Blog by Sumana Harihareswara, Changeset founder

15 Sep 2007, 1:42 a.m.

"How can I provide for this right thing to be always done?"

Hi, reader. I wrote this in 2007 and it's now more than five years old. So it may be very out of date; the world, and I, have changed a lot since I wrote it! I'm keeping this up for historical archive purposes, but the me of today may 100% disagree with what I said then. I rarely edit posts after publishing them, but if I do, I usually leave a note in italics to mark the edit and the reason. If this post is particularly offensive or breaches someone's privacy, please contact me.

I've had some recent success in using spam as a source for recommendations, but more useful by far were Crooked Timber and Ask Metafilter. Speaking of CT, title ideas for your blog posts and a game theory question of sorts.

Thanks to "foobario"'s Ask Metafilter recommendation, I'm currently reading the Project Gutenberg text of Florence Nightingale's On Nursing and it's tremendous. This post's title comes from it. I thought it would be like Martha Ballard's diary, but instead it has a lot in common with Spolsky or my business-ish textbooks. Nightingale focuses on executive energy, attention, and putting the proper processes into place such that patients (employees) have the resources and quiet they need to get better (do their work). Once you get to a certain administrative level, instead of solving problems ad hoc you have to think strategically.

But it's still fun to solve a good puzzle, or to hear a good problem-solution story.

On New Year's Day, 2002, I was working on Douglas Hofstadter's Gödel, Escher, Bach -- Adam P., that's the exercise I mentioned at lunch the other day. I met Zack Weinberg on January 2nd, when we were both living in Berkeley.

Now we're living on opposite coasts. I go to Columbia, where Zack did his undergrad. He lives in/near San Diego, where my sister did hers. OK, maybe that's too forced.

Zack criticized The Atlantic, at least the 2003-era Atlantic Monthly. I've been subscribing for at least a year since I find it good for long trips, so Zack, I'd be very interested in hearing what it was you found unimpressive. I try not to pay too much attention to Hitchens or Flanagan, but Fallows and Bowden seem solid. Am I wrong?

And it's light enough for good not-class reading, a.k.a. cardio-machine reading. Elliptical, stationary bike -- the machines in the Columbia gym have little perches just big enough for paperbacks or magazines, but there's really no way I can take notes during the experience. Some people have beach novels; I have Colson Whitehead's fun and moving John Henry Days and Atlantics from the past ten months. And once I've finished the mag, I can leave it in the mag-swap slots on the wall under the clock, next to the Columbia Spectators and Entertainment Weekly that people bring in. ("So that's what Chuck's about!")

Speaking of Fallows -- James Fallows, former Microsoft employee, current China correspondent -- he had an interesting article in the July/August issue: "China Makes, the World Takes." I can't glibly agree with the cover headline, "Why China's Rise Is Good For Us." Fallows does what the business folks would call a SWOT analysis (Strengths, Weaknesses, Opportunities, and Threats) for the economic relationship between China and the US.

Right now, it's the half-automated processes, like snapping a part onto an electric toothbrush, where Chinese manufacturing excels. At the beginning (design, branding) and end (retail and service) of a product cycle, IP-heavy firms based in first-world countries do great. Manufacturing is a cost center; design and retail are revenue centers. It's classic division of labor to offshore the parts of your business where you have no competitive advantage, can't add value for the customer, and can't make profit for yourself.

That's how US businesses are thinking strategically. And Chinese manufacturers, optimized for cheap prototyping and quick turnaround (hmmm), can do quite well partnering with such firms. But the Chinese government is thinking strategically at a higher level of abstraction. How can China become a revenue center? How can China add value? By building or enticing the institutions that grow intelligent, cosmopolitan executives and entrepreneurs. So the government, being in charge, provides that these things will be done. Schools, Microsoft design labs, whatever you could imagine the frickin' communist dictatorship of the PRC coercing or encouraging. China's not content with being China; China wants to be India too.


15 Sep 2007, 14:30 p.m.

It was long enough ago that I don't remember exactly, but my memory is that every article I read was making generalizations from not enough evidence; in all the cases where I knew something about the subject matter, they were generalizations contradicted by evidence I knew about that they had failed to include. Which naturally led me not to trust any of the other articles.

The article you're citing makes me suspicious but not quite for the same reason. I have this deep-seated mistrust of the argument that one ought to outsource one's cost centers. More often than not, it seems like the only gain from that is short-term profit, and you lose internal flexibility plus you have to trust the outsourcee's interests will remain aligned with your own. (Consider just how screwed the US would be if China stopped lending us money, to pick an example not at all at random.)

Sumana Harihareswara
16 Sep 2007, 16:56 p.m.

Thanks, Zack. Your concerns about the Atlantic make sense, and I'll keep an eye out for that. Where I know something about the topics that Fallows writes about, I don't see him making bonehead errors. I've found myself following bylines more than magazine stables recently.

<br/>I agree that there is a big trust problem in outsourcing (whether it's offshore or domestic), and that many businesses misunderstand their key competitive advantages and thus outsource too much for the sake of the quarterly earnings. (The dictatorship of the shareholder makes more sense now that I know declining share price -} a lower market capitalization for the company -} fewer financial resources and, depending on how you look at it, destroyed value.)

<br/>The thing about cost centers is that the decisions become very different depending on whether or not you already have in-house expertise in the area. If I already have an up-and-running manufacturing facility, and the people there are smart and quick enough that it gives me flexibility and some rough R&D as a bonus, then I'm stupid to throw it away. But if I'm just starting out, or I can see that within the next 3 years I'll have to change my business/products/what have you as radically as if I were starting anew, then in-house facilities might not be a great long-term asset.

<br/>Outsourcing is sometimes a good idea but the fad sucks.