Several years ago, as I was coming up on ten years of experience in project management, I decided to obtain professional certification. My reasons included a desire for professional growth, to interact with peers, to receive and in turn transmit knowledge, building contacts, and of course the cynical potential of personal gain. A bit of research confirmed that the Project Management Institute, which I'd heard of before, remained the 800 pound gorilla of the field. They offer the "Project Management Professional" credential, or PMP, which requires a college degree, 4500 hours and 36 months of project management experience, 35 hours of classroom training, and passing an exam.

In a series of upcoming posts, I will describe my experience with the PMP in detail. I will try to answer questions such as, Is the PMP helpful for employers? Is the PMP helpful for project managers? How practical is the knowledge covered in the PMP exam? How hard is the PMP exam? How well does the PMP satisfy its goal to "advance the project management profession and to recognize the achievements of individuals"? How well does PMI "promote a unifying influence in the advancement of Project Management"?

To start things off here in part 1, I'll summarize the controversy I saw surrounding the new exam released in 2005 with the Third Edition of the Project Management Body of Knowledge guide, or PMBOK.

---

I started thinking about getting a PMP in late 2004, but didn't do much research until early 2005. I discovered that a new version of the exam would be rolled out in September 2005. But before you can take the exam, you have to have 35 classroom hours. One option was to take a class at one of the local universities. San Diego State University and University of California San Diego both offer PMP programs, which are one or two-year affairs with a total cost in the range of $5,000 to $10,000. However, you can take a single class and pay in the neighborhood of a thousand dollars. But, it turns out that the San Diego Chapter of PMI offers specific Exam Prep classes that provide the required number of hours. And, because they are taught by volunteers who are themselves PMPs (seeking the 60 Professional Development Units, or PDUs, that PMPs must accumulate every three years to maintain certification), the classes cost only about $400.

The summer classes coincided with my annual July Vancouver vacation, which ended the chance of taking the old version of the exam. So I signed up for the fall class, six full-day Saturdays in a row, and after completing the classes registered to take the new exam in early 2006. During the class, many students were quite nervous about the new exam, and I found out why: The previous exam required 141 correct out of 200 questions, or 70%. The new exam still had 200 questions, and still required 141 correct, but 25 of the questions would be present solely for "testing the test", and would not count towards correct answers. So, on October 1, 2005, not only did all the questions change from "second edition" to "third edition", but the passing score rose from 70% to 81%.

Rumors passed around class included that the summer classes had forty or fifty people (we had about 12 regulars), that all available test slots in the San Diego area had been booked for months before the cutoff date, and that some students had flown to Nebraska and other underpopulated mid-West states solely to be able to take the old exam before the cutoff. The instructors said that this was part of an effort by PMI to "raise the bar" for new PMPs and make sure the credential didn't get diluted. As the class continued into OCtober, I heard more rumors, both from within class and without, that the percentage of test-takers passing the test on the first try was plummeting from 70%+ to 40% or even lower. (The test costs $405 (for members of PMI, itself a $119/yr cost, plus $10 for application and $30 for the local chapter), and $275 for re-takes.)

Eventually, I received an PMI email—not from PMI, but forwarded from an early test-taker:

...Before offering the new examination, PMI assembled a group of volunteers to help establish the passing score. Using a method known as the "Modified Angoff Technique" (a proven exam development method), a group of global PMPs in the summer of 2005 assessed each test question and independently evaluated the questions to determine their difficulty level. Their responses were then sent to PMI's psychometric (exam development) experts and averaged. From that information, PMI?s psychometricians recommended that PMI adopt a passing point of 81 percent (141 correct questions).

... After performing a statistical analysis of [actual test results], PMI and its independent psychometricians were able to make conclusions about the performance of questions as well as candidate performance ... PMI revised the passing score for the exam to 61 percent (106 correct questions). PMI then applied the new passing score to all examinations taken since 30 September 2005 by candidates who sat for the new exam. PMI is in the process of updating the candidates' records.

While we remain extremely sensitive to candidate and trainer concerns about such a change in the passing score, these considerations must be weighed in context of the overall purpose of the exam: to provide a consistent global standard that all practitioners must meet to ensure the credential is awarded to qualified individuals. ...

PMI understands that the changes to the exam and its passing score raise
numerous questions. ...

—forwarded email from Drew Ihlenfeld, PMI

It's hard to see such a drastic rescoring of the exam as anything other than a major failure of the exam development process. Of course, the retroactive rescoring effectively corrects the problem, but that's after-the-fact quality control, and it means that their before-the-fact quality assurance failed. And it doesn't address the wear and tear on PMP applicants, both those that were temporary failures and all of us in the months before and after the transition that had to make decisions in an atmosphere of uncertainty. For example, the email was dated 30 November, but it was weeks later before anything about it was posted on the website.

In the event, my classmates and I were very relieved to learn of the change in scoring, and I finally took and passed the exam on April 1, 2006 with a score in the high 70s. Nonetheless, I was somewhat bemused to recently read PMI's version of the fiasco in the 2005 annual report:

As part of best practices for exam development, PMI proactively reviewed data collected on the revised Project Management Professional (PMP) certification examination.

Driven by a role delineation study identifying volunteer-credentialed practitioner recommendations on how the skill sets for a PMP had evolved, PMI conducted this evaluation and revision based on the Institute's desire to accurately provide a consistent global standard that ensures the credential is awarded to qualified individuals, and meets the needs of PMI's stakeholders. Specifically, the attributes to "lead" and "direct" were identified as key elements of a more mature PMP.

