|Wednesday, October 14, 2009
10:05 - Is this from global warming too?
The following are measured wind gusts in mph from this recent early
Point Reyes lighthouse... ... ... ... 63
Woodacre RAWS... ... ... ... ... ... ... 52
Middle Peak RAWS... ... ... ... ... ... 53
Big Rock RAWS... ... ... ... ... ... ... 52
San Francisco Airport... ... ... ... . 56
San Francisco State University... . 58
Twin Peaks in San Francisco... ... . 75
Golden Gate Bridge... ... ... ... ... . 63
Point Blunt Angel Island... ... ... . 77
las Trampas RAWS... ... ... ... ... ... 60
Oakland North RAWS... ... ... ... ... . 58
Half Moon Bay... ... ... ... ... ... ... 49
Los Gatos RAWS... ... ... ... ... ... .. 87
Calaveras Road RAWS... ... ... ... ... 66
Spring Valley RAWS... ... ... ... ... . 62
Ben Lomond RAWS... ... ... ... ... ... . 46
Cabrillo College... ... ... ... ... ... 51
Pinnacles RAWS... ... ... ... ... ... .. 52
Monterey Airport... ... ... ... ... ... 51
And record rainfall for the last few days, too. What the hell?
Should I be glad I'm out east now where all we get is blizzards and protracted sub-freezing periods that require you to buy special cars?
UPDATE: And now it's snowing.
UPDATE: And it's frickin' sticking. Not a lot, and only in certain colder areas, but... damn.
I know. I'll dress up as Winter for Halloween!
|Monday, October 12, 2009
13:05 - Clown Computing
I can honestly say I didn't see this one coming, but maybe I should have.
Currently the rumor with the most weight is as follows: Microsoft was upgrading their SAN ... and had hired Hitachi to come in and do it for them. Typically in an upgrade like this, you are expected to make backups of your SAN before the upgrade happens. Microsoft failed to make these backups for some reason. ... Microsoft should know better. So Hitachi worked on upgrading the SAN and something went wrong, resulting in it’s destruction.
This is perhaps what unnerves me most about Microsoft: it would be one thing if it were a company whose sluggishness and ossification were attributable to an overabundance of procedure and paperwork, like a government agency or a company like Ma Bell or GM. At least when an organization of that much size fails to serve its customers in a speedy and satisfying way, it has the excuse that at least it was well documented, and at least proper channels were followed. At least everyone signed off on all the forms in triplicate. Maybe it could have been done faster, but would you rather it be done fast, or done right?
But apparently Microsoft still has a seat-of-the-pants mentality in some ways, still thinking of itself like a nimble Silicon Valley startup who has no time to take onerous precautions when there's good money on the table. We can't draw too many conclusions just from the rumors that are out at this stage, but it sounds like the case is fairly open-and-shut; not having a backup during a server rebuild is such an easy trap to fall into—and so well-known as such to anyone in the tech world—that there's no implausibility whatsoever to the story. Much as we might like to imagine that a company like Microsoft, more than anyone else, could be trusted at least to get the I-dotting and T-crossing parts of the business right even if they can't execute on anything else, it doesn't appear that that's necessarily the case.
It apparently will serve as a lesson to all companies destined for the same stature that treating themselves like small companies won't cut it. Hell, even small companies can't treat themselves like small companies once critical end-user data is involved. Data integrity and high availability ought to get built into any company's infrastructure from the outset, even long before the company has any need for it, because by the time it does have the need for it, it won't have the time to buy it.
People often ask each other why we ought to trust Apple and Amazon.com with their credit card numbers. Well, because if Apple or Amazon.com were to suffer a compromise of their customer data, that would be a brand-killing event. Same goes for banks: the one unforgivable sin for a bank, the one that would mean they might as well roll up the signs and board up the windows, would be to have its internal network infiltrated and/or its customers' information stolen or lost. These companies have even more incentive to protect your critical information than you do. Never mind wondering whether So-And-So Company is some kind of criminal front and will go selling your info to the highest bidder; it's death even if this happens accidentally. This is why it's often so hard for an upstart company to move from the world of obscurity to stardom: customers don't know whether they can trust it until they know everyone else is trusting it. But once they've reached that level of trust and public awareness, a company will find it pays off to a degree that more than justifies the hard initial slog. Mindshare is a powerful force; customers will gravitate to a name they've heard of even if it's not promising more than a small obscure upstart does.
That's why Microsoft continues to exist, is all I can figure. It seems like every story that comes out of Redmond these days is a tale of woe; and yet Microsoft hasn't imploded or lost enough of the public's trust to cause their key revenue streams to dry up. There's still nobody providing the solutions they do, in many important areas, and as long as that's true they'll always be around; but even with energetic next-generation competition nipping at their heels, they'll still keep trundling along.
Unless, of course, this Danger fiasco turns out to be one of those brand-killing events that people talk about.
That'd be a hell of a drama to watch unfold. And unless Microsoft is able to make some enormous changes really quickly, it'll just be a matter of time—and one or two more missteps like this one—before it happens.