Archive for the ‘Notes’ Category

There should be two kinds of insurance.

Friday, November 15th, 2013

There should be two kinds of insurance: insurance that covers random acts of god and insurance that covers the expense of protection money.

If you own a car and it’s parked in your driveway, and there’s a storm, the wind blows, and knocks the tree onto your car and damages it, this is a random act of kindness on god’s part, as you probably wanted a new car anyway, or at least an excuse to buy a new car.

This kind of loss should be covered by one kind of insurance, it would insure against accidents where there was no intentional effort on anybody’s part to cause the loss, it was just a random sequence of occurrences that ended in your losing out.

The other kind of insurance would be for protection money. Consider the guy who, instead of wrecking your car by having a tree fall on it, steals your car.

In this case, with traditional insurance, you would be compensated (poorly) for the loss of your car. But unlike an accident, somebody is profiting from the loss. When a tree falls on your car, I suppose there’s some satisfaction god gets by laughing at your loss because he knocked a tree on your car, but there’s really not much you can do about it, and god can’t spend that. There’s no financial gain on anybody’s part.

But when somebody steals your car, you are out the car, the insurance company is out the lesser of the fair market value of your car, or whatever they can get away with paying you, whichever is less. But the guy who stole your car gets a free car out of it.

If you remove all the middle layers of the transaction what’s actually going on is that people who pay insurance premiums are giving the thief a pile of money.

Okay so this isn’t exactly protection money, but it’s close. God can’t be stopped from knocking down trees unexpectedly, but people don’t have to steal cars and cause this loss.

It seems to me these are two very different problems that should be treated differently.

The cost of theft insurance (the random acts of paying thieves) should be separated from the cost of random acts of damage. So if you have a strong belief in the wonderfulness of humanity, and don’t think your car will ever be stolen, you shouldn’t have to pay the theft insurance. And indeed, you don’t have to get theft insurance on your car, but because all insurance companies lump them together, it seems to me people who don’t want to pay for theft insurance are paying via the rest of
their premium.

There are obviously grey areas. If a hurricane blows down a house that’s on beach-front property, and you rebuild, and then another hurricane blows that one down, and then you build AGAIN, it’s hard to say you had no intention of collecting insurance money.

I don’t think that would fall under act of god. That would more be an act of stupidity or an act of selfishness, because you’re building the house again at other insurance-premium-paying-people’s expense.

But it seems to me, there’s something unfair about the way all kinds of loss, whatever their cause is lumped together under the same insurance umbrella, when in fact some types of loss are man made and possibly can be prevented.

 

Sharkbite

Sunday, November 3rd, 2013

So I had to do a plumbing job at my house where I needed to run pipe from my basement to a crawlspace that would not accommodate 10 foot lengths of copper, and I didn’t want to cut a 4 foot length feed it down the hole then solder another 4 foot length, that would take forever. So it was time to try pex.
Pex is basically a really big hula hoop that isn’t in a loop. And you run water through it.
It’s flexible and can go places copper can’t.
It’s got lots of downsides, but for small things it’s handy.

But that’s not where the magic is. The sharkbite connector thing is the magic. A copper fitting elbow or straight connector costs like 30 cents or something like that, whereas a sharkbite connector costs $7. Interestingly the elbows cost $7, the straight connectors cost $9. You are penalized for being cheap and not buying a longer piece of pex.

But it works. It really does. No solder, no torches, no worrying about catching the insulation on fire. You just snap it together and you’re done. Takes less than a minute. You spend all your time hanging the pex, but the connection itself is trivial.

The best part is that these connectors work on pex and copper so you can make a junction between old and new plumbing with these things.

And if you weren’t convinced by me saying nice things (which I never ever do) go look on all the plumber’s boards and read about how they rag on it left and right. They’re angry and afraid, and they have every reason to be.

There will always be a need for plumbing and there will always be plumbers, but there’s no question sharkbite takes the bottom out from under plumbers for the easy work.

Plumbing is one of those technologies that hasn’t progressed in decades because everybody making money liked the status quo, but these silly little things are game changers for do it yourself work.

