At work we are still in hiring and training mode. The hiring is now not being driven by an expansion of our business, or even the turnover of experienced personnel from retirement or relocation. No, we are now hiring trainees to replace the trainees who washed out.
One guy found a better job, and we determined that a couple of others were not going to work out long term. From management’s point of view, it is always best to cut ties as early as possible, once you have decided that you have an unsatisfactory candidate. Why continue to invest training dollars in someone who is not going to make it?
Despite our best attempts at interviewing and upfront testing, hiring is largely a blind process. I know of no way to determine is someone is truly going to perform in a job, prior to having them start in our unique environment. With some people, it is apparent early on that they will excel, going far in our organization. Sometimes it’s an absolute disaster: the person who creates a quality problem, or breaks equipment, or gets someone hurt. But they all look pretty much the same coming in the front door.
It is also a linear process. When I say that it is linear, I mean that we can only try out one candidate at a time for a position. It would be tempting to hire multiple candidates simultaneously, and then select the best one at the end of training. You retain the star, and drop the others.
A Darwinian selection process would probably be more cost effective than working with one candidate at a time. But that approach strikes me as fundamentally flawed and unfair. We make hiring decisions in good faith. Hiring multiple people, with the intent of cutting some of them, breaks that faith. If a person can do the job, they get to keep the job. I will forego the possibility of finding someone even better.
So, hiring is both, blind, and linear. The metaphor I use to describe this process is playing solitaire. You can shuffle the deck all you want, but eventually you’re going to have to turn over a card. Most of the time you get a number card, a three, or an eight, or a ten. And that’s how many days they last. Then you have to turn over a new card, and try again.
You’re hoping for an ace, you’ll settle for a face card, and you’re praying you don’t get the joker.
Thursday, May 27, 2010
Monday, May 24, 2010
Cell Phones for All
I discovered a fresh outrage this weekend. Unusually for me, I was glued to the tube, watching the series finale of Lost. Four minutes of show, three minutes of commercials. Anyway, from sheer repetition, commercials for Assurant Wireless began to penetrate my consciousness.
Assurant Wireless offers a free cell phone, along with 200 minutes of free nation-wide calling every month, as long as you qualify. The primary qualification is to be receiving food stamps, or to have income below 135% of the Federal poverty level.
The first problem with this is that the “free” cell phone service is not free, of course. It is being paid for out of the taxes attached to the paying customers’ phone bills every month. So I’m being taxed, not to pay for a common good like police or fire protection, but for someone else’s specific good. But the irritation doesn’t stop there.
I can kind of wrap my mind around the concept of food stamps. No one should starve in a nation full of food. But when was it decided that cell phone service is a right, not a discretionary expense?
Then there’s that 135%. The whole point of having a Federal poverty level is that is supposed to be the line demarcating the poor (who presumably need help from the government), from the not-poor, who should be able to fend for themselves.
Those are all bad enough to give me dyspepsia. The part that really frosts me, however, is the television commercials. The contract to provide the “free” cell service is so lucrative that the company (a division of Virgin Mobile) can afford a primetime TV ad campaign. They want to reach anyone who might be qualified, and get them to sign up. There used to be at least a little bit of a stigma associated with government assistance. You were supposed to have enough pride to take care of yourself. Not any more. Don’t be embarrassed to get free cell phone service. You’re entitled to feed at the government trough.
This whole situation irritates me, but in reality, it is only an irritant. The taxes that pay for this program are not unduly onerous. But I wonder how the folks who earn 140% of the poverty line feel. If they want cell service, they have to choose what to sacrifice to afford the luxury of a cell phone. I’ll bet this makes them just wild.
Assurant Wireless offers a free cell phone, along with 200 minutes of free nation-wide calling every month, as long as you qualify. The primary qualification is to be receiving food stamps, or to have income below 135% of the Federal poverty level.
The first problem with this is that the “free” cell phone service is not free, of course. It is being paid for out of the taxes attached to the paying customers’ phone bills every month. So I’m being taxed, not to pay for a common good like police or fire protection, but for someone else’s specific good. But the irritation doesn’t stop there.
I can kind of wrap my mind around the concept of food stamps. No one should starve in a nation full of food. But when was it decided that cell phone service is a right, not a discretionary expense?
Then there’s that 135%. The whole point of having a Federal poverty level is that is supposed to be the line demarcating the poor (who presumably need help from the government), from the not-poor, who should be able to fend for themselves.
