Latent Content

Like This Product to Enter

I was at the dentist yesterday. We had a nice chat in which he told me one of my teeth has a crack and is going to be a problem. Then, while my mouth is occupied with picks and mirrors and tubes and crap that normally happens during a dental exam, he starts talking to me some more. He asks me to recommend him to my coworkers, since he takes their insurance, and his prices are good.

Keep in mind this is the dentist that botched a filling and is going to subsequently cause me to have the tooth extracted if a root canal and crown can’t save it. If he really wants me to leverage social media or tell my social network about him, he might get more than he asks for.

We’ll ignore that. If I like my dentist, I’m going to talk to people about him, or at least offer him up if people are looking for a recommendation. This happens. Instead, what happened is that this discussion solidified my dislike, and I’m now in the market for a new dentist. What’s the first thing I did? I asked my social network for recommendations for a dentist.

Point being, if you’re going to ask me to recommend you to my friends or coworkers, you’ve already lost.

Businesses have been trying to harness the social recommendation network for as long as there have been businesses. What’s weird lately is how artificial the practice has become with technology. Somebody figured out that being liked on facebook or mentioned on twitter increased something or other, so now everybody asks to like them on facebook or retweet something specific on twitter. Maybe they offer a prize. “Just like us on facebook and fill out the entry form to win!”

The problem is, are my friends really endorsing this, or do they just want to win a free trip to Aspen? The gaming of the system makes the system unreliable. So I go back to the same old thing, and ask people what they recommend.

Plumbing

This is a yarn about software engineering¹.

There’s a luxury apartment complex. One of the nice features of the luxury apartment complex is that the master bathroom in each luxury apartment is well-appointed. It features a shower, bath, toilet, two sinks, and a bidet². These apartments lease for many thousand dollars per month, so such features are expected by the renters.

Once every few months, some circumstances combine and violent water hammer breaks out in the bathrooms of one of these apartments. This causes noisy shaking in the walls, sure, but if allowed to continue for more than a couple of seconds, the pipes start breaking, water sprays all inside the walls, and all hell breaks loose. Substantial damage is done to the bathroom and adjoining walls. The tenants have to use their second bathroom (not nearly so luxurious, lacking the bidet, separate shower/tub, and one of the sinks) or stay somewhere else while weeks of repairs and renovation are done. It’s not really clear what combination of devices causes the behavior, but it’s clear that in each situation more than one thing was using water at a time.

A couple of tenants experience this, and when it comes time to re-lease, they decide they’ve had enough and move out. These tenants go out on the Internet and complain about the apartment. The company has to lower the rent and do its best to improve the situation so it can increase occupancy and raise rent again. They send around a memo to all existing and future tenants telling them about the water hammer. They’re advised not to use more than one water-consuming device in their bathrooms at a time.

This seems a perfect solution, but the residents complain. They have two sinks, they should be able to use two sinks, right? So the management office says “well, we’re sure that two sinks should be fine, it’s not like you’re using everything at once. Feel free to use two sinks.”

So residents start using the two sinks (in fact they never stopped in the first place, because nobody pays attention to memos and rules), and sure enough, using two sinks is fine.

But a few residents experience the water hammer, have to survive in their third-world bathrooms for a few weeks, and complain. They’re told that they’re idiots. They should have read the memo and not used more than one thing at once, though they could have used two sinks, because that’s okay.

The residents of course say “yeah, whatever” and decide not to re-lease and we’re back to the same problem.

So, it’s clear the apartment complex has to do something. It calls in its plumber, and the plumber asks what residents were doing when they experience the water hammer. They were “just using a few things” but can’t remember what. The plumber gives each device a try and says they all work and packs up for the day.

The management company sends its maintenance staff to try to reproduce the problem. It’s just six devices, it can’t be that hard to figure out the problem, right? Basic combinatorics, which every maintenance staff knows, tells the management company that they just need to test every combination of on and off for the six water-using devices. 64 tests later, they’ll know what combination of usage causes the problem, right?

Well, that sounds like a lot of work, so the maintenance staff takes some shortcuts and just tries all trios of things together. 20 tests later everything still works and they attribute the numerous incidents in the past to freak accidents that won’t ever happen again.

The plumber confirms this, since they’ve had to re-do a bunch of the plumbing anyway, it’s probably fixed.