Any idiot can do plumbing with these. The only thing you can do wrong, is not push it on all the way. There’s nothing to screw up.

 

 

Reset buttons

Wednesday, October 9th, 2013

My computer has a reset button.

My DVD player has a reset button.

Many computerized toys and fancy electronic gizmos come with reset buttons nowadays.

It’s a sorry state of affairs, but it’s a universal acknowledgement that computer software is getting complex beyond its creator’s ability to manage, and it’s cheaper to create a reset button than to write nearly perfect software and expect to get your product to market at or before the time your competitor does.

Since the era of good software development is over, and competition for market share rules the roost, it’s almost understandable that companies have to sacrifice software quality for ship-to-client-before-we-run-out-of-money.

And I have only found one thing worse than having a reset button.

And that is not having a reset button.

Chrome comes to mind.

Large software packages that don’t include a reset buttons or restart functionality at this point, are just snobby.

Eclipse comes with a restart button built in. They know they’re not perfect, and they know their audience.

In firefox you can add a restart button as an extension, but last time I checked, there’s no restart extension for chrome, and it certainly not built in. Chrome is so wonderful you never need to restart it.

There are workarounds, you can change a config setting and change it back and an option to restart will appear, but there’s no menu option.

Presumably google believes that their software doesn’t have problems, or leak memory and doesn’t need an easy way to restart it.

Microsoft word doesn’t have a restart button, but it does have an incredible amount of crash recovery built into it, which seems to me is almost as good. Certainly by doing that microsoft is at least admitting that they know there’s a problem and have gone to great lengths to mitigate the misery for the user.

Of course it would be a better place if people had the time to write good, problem free software in the first place, but I guess that’s a dream for a bygone era.

 

Washing machines.

Tuesday, September 24th, 2013

My washing machine has a light on it with the label “Add a garment” next to it.

This light goes on for a few minutes after I’ve started running a load of laundry. I don’t actually read the instructions that come with appliances like washing machines, because I am a man, and it’s in the man-code that you don’t do that.

But presumably, the idea is that for a few minutes after the washing machine starts doing its thing, you still have the opportunity to add something to the mix and you won’t miss out on any serious or meaningful washing.

So sometimes what happens is that I start a load of laundry and as I’m walking around the house a few minutes later, I see some article of clothing, a garment if you will, lying on the floor where my son deposited it when he took it off in a whim of clotheslessness.

I immediately, like the rush of a typhoon, grab said garment and run it to the washing machine in hopes that the happy little “add a garment” light is still on. Often it is, I push the pause button, open the door, add my garment, close the door, push the start button and voila everything gets washed, including my newly added garment. So now I’m thinking, if my newly added garment is going to get just as much washing as the original contents of the load I put in, what was it doing between when I initially hit start and when I interrupted the machine?

It couldn’t have been all that useful because any newly added garments presumably get just as much good washing action done when added later. Arguably, you could walk up to your washing machine and press start without putting any clothes in, then come back a few minutes later before the ‘add a garment’ light has gone off, and hit pause, add an entire load of laundry and let it continue and have everything get washed.

Is it serious? Can you only add “A” single garment? Will it get mad if you dump an entire load in? What if you add a single garment multiple times, one at a time, instead of adding a bunch all at once. I think you can get around the 1-at-a-time implied restriction this way.

So this brings me back to … what is it actually doing? And does it really need to do it?

Is it just wasting time putzing around for a few minutes with that light on because it knows humans tend to first find dirty clothing after the wash has started? Maybe they should have an option for those of us who don’t have kids and are sure footed to SKIP the warm-up pre-game show and get right down to the serious washing right away. This would save me many minutes of wash time I expect, which would be far more valuable to me than giving me the option to add clothing later.

Which, lets face it, I’m going to do anyway whether that stupid little light is on or not. Fancy options be damned.

I am after all, a card carrying member of the man club.

First world problems, I know.

 

Investment diversification.

Tuesday, September 24th, 2013

I had this novel thought the other day.
For as long as I can remember, I’ve heard nothing but “diversify, diversify, diversify” from every investment advertising TV commercial and radio ad I’ve ever heard. That seems to be the conventional wisdom, spread your money out wide and thin so if anything bad happens to one or more investments, your personal hit will be small.