PMI utilized proven examination development methods in revising the PMP examination, and conducted all appropriate due diligence to assess the applicability and difficulty of each test question as determine the overall percentage passing point.

Quality control measures implemented prior to and following the exam launch data indicate the exam is functioning as designed, including the fact that no inaccuracies in scoring and no malfunctions in the test's administration or its translations have occurred.

—PMI 2005 Annual report, p 23-24

Harper's Magazine is famous for its index, a list of numbers about the world. Like the game show Jeopardy's "answer in the form a question", this gimmick is rarely especially effective. For example,

Year that a signboard tallying the U.S. national debt was erected near Times Square: 1989 [Douglas Durst (N.Y.C.)]

Year in which it is expected to run out of digits: 2007

—Harper's

is pretty awkward way to deliver this factoid.

Talking Points Memo presents a few numbers today which are better suited to the Harper's Index format:

Number of Technorati mentions of Ned Lamont (CT): 26,578
                           of Sherrod Brown (OH):  6,764
                                of Jim Webb (VA):  4,516
                          of Bob Casey, Jr. (PA):  3,157
                              of Jon Tester (MT):  2,325
                             of Jack Carter (NV):  2,077
                        of Claire McCaskill (MO):  1,976
—Talking Points Memo

Who are all of those people who aren't Ned Lamont? They are Democratic challengers to incumbent, vulnerable Republican senators. And they're being ignored by the netroots (and the netroots' wallets) who are distracted by a fight between a Democrat and an independent with a track record of voting strongly Democrating who has pledged to caucus with the Democrats. Karl Rove is laughing all the way to the End Times. (For the record: I've never liked Lieberman. I think he's the political equivalent of an empty .300 hitter (a baseball player whose superficial statistics are good but whose fundamentals are bad), and I hope he loses. But if I contribute money, it won't be to help Lamont.)

And the second Harper's figure from this morning:

Number of reporters contributing to Friday's front page New York Times story on the JonBenet Ramsey case: 13

Number of reporters contributing to Friday's front page New York Times story on the federal court ruling that the NSA warrantless wiretapping program is unconstitutional: 2

—Talking Points Memo

I read the Becker-Posner blog because they are both extremely smart, extremely opinionated, very articulate, and often spectacularly wrong in the very special way that only utterly, arrogantly brilliant people can be.

In the middle of a discussion of airplane security, Becker had this gem:

Although civil libertarians criticize "profiling" of travelers and others, and government officials deny they engage in it, profiling is a necessary part of any reasonably effective security system. Groups that should be scrutinized carefully differ over time and among region of the world. [...] Young Muslim males of Pakistani and Arab background have been responsible for the vast majority of recent terrorist activities in America, Britain, and continental European countries. [...] . Therefore, young males from these groups should receive especially close scrutiny at airports and other public places. [...] To limit the discomfort and anger caused by profiling, members of the profiled groups should be treated politely and with dignity. They should also be reminded that they too are being protected from terrorist activities by a small fringe.

Those objecting to profiling potential terrorists usually want to subject everyone to the same detailed examination and inquiry. However, when potential terrorists are part of a group that constitutes only a small fraction of the population, searching everyone with the same detailed care at airports or at other venues would be needlessly costly and time consuming. This would slow down and thereby reduce air travel and other vulnerable group activities. It would also lead to loud complaints by those affected after the fear of terrorism had abated.

I don't want to address the arguments for or against racial profiling. I just want to point out that Becker is making two contradictory arguments:

• young Muslim males of Pakistani and Arab backgrounds should be profiled closely, over their own objections if any
• One reason not to profile everybody equally is that people would complain

It's hard for me to see any interpretation other than that the complaints of young Muslims count for less than the complaints of others. I suppose Becker would say that their complaints are valued equally but they have to suck it up for the team, basically. To which I say that a fundamental characteristic of pluralist, democratic cultures must be that nobody is asked to bear more than their fair share of a burden based on attributes that are out of their control. I think the word for that is discrimination.

It may be empirically true that young Muslim males of Pakistani and Arab background are a greater threat to, say, US security than any other group defined by age, ethnicity, gender, and religion. I am skeptical of that claim for several reasons: several other profiles, such as white supremacists and Christian anti-abortion terrorists, compete hard in the rankings; and successful Muslim terrorist groups already include, and are actively recruiting more, people out of that profile.

But even if that were empirically true, any stance by the rest of us other than demanding equal inconvenience and suspicion for all is unacceptable and un-democratic. Saying "sorry, brown guy, but you have to stand in line and get dirty looks and just maybe go into secret detention for a few years and get shipped to Romania for outsourced torture sessions" is the exact opposite spirit of Danish King Christian's classic response to the Nazi invasion: "We'll all have to wear yellow stars." (Of course, he never actually said that. I'm not sure if it qualifies as truthy; regardless, Denmark did indeed have an excellent record of protecting its Jews during the Holocaust.)

When I got back from a trip to Vancouver and Seattle, I booted up my desktop, put my Palm Pilot in the dock and synced. Something about the port configuration that the rebooting machine put together didn't satisfy the Palm, and it crashed. Hard. Soft reset did nothing. Hard reset did nothing.

This happened once before, a few months ago, and at the time it occured to me to let the battery run down and see if it would really reset itself. It has a remarkably long-lasting, built in lithium battery, and it took a day or so before it occured to me to take it out of the charging cradle; from that point, it was over a week before it wound down. In fact, I happened to check it when it was down to maybe 5% battery, and it had awoken from its mystery crash in order to announce that it was almost out of battery power. I promptly put it back