Those are all bad enough to give me dyspepsia. The part that really frosts me, however, is the television commercials. The contract to provide the “free” cell service is so lucrative that the company (a division of Virgin Mobile) can afford a primetime TV ad campaign. They want to reach anyone who might be qualified, and get them to sign up. There used to be at least a little bit of a stigma associated with government assistance. You were supposed to have enough pride to take care of yourself. Not any more. Don’t be embarrassed to get free cell phone service. You’re entitled to feed at the government trough.
This whole situation irritates me, but in reality, it is only an irritant. The taxes that pay for this program are not unduly onerous. But I wonder how the folks who earn 140% of the poverty line feel. If they want cell service, they have to choose what to sacrifice to afford the luxury of a cell phone. I’ll bet this makes them just wild.
Friday, May 21, 2010
Synthetic Life
This was a big week for molecular biology geeks. Craig Venter and his team at the Venter Institute in Maryland announced that they had successfully created what they call “synthetic life.” Like Dr. Frankenstein, we have now created life from non-living material. Surely designer organisms that eat carbon dioxide and secrete oil are not far away, thereby solving both global warming and the energy crisis!
Well, maybe not.
When you get beyond the hype of the press release and examine what Venter’s team really did, it boils down to three things. 1) They completely decoded the DNA of an existing bacteria. This is known as sequencing the genome. 2) They artificially created a copy of the sequenced DNA, starting with the nucleotide bases that are the building blocks of the DNA molecule. As part of that process they modified some of the nonfunctional sections of the DNA to insert what are called “genetic watermarks.” 3) They took a different bacteria, removed the genetic material that was native to the germ, and implanted their prebuilt DNA. The newly implanted DNA then took over the machinery of the cell and began making copies of itself in the normal process of cell division.
These are all neat tricks, to be sure, and incredibly difficult to boot. But I think they fail two key tests for the claim of truly creating artificial life. First, true artificial life will have a genetic code designed from scratch. Venter borrowed the genetic code from an organism that had developed over a billion years of trial and error evolution.
Second, to be truly synthetic, they would have to combine their novel DNA with both a lab grown cellular membrane and built from scratch cellular metabolic machinery. If that assembly then began to grow and reproduce, then they could truly claim to have built an artificial life form.
We’re still a long way away from gleefully cackling “It’s alive! My creature lives!,” as we rub our hands together with glee. Consider: the bacterial DNA Ventner synthesized consisted of a little over one million base pairs, combined to make a single chromosome. Human DNA includes over three billion base pairs, spread out over 46 chromosomes. That’s three thousand times more genetic code.
Those caveats aside, it is still an astonishing technical achievement. Clearly with some fine tuning, they will soon be able to use their techniques to do the kind of design work mentioned above.
Well, maybe not.
When you get beyond the hype of the press release and examine what Venter’s team really did, it boils down to three things. 1) They completely decoded the DNA of an existing bacteria. This is known as sequencing the genome. 2) They artificially created a copy of the sequenced DNA, starting with the nucleotide bases that are the building blocks of the DNA molecule. As part of that process they modified some of the nonfunctional sections of the DNA to insert what are called “genetic watermarks.” 3) They took a different bacteria, removed the genetic material that was native to the germ, and implanted their prebuilt DNA. The newly implanted DNA then took over the machinery of the cell and began making copies of itself in the normal process of cell division.
These are all neat tricks, to be sure, and incredibly difficult to boot. But I think they fail two key tests for the claim of truly creating artificial life. First, true artificial life will have a genetic code designed from scratch. Venter borrowed the genetic code from an organism that had developed over a billion years of trial and error evolution.
Second, to be truly synthetic, they would have to combine their novel DNA with both a lab grown cellular membrane and built from scratch cellular metabolic machinery. If that assembly then began to grow and reproduce, then they could truly claim to have built an artificial life form.
We’re still a long way away from gleefully cackling “It’s alive! My creature lives!,” as we rub our hands together with glee. Consider: the bacterial DNA Ventner synthesized consisted of a little over one million base pairs, combined to make a single chromosome. Human DNA includes over three billion base pairs, spread out over 46 chromosomes. That’s three thousand times more genetic code.
Those caveats aside, it is still an astonishing technical achievement. Clearly with some fine tuning, they will soon be able to use their techniques to do the kind of design work mentioned above.
Thursday, May 20, 2010
Educational Qualtiy Control
I want to build on my concept of testing as educational quality control that I started in my last post.