And there are many historic examples of bad calls and poorly placed investments where people have suffered miserably, because they didn’t diversify.

Enron comes quickly to mind. Here’s, (or rather, there’s) a company that managed to convince a way too large portion of its employee population to invest their retirement money in the company’s stock.

Breaks the first 3 rules of investing: 1) Diversify, 2) Diversify, and 3) Diversify. (Rule number 4 by the way, applies to all facets of life: if anything weird happens, hit the clutch.)

So when enron quickly tanked, so did the retirements of many of its employees. Somebody at enron should have known better, particularly somebody in HR, but that’s history, and we should learn from it.

Enron makes a particularly interesting example because it highlights a similar non-diversification problem. Pensions. Now if you start working at a job after the 2008-2012 era, you’re likely to not get a pension anyway, so this won’t apply to too many people anymore, but pensions create a significant diversification problem for long term employees.

Imagine you start working at GM in the 60’s. There you are, young and in your 20’s and you start making widgets on the assembly line at GM. You power through your best years making the best damn widgets known to man, slowly accruing equity in a fantastic pension plan.

When you retire, you’re going to be set for the rest of your life.

Then 2008 comes along, near retirement time, and all of a sudden, GM finds itself with lots of really expensive unfunded pension obligations because the children of the greatest generation chose to shove off those problems to the future, and the future has finally arrived.

I’m far from the first to preach this particular song, but what I haven’t heard anybody suggest is that maybe everybody should be diversifying their pensions. I don’t remember the details of what happened to GM, quick reading tells me they’re trying to buy everybody out, but haven’t actually screwed anybody over yet, but if you were that GM employee in the 60’s maybe sometime in the 70’s you should have gone to work for ford.

They’re in the same boat as GM, but you can still diversify and limit your risk.

Then in the 80’s you could have worked for honda. in the 90’s you could have moved to toyota, and in the 2000’s moved to hyundai. I don’t know if any or all of those companies offer pension plans, but the idea is that if you stay at one company forever, and start banking on that one single point of failure for your retirement, you may be waiting decades to find somebody is going to screw you over at the last minute.

It’s an interesting argument for why you shouldn’t stay at one job for a long time. Diversify your investments, including your pension.

 

Hydraulic Door Closers

Wednesday, September 11th, 2013

Ahh the innocent, ubiquitous self closing door.
Doors keep out the cold, they provide privacy, they can be a form of art even. But most importantly, they create social angst.
But only among strangers.
If you’re hanging out with a friend or relative and you’re walking towards a door, one of the two of you will grab for the door first and hold it for the other. It is a common courtesy and a reasonable assumption that you want to keep walking with the person.
I seem to recall some cell phone carrier that couldn’t work out how to reliably transfer ongoing calls from one cell tower to the next if you were on the phone in a moving car.
They called that a feature, because having disconnected your call, it gave you the opportunity to not call the person back.
But assuming you both want to walk through the door, you manage it because you’re walking together and in close proximity to each other when you reach the door.

But what if you’re in a public area and you’re walking towards a door by yourself. You open the door, go through it, then notice that there’s somebody behind you. But they’re not right behind you, maybe they’re 20-30 feet behind you.
So you hold the door open for them because it’s rude not to. Letting a door close in somebody’s face is inherently seen as a slight when really it is just a side effect of hydraulic door closer physics.
The hydraulic door closer doesn’t know that, but the person does.
So now the person who’s 20-30 feet behind you sees that you are being considerate to them, and they now feel obligated not to inconvience you any more than need be, so they hurry up and rush towards the door.
It was kind of the person to minimize your wait since you did them a nicety by holding the door for them.

But what do you do if the person is 35 feet away? 40? 50? Where do you draw the line? Do you take into account how fast or slow they’re walking?
Do they look like the kind of people who are likely to speed up to get the door so you don’t have to wait, or are they going to maximize your suffering by making no effort whatsoever to reciprocate your kindness?