Sure enough, another resident experiences the problem, gets pissed, and moves out. It’s time to get to the bottom of this. The shortcuts taken by the maintenance staff were uncovered via an inquisition, and the management staff demands they go through all the combinations, not just the combinations of three devices!

They go through all the combinations and nothing happens.

The plumber is consulted again, and they smartly tell the management company that that result is obvious: It’s a master bathroom, which means two people at most, most likely. Is it really likely they would be using more than three devices at a time? Why would they think that more than three devices at a time cause the problem?

So now the search space is pruned, again, but still nothing has been uncovered.

Somebody (attributed to the management company, but nobody can quite remember) has the bright idea to start thinking the problem through more carefully. There may be six water-consuming devices, but they all consume water in different ways!

After all, the sink, tub, and shower can all adjust their temperatures, which means a seemingly infinite series of output water temperatures and input hot and cold water consumption can be experienced.

Perhaps if they use enough maintenance staff and approximate various gradations of temperature along the way, they can reproduce the circumstances. The maintenance staff dives into the problem head-first.

Hours elapse, but still no water hammer. Back to the plumber, who again reminds them that they’ve over-simplifying things. The sink faucets have pressure adjustment as well as temperature. The shower head has a lever to adjust the spray which adjusts the pressure. They get into topics of differential pressure between the hot and cold water lines, and how adjusting the shower head’s spray may ultimately adjust not just the amount, but also the ratio of each type of water flowing through the shower head.

Bewildered, but satisfied, the conference call between the management company, maintenance staff, and plumber is about to end. The plumber, however, is not finished!

The plumbed starts explaining how bad the situation is. The pipes leading to the bathroom change temperature over time. The hot water heater doesn’t deliver water consistently at the same temperature. The cold water isn’t always the same temperature. The ambient temperature varies in both the apartment and the walls.

The pressure in the building varies between negative and positive pressure versus ambient. There are other water-consuming appliances in the apartment, including another bathroom and a half, dish washer, several sinks, washing machines, humidifiers for the central air, and so forth.

There are other apartments in the building and other buildings in the neighborhood, and there are pressure boosters on every 4 floors to keep the water pressure up, which also means that the pressure varies between floors. Seals and valves degrade in performance and behavior over time, such that the same handle input doesn’t produce the same result. Toilet ballasts and seals change over time, and some of the newer toilets have two different flush settings. Further they have very periodic and definite on/off behavior, whereas some faucets and things like bidets aren’t necessarily turned on full blast right away. Different shower and faucet heads have different classes of flow restriction devices depending on when they were manufactured.

Also, sometimes people take longer showers or decide to take a bath instead, and …

You get the idea. Now what?

Chicken Breast Mezcal

Check this out.

First off, I’m going to assume some basic familiarity with mezcal. Refresher course: maguey pinas are harvested and trimmed to just their hearts. The hearts are dumped in the ground for a few days and cooked over a wood fire. The roasted piñas are then ground up on a stone wheel, mixed with water, and then they ferment naturally.

The resulting slurry is distilled over hand-fed fires and potentially aged in oak. For more mass-produced mezcals things are a bit different, but those mezcals are not interesting, so I’ll ignore them.

This brings us to Pechuga. After the mezcal is distilled, it is combined with wild mountain apples, plums, plaintains, pineapples, almonds, and uncooked rice.

Don’t be fooled, this isn’t a crema de mezcal being made. This is far cooler than that. This infusion is distilled again, but this time a raw chicken breast is suspended in the still.

How cool is that? Chicken breast mezcal. This is happening. Stay tuned.

Browsing For Books

Krugman was interviewed in the Boston Globe. Everybody’s making a big deal out of how he’s a hard sci-fi buff. To be fair, that’s pretty damn cool. What I latched onto, however, was something I’ve been frustrated by before:

The problem with digital books is that you can always find what you are looking for but you need to go to a bookstore to find what you weren’t looking for.

This is one of those obvious things that seems to come up every time somebody muses over the problem of e-readers and such. Amazon has its Kindle store, and there are a myriad of “suggest a book” websites that leverage varying degrees of effort.