Quality control, or quality assurance, as it is sometimes called, is a core function in every manufacturing plant I’ve ever seen. We continuously check product at every stage, from incoming raw material to assembly to final audit before shipping. The question is always the same: does the product meet specifications? Put another way, what we are doing is verifying that the manufacturing process is generating the desired result.
If you check parts, and find you aren’t getting the correct result, then there is a problem with the process. But until you do your quality checks, you don’t know what you are producing. Also, one of the key guidelines of quality control is to check your quality as early in the process as possible. It makes no sense to wait until after you have finished final assembly to test the product and discover there was a problem with the first step in the assembly process. Finally, a good quality control program relies on objective evidence. If a good part has to be round within .004”, everyone agrees on that, and everyone agrees on how to measure the part. There is never any discussion of “the parts look round enough to me.” It is either round within .004”, or it is not. If not, then there is a problem, and we have to fix the problem.
In education, standards and standardized testing take the place of quality control. That’s why I can never understand why the educational establishment fights so hard against standardized testing. Actually, that’s not true. I completely understand why teachers fight so hard against testing. If I had a complete lack of accountability for producing the desired result in my job, I would want to protect that too.
What I don’t understand are the lame excuses offered as to why we shouldn’t use standardized tests to evaluate the performance of our education processes. For example, “all we do is teach to the test.” If the test is a representative sample of what we want students to know, than test scores should accurately show the students’ mastery of that body of knowledge.
Conversely, without testing, there is no way to establish that the student has learned anything. And based on my experience trying to employ a number of high school graduates, many of them really haven’t learned anything.
Quality control, or quality assurance, as it is sometimes called, is a core function in every manufacturing plant I’ve ever seen. We continuously check product at every stage, from incoming raw material to assembly to final audit before shipping. The question is always the same: does the product meet specifications? Put another way, what we are doing is verifying that the manufacturing process is generating the desired result.
If you check parts, and find you aren’t getting the correct result, then there is a problem with the process. But until you do your quality checks, you don’t know what you are producing. Also, one of the key guidelines of quality control is to check your quality as early in the process as possible. It makes no sense to wait until after you have finished final assembly to test the product and discover there was a problem with the first step in the assembly process. Finally, a good quality control program relies on objective evidence. If a good part has to be round within .004”, everyone agrees on that, and everyone agrees on how to measure the part. There is never any discussion of “the parts look round enough to me.” It is either round within .004”, or it is not. If not, then there is a problem, and we have to fix the problem.
In education, standards and standardized testing take the place of quality control. That’s why I can never understand why the educational establishment fights so hard against standardized testing. Actually, that’s not true. I completely understand why teachers fight so hard against testing. If I had a complete lack of accountability for producing the desired result in my job, I would want to protect that too.
What I don’t understand are the lame excuses offered as to why we shouldn’t use standardized tests to evaluate the performance of our education processes. For example, “all we do is teach to the test.” If the test is a representative sample of what we want students to know, than test scores should accurately show the students’ mastery of that body of knowledge.
Conversely, without testing, there is no way to establish that the student has learned anything. And based on my experience trying to employ a number of high school graduates, many of them really haven’t learned anything.
Monday, May 17, 2010
Training the Trainer
We’ve gone into hiring mode for the first time in a couple of years. This is due to two factors. First, business continues to strengthen, necessitating more workers to get the job done. Second, some of our people have moved on to greener pastures. When somebody leaves, it creates an opening for somebody else. The end result is that we’re not just calling back former employees who were laid off. We’re bringing new people into the organization.
Since it has been a couple of years since we have done that, we are relearning lessons on training new people. Our usual method of training is OJT or job shadowing. We assign the trainee to follow an existing employee for a few weeks. In theory, the trainer shows the trainee all of the ins and outs of the job. Anything the trainer doesn’t teach, can be picked up from the work instructions and operating procedures.
At least, that’s the theory.
We had one guy training for five weeks at one position. When we gave him a written test at the end of that time, he failed miserably. It turns out he couldn’t read the questions, either on the test paper or the work instructions that contained the answers. We had to let him go. Oh yeah, we’re supposed to test for literacy on the front end. So now we’ve reinstalled that part of our hiring procedures.