Now imagine you’ve got 2 doors in an airlock type fashion. And there’s 3 people. The situation gets very complicated very quickly.
All because of the stupid hydraulic door closer.

Oh but wait, it gets better.
What if you’re walking towards said door, but you’re walking slowly, and there’s a person behind you who is walking quickly and given the distance from the door, and your relative speeds, they will arrive at the door before you do.
Imagine that. Somebody walking behind you (you being the slow one) can see their miserable future, that they’re going to have to wait for you as they hold the door open for you because they got to the door before you and will kindly hold it open for you. And neither of you are even near the door yet.
Being the slow walker you are, they can tell you’re not going to speed up for them.

Oh those hydraulic door closers. How did they invent such social angst.

Maybe everybody should walk in clumps. You gotta get a ticket to go on this ride. That way, the clump all goes through the door in one stream, and like body surfing, everybody holds the door just a bit until the next person gets it.

I think it’s 2013 and it’s high time we invented a door that doesn’t have this problem.
Revolving doors don’t have this problem. And they’ve already been invented. They also solve lots of air conditioning pressure related problems. Why aren’t there more revolving doors?
Of course revolving doors have their own problems, but that’s a story for another day.

 

Why is there a save button?

Friday, August 30th, 2013

If you look on google for “why is there a save button” you get a lot of results with “why is there no save button.”

People NEED save buttons. That must be why they still exist, because certainly the computers don’t need them.

The save button seems to have come from the technical nature of computers from way back when. There is volatile random access memory, and there is more persistent storage, a hard disk or SD card, or think way back, floppy disks, magnetic tape, and punch cards.

In addition, you could sort of use the two layers of storage, one temporary and one permanent as an undo mechanism. You play around with your data in the application you’re using, and when you’re sure it’s right, you save it permanently. But if you messed something up, you always knew you could go back to the copy saved on the permanent storage.

Well, let’s fast forward to 2013. There’s an already old joke going around, “Why is the icon for ‘save’ a picture of a floppy disk?” Most kids wouldn’t even know what that is. That just goes to show icons are in fact arbitrary which probably explains why I have such a hard time figuring out what they’re supposed to mean.

At least there are hovers on PCs, alas, tablets and phones don’t support hovering as there is no pointer, so it’s pretty much pot luck what’s going to happen when you press one of those buttons with the funny picture on it on your phone.

But I digress.

It’s 2013 again and you still have to hit save to keep from losing your work.

Some higher-end applications like ultraedit and Microsoft word will actually save what you’re doing without you telling it specifically to do so, and they have recovery mechanisms if the program ended for some reason without you actually hitting save.

To me, it seems kinda backwards. Imagine the real world for a second. You have a bag full of groceries on the kitchen table. You take out each item and put it in it’s final resting place, the fridge or the cabinet or the pantry. You don’t then have to say to the bag “ok, I mean it now, keep everything where I put it.”

The contents of the bag are going to stay where you put them, they’re not going to magically appear back in the bag if you leave the room unexpectedly while unpacking.

Why is it so backwards with computers?

The permanent storage is now fast enough that every little change you make can be stored as you do it.

And if you’re worried about having that undo mechanism, there’s plenty of software technology behind undo already built into lots of popular software.

Why on earth do we still have to hit save?

Why is there a save button at all?

I notice that google docs spreadsheet has no save button, but includes a message saying “all changes saved in drive” all the time. Perhaps google got something else right?

A treatise on being average

Wednesday, August 28th, 2013

A friend of mine recently suggested I start writing more, so I’m writing this.

It occurred to me that given a set of people, half of them will be above average and half would be below average.
If there were an odd number of people in that set, there would be one lucky soul who was actually the very definition of average.

Now let’s apply this to the workforce.

The way I understand it, an official unemployment rate of 4% is basically zero because it doesn’t take into account the
off the books employees. According to the bureau of labor statistics: http://data.bls.gov/timeseries/LNS14000000
the current unemployment rate is about 7.4%. That means not everybody who wants to be working is working.
This number does not include mothers staying home with their children who are not looking for work. It doesn’t include
retirees, and presumably most those people walking around at the mall during the day that you wonder “doesn’t anybody work anymore?”
about, are part of this group not reflected in the unemployement rate.

