce will need it to get the job done.

但出于体面,微软不应靠着胡思乱想来解决问题,而是应该迅速决定,让上周故障中受影响的所有用户自动获得服务赔偿,而不是要求他们申请自己应得的补偿。如果微软不采取主动,这次故障的严重程度将成为一个契机,让其客户向这家软件巨头表明它必须做得更好——或许可以通过提交大量的赔偿申请,让负责雷德蒙德系统和财务的管理人员不得不注意到激增的业务量。可以把这看作是一次分布式拒绝满意攻击,它表明微软无法用一句俏皮的道歉和一篇博客文章来解决这次特定的故障。®

But rather than vibe-coding its way out of this, decency demands Microsoft quickly decide that all users impacted by last week’s automatically receive service credits, rather than asking them apply for their entitlements. If Microsoft doesn’t get proactive, the magnitude of this outage represents a moment in which its customers can show the software giant it must do better – perhaps by filing so many applications for credit that managers who oversee Redmond’s systems and finances can’t help but notice a spike of activity. Thinks of this as a distributed denial of satisfaction attack signaling that Microsoft can't solve this particular outage with a glib apology and a blog post. ®

阅读媒体原文 ↗