Another lesson we’ve relearned is to provide milestone tests to be administered during the training period. We fell into the lazy trap of waiting until the very end of training, and then testing everything at once. We had one person fail all of the tests. It turned out the trainer had used the “monkey see, monkey do” technique to have the trainee go through the motions of the job. But the trainer had never bothered to explain what the individual actions meant, or why they were included in the job.
It’s possible that the trainer had tried to explain these things. But this was the first time that particular trainer had ever trained anyone, it is very loud out on the shop floor, and the trainer was not particularly articulate in the first place. However, we realized that by virtue of hindsight. The bottlenecks to successful training could have been removed weeks earlier if we had enforced a testing regimen on both the trainee and trainer.
In our training model, testing performs a quality control function. If you want to know if your training is effective, than you devise a test that covers the material being trained. If the trainee can pass the test, then the training is effective. If they cannot pass the test, then your training is ineffective.
Obviously, if your training is going to be ineffective, you want to know as early in the process as possible. That way you can take steps to improve your training (maybe replace the trainer, maybe replace the trainee) before too much cost has been expended.
Because the testing is a pain in the butt, the trainers tend to push it off as long as possible. It’s up to management to enforce the testing regime. And as this whole incident has shown, everybody needs refresher training every now and again.
Since it has been a couple of years since we have done that, we are relearning lessons on training new people. Our usual method of training is OJT or job shadowing. We assign the trainee to follow an existing employee for a few weeks. In theory, the trainer shows the trainee all of the ins and outs of the job. Anything the trainer doesn’t teach, can be picked up from the work instructions and operating procedures.
At least, that’s the theory.
We had one guy training for five weeks at one position. When we gave him a written test at the end of that time, he failed miserably. It turns out he couldn’t read the questions, either on the test paper or the work instructions that contained the answers. We had to let him go. Oh yeah, we’re supposed to test for literacy on the front end. So now we’ve reinstalled that part of our hiring procedures.
Another lesson we’ve relearned is to provide milestone tests to be administered during the training period. We fell into the lazy trap of waiting until the very end of training, and then testing everything at once. We had one person fail all of the tests. It turned out the trainer had used the “monkey see, monkey do” technique to have the trainee go through the motions of the job. But the trainer had never bothered to explain what the individual actions meant, or why they were included in the job.
It’s possible that the trainer had tried to explain these things. But this was the first time that particular trainer had ever trained anyone, it is very loud out on the shop floor, and the trainer was not particularly articulate in the first place. However, we realized that by virtue of hindsight. The bottlenecks to successful training could have been removed weeks earlier if we had enforced a testing regimen on both the trainee and trainer.
In our training model, testing performs a quality control function. If you want to know if your training is effective, than you devise a test that covers the material being trained. If the trainee can pass the test, then the training is effective. If they cannot pass the test, then your training is ineffective.
Obviously, if your training is going to be ineffective, you want to know as early in the process as possible. That way you can take steps to improve your training (maybe replace the trainer, maybe replace the trainee) before too much cost has been expended.
Because the testing is a pain in the butt, the trainers tend to push it off as long as possible. It’s up to management to enforce the testing regime. And as this whole incident has shown, everybody needs refresher training every now and again.
Thursday, May 13, 2010
A New Passion: Off Topic Post
In the last couple of weeks a have discovered a new area of interest: molecular biology.
It started last year when I bought The Teaching Company course on Big History: The Big Bang, Life on Earth, and the Rise of Humanity. Fourteen billion years condensed into 48 lectures on CD. Woo-hoo!
One of the critical thresholds discussed in Big History was the origin of life on Earth. So a couple of months ago I bought the course The Origins of Life. To talk about the research into the origin of life, you have to talk about biochemistry. There is also a chicken and egg type of problem with the origin of life. To be recognizable as life, the organism has to have both metabolism, a way of extracting energy from the environment, and inheritable reproduction, a way of making copies of itself. Which came first, and how did they combine? Nobody knows ... yet.
One of the lectures in The Origins of Life course covered the basics of genetic coding, how RNA is built out of DNA, and how proteins are assembled by RNA. I was hooked. I know we can splice DNA, and I know we have sequenced the human genome, but I don't know what that means. So I'm on a journey to find out.
My current Teaching Company course is on fundamentals of biology, which includes a number of lectures on cellular mechanics. After listening to a couple of lectures, I went online and did some searches for molecular biology. I found a Berkeley molecular biology course, and listened to a couple of lectures. Different perspectives, but not as well structured as The Teaching Company course.