But that leaves us with lots of people who are working.

If you smear that group of working people across all the companies in the country, given an even and fair distribution, each company would
have half their employees being above average, half below average, and if there were an odd number of employees, there’d be
one average person.
But people most certainly aren’t smeared evenly. In the business field, bloomberg takes great care to hire more above average
employees than below average employees. For the moment, I’m leaving out the discussion of what traits define the position on the
scale to which somebody is above or below average, right now, I just want to present an idea.
In the tech field, google expends great energies to make sure they get the best, and thus above average people.
There are also all those mothers who are staying home with their kids who might otherwise be working, though they
may not be looking for work right now. They may be above average
or below average. There is probably some statistic somewhere that would point out that above average mothers tend to have
more or less kids, and thus be out of the workforce for longer, and it doesn’t count people who stay home because their spouse
is working and they don’t have to who also may be above or below average. Since we can’t really know, for now, let’s call them a wash.
But for the rest of them, all these things skew the smearing of people across companies from being even and fair.
Unlike bloomberg and google, there probably aren’t too many companies that go out of their way to hire below average people.
There are certainly companies that tend to accumulate them, but unless there’s some type of insurance fraud scam being run
it is unlikely they are sought out. Sorta like ‘the producers’. Even if it is true, the number is probably smaller than the number
of google and bloomberg-like companies.

(A funny little anecdote here. When I worked at one of my former jobs, one day an edict came down from on high that at the end
of some time period (quarter, end of year, whatever) all the managers would have to sort their employees, and the bottom 5% were
going to be laid off. The end result of this edict was a scramble for below average people. The idea was that if you had amassed
a group of really good people, you didn’t want to have them get laid off, so groups with too many good people, had to transfer some
people out of their group in trade for less good people so they’d have somebody at the bottom of the list to lay off. So while
there was in fact a scramble for below average people, the company wasn’t actually going out of their way to hire them.)

So now we are left with all of the millions of companies employing all these people. Since there are companies that go out of
their way to skew the working population so that there are more above average employees at their company, that leaves less
above average employees to be spread over the rest. There is no counter collection of companies trying to hire below average
employees to compensate. So this would mean that at any company that isn’t egregiously scraping away all the good people
most of the employees are below average.

Does this sound like your place of work?

Now keep in mind the “average” could mean “able to walk and talk mostly at the same time” which would mean everybody is
still competent to do most any job, which would explain how all these companies with more below average people than above
manage to stay in business.

Okay okay. So the obvious answer is that different people are above average at different things. So a competent baker might not be
a competent accountant. So he would get a job as a baker not an accountant.
In this way you can end up with companies filled with people who are good at their job without taking away from other companies
by leaving them the people not good at that job. Those people simply get a job doing something their good at.

The counter counter argument however, is that if you take only a cross section of people who are all in the same business for
your sample set, within that group, the above and below average rankings still apply. But at that point the average can
be pretty high so even the worst of the worst can still be really good employees.

 

Why did evolution invent eyelids but not ear flaps?

Wednesday, July 10th, 2013

I dunno. Comes from all those people who say they are offended by seeing things they were taught are bad to see.

“If you don’t like what you see, don’t look.” Easy enough, you just close your eyes.

But when you’re hearing something you don’t want to hear, you have to at best put fingers over your ears, because you don’t have any earflaps.

I guess the human ear didn’t evolve around country music, thus creating a need to block the sound. Snicker,

Something good.

Saturday, May 18th, 2013

Everybody who knows me has heard me whine many times about how everything sucks. Progress actually goes backwards and despite our richness and modernity and technological prowess, things always seem to fall terribly short of where it’s obvious they could be.

But every once in a while somebody does something so impressive that I can’t help but go on and on about it and tell anybody who will listen.
The last time I remember being impressed about something was my car radio that for $100 played every kind of audio and video media current at the time I bought it. I was just so impressed that for $100 I could put any kind of video or audio format data on a DVD, sd card, or usb device and it would play through my silly $100 car radio.

Well, over the past few months I’ve seen a series of small shifts at a retailer that really put it up there with google on the almost-not-possible scale. And it’s damn impressive.
That retailer is home depot.
I don’t know if they’ve rolled out all these changes to any other stores or if they’re using my local home depot as a test bed but what they do compared to everybody else is nothing short of amazing.

To start, a number of months ago, maybe a year or two now, I lose track of time, they changed the hardware on their self service checkout aisles to accept coins via a coin slot and paper bills so you could pay your entire bill with pennies if you wanted. I particularly liked this feature, as I was able to make use of all my collected pocket change without incurring a 10% fee like you would at coinstar.
At the same time, I also noticed that credit card transactions would complete in maybe 8-10 seconds after I put my credit card in the machine. It would churn rather quickly and spit me out my receipt. I was impressed in that they were making a lot of progress compared to what they used to do.
The checkout system had a scale and would measure the rough weight of what you scanned with the item’s recorded weight. Which means they kept track of every barcoded item and its weight, and they could measure and compare as quickly as I could scan my items. Good job. Neato.

But now, they’ve really outdone themselves. And as far as I know, everybody else, by far.
A month or two ago, they again replaced their self checkout hardware. These machines did one better. Instead of having a coin slot, it’s got a bowl. Yes, you can actually dump a pile of change in this thing and it will sift through it all and rack up the credit. I have noticed you can overwhelm it, but if you feed it at not too fast a pace, it will correctly tally up all your coins, including pennies, and you won’t have to feed the machine one coin at a time. First I thought this was an amazing step forward, but it gets better.
The credit card transaction processing speed, is unlike anything you will see anywhere. You put your credit card in, and before you have finished putting your credit card back in your wallet, your receipt is already printing. Quite literally, less than TWO SECONDS, from reading card, to getting a receipt.
I didn’t even know the credit card companies were capable of that on their end. Just amazing.
But we’re not done, there’s one more.
They may have had this for a while, but I don’t go out to the garden section too often, so I wouldn’t have noticed.
The checkout lines tend to grow quickly in the garden area because there’s only 1 or 2 checkout lanes. Maybe this was a design mistake, more likely it was not, as there’s no point in manning 5 lanes outside because most of the time it’s probably dead out there. But on a nice Saturday, it gets busy quickly.
So they solved the long line problem a different way: They changed the process.

I have long bitched about supermarkets that have 15 aisles and 15 queues rather than 1 queue, and I assume the reason they do it that way is because 1 queue would take up a lot of valuable real estate, or possibly they don’t care, I don’t know. But from an efficiency point of view, 15 queues is dumb.
But in home depot’s case, they only have 2 aisles and therefore 2 queues, so at max load, they back up quickly.
So here’s what they did: They got one or two extra people with portable scanners to work their way up the line, and ring up all the stuff in the cart of the next person waiting on line. They give you a little credit card sized barcoded card that ties your order to you. When you arrive at the till, you hand over your card, they scan it, and vwoom, your entire order is instantly on their till. Even better, the guy who gives you the card from the scanner, can tell you your total so you’re prepared for it. Then he goes to the next guy waiting on line.
What this does is split the work of checking out one person across two people who can work in parallel. The till person only accepts money and spends exactly 0 time sifting through the stuff in your cart you want to buy.

What’s amazing about all this, is that it’s home depot.
Somebody over there deserves a lot of credit for trying to fix the customer service problem, and as a result has achieved real world process applications that blow away anything I’ve ever seen before.
Hopefully they will set the bar for everybody else.

There are even a few more things. I talked to a guy at home depot who told me they were instructed to break the rules to get what the customer wants.
In my case I needed a length of wire, and he’s not supposed to rip open a new otherwise-sellable roll to cut a length for somebody, but that’s what the customer wanted, and they had run out of their wire on the available roll, so he dug it out opened it up and cut me the length I asked for.

Somebody over there is really trying hard, and I hope they get sufficiently rewarded, because it’s rare that I’m impressed by anything, and I am damn impressed by home depot. They are really making progress in the way the word progress was originally intended.