But, I’m reminded of Books Inc. in Burlingame. For years, when I’ve gone to the bay area, I’ve tried to stop here when I’ve had time. It’s an independent chain, but it’s hard to realize that when you visit just one store. It feels like it’s been there forever. It’s not particularly fancy, wedged into a strip mall with sufficient, but not particularly attractive, furnishings. The selection is relatively small, with a few shelves dedicated to each genre.

What I love about this book store is that I can browse around and easily find something completely unexpected. I can pick it up, flip through it, and go on from there. Sometimes I browse within genres I know, sometimes I find myself picking up strange offerings from biography and history sections.

What’s impressive to me about the store is that it is a well-curated selection of books that are more frequently interesting than not. There are some staff picks with hand-written descriptions (I still love this custom), but one need not rely on those to find something new.

In the end, Books Inc. is not a bookstore you visit to find what you’re looking for. I’ve never, in fact, found a book that I was looking for there.

I’ve certainly discovered new books via the Internet and various online tools and storefronts. Some of them have been enjoyable. Somehow the experience never feels the same. I can’t tell if there’s a genuine difference or if it’s some foolish romantic notion.

Surely there are some differences. A suggestion website or commerce system has a massive catalog, rather than a constrained selection of titles. When confronted with a large inventory, there’s a tendency for obsessive min-maxing of sorts; star ratings and reviews quickly screen out things on the periphery that might be interesting. It’s easy to find things that are generally appealing to the general target market, but finding things that are personally resonant becomes a chore. It’s the problem of Oprah’s book club and similar efforts for me. The books are meaningful, deep, poignant, and whatnot, but they feel like they’re deep and meaningful for a population.

I take it as a given that brick and mortar book stores will continue to die in their current forms. The tide of electronic commerce and electronic books is going to relegate what we thought of as book stores to the same fate as music stores. There’s always things like used book stores, but those are largely full of junk that nobody wants. They’re good for decorating shelves with books, but the signal to noise ratio is horrible. One’s never (seldom?) going to find a used book store that says “no, I don’t want to buy that, it’s not interesting enough to be in my collection.”

I want little well-curated book stores to survive, but I don’t know if they will. If they don’t, I’m not really sure what fills that hole.

I can’t help but feel there’s something here similar to channel flipping on television. I haven’t had cable in half a decade now; I watch television on demand with what I purchase when I want to. I don’t discover anything randomly. The things I venture out and try are critically acclaimed or create a buzz among friends. Television never really had a phase where there was a well-curated selection of shows or whatever that I can be nostalgic about. There’s nothing there I miss. If a generation grows up without the well curated small bookstore for books, it probably will not miss the concept either. So perhaps I’ve answered my question, but that’s a little sad.

Metrics

This is a bit of yarn.

A company was evaluating upgrading to a new laptop. The old laptop was generations behind on architecture and storage. The upgrade should have been a formality.

The new laptop was thicker. It was heavier. That was sort of weird.

There were a lot of complaints about the screens. The aspect ratio changed from 16:10 to 16:9 (“it’s just the way the industry is going”)¹, so some complaints were expected.

Upon seeing the screen, it became clear that the problem was not just aspect ratio. The new screen had more pixels. Unfortunately, there were fewer vertical pixels. Vertical pixels are life for a software engineer. The new screen was also an inch shorter and featured a smaller pixel pitch. So there was less content, in less room, and it was harder to see. A thick bezel was added to eat up all of the space freed up from the previous design.

How did this happen?

At the tail end of the life cycle for the old laptop, the vendor had a focus group with its biggest clients. Makes sense. The vendor wanted to make money. Getting real world feedback would help it retain clients and attract new ones.

Chief complaints in the focus group were that the battery life was poor and that the device felt flimsy and cheap. Both of these complaints were valid. On the old laptop, the screen in particular felt about as sturdy as a shrink-wrapped legal pad.

The vendor listened to this feedback. It set specific goals for the next laptop. Dramatic increases were required in battery runtime and chassis rigidity.

More efficient components were procured, but battery weight was required to achieve the runtime goal.

Reducing chassis flex was accomplished by adding more internal and external structure (read: plastic). The smaller screen helped with this, since it freed up some room and decreased the weak flexible cross section. The laptop could now be carried by the opened screen² with confidence.

The effort was a success. The new laptop hit its targets.

Except there was one little problem. Nobody was asking for a bigger laptop with a smaller screen.