Then, on a whim, I went to YouTube and typed in "DNA Replication." I hit the mother lode. It turns out there are tons of animations for all kinds of biological processes: DNA replication, translation, and transcription. Under metabolism, there are animations for glucolysis, the Krebs cycle, the citric acid cycle, and ATP synthase.
The animations run the gamut from block diagrams to real time scale reproductions of cellular processes. What is most amazing to me is that we now actually know the shapes of the various large molecules in these processes.
This is a whole new area of human knowledge for me to explore, and I can totally get my geek on doing so. I foresee hours of fun chasing down the minutiae of various biochemical processes.
Yeah, I'm a nerd. You got a problem with that?
Monday, May 10, 2010
Greece's Problems
Greece’s fiscal problems have been much in the news recently. Essentially, the country is bankrupt. They have bond payments coming due this month, and they don’t have enough euros in the treasury to pay back the bond holders.
There is nothing unusual about that. Most governments don’t actually pay off their bond holders when the note comes due. What they do is issue new bonds, and just keep rolling the debt over. Greece’s problem is that they have hit their credit limit. The international financial markets are so nervous about how much debt has already been issued that they don’t want to allow Greece to continue digging the hole deeper.
This is not the first time a sovereign nation has run into this problem. Nations can’t handle their credit cards any better than the average American. The nation state playbook says that in a circumstance like this, you devalue your currency. Devaluation makes your exports cheaper, imports more expensive, and pays back the bond investors with a cheaper currency than they loaned you. The inflationary effects make everyone poorer, including the bondholders, who have to take a haircut on the value of their investment.
This isn’t an available option for Greece, because the Greeks don’t have their own currency anymore. They use the common European currency, the euro. If Greece defaults on its bonds, all of the countries in the Euro zone are in the splash zone. Hence the incentive for the other European Union countries to bail Greece out.
The other European nations, notably France and Germany, along with the International Monetary Fund, have agreed to be the lender of last resort to the Greek government. But there are conditions. They are requiring Greece’s government to reduce the government budget deficit from 13.9% of GDP to 3.9% over the next three years.
With their back against the wall, the Greeks are agreeing to the plan. They are cutting pensions, cutting salaries of government employees, and raising the retirement age. On the revenue side, consumption taxes are being increased one tenth, from 20% to 22%.
How big a cut is this going to be? Government spending makes up about 43% of the total Greek economy. The proposed austerity package of tax increases and budget cuts aims to get that down to about 35%. The government in Greece is going to have to shrink by about 20%. Overall, the average man on the street is going to get 10% poorer over the next couple of years, but the effect will be concentrated for government employees and retirees.
No wonder they’re protesting.
There is nothing unusual about that. Most governments don’t actually pay off their bond holders when the note comes due. What they do is issue new bonds, and just keep rolling the debt over. Greece’s problem is that they have hit their credit limit. The international financial markets are so nervous about how much debt has already been issued that they don’t want to allow Greece to continue digging the hole deeper.
This is not the first time a sovereign nation has run into this problem. Nations can’t handle their credit cards any better than the average American. The nation state playbook says that in a circumstance like this, you devalue your currency. Devaluation makes your exports cheaper, imports more expensive, and pays back the bond investors with a cheaper currency than they loaned you. The inflationary effects make everyone poorer, including the bondholders, who have to take a haircut on the value of their investment.
This isn’t an available option for Greece, because the Greeks don’t have their own currency anymore. They use the common European currency, the euro. If Greece defaults on its bonds, all of the countries in the Euro zone are in the splash zone. Hence the incentive for the other European Union countries to bail Greece out.
The other European nations, notably France and Germany, along with the International Monetary Fund, have agreed to be the lender of last resort to the Greek government. But there are conditions. They are requiring Greece’s government to reduce the government budget deficit from 13.9% of GDP to 3.9% over the next three years.
With their back against the wall, the Greeks are agreeing to the plan. They are cutting pensions, cutting salaries of government employees, and raising the retirement age. On the revenue side, consumption taxes are being increased one tenth, from 20% to 22%.
How big a cut is this going to be? Government spending makes up about 43% of the total Greek economy. The proposed austerity package of tax increases and budget cuts aims to get that down to about 35%. The government in Greece is going to have to shrink by about 20%. Overall, the average man on the street is going to get 10% poorer over the next couple of years, but the effect will be concentrated for government employees and retirees.
No wonder they’re protesting.
Subscribe to:
Posts (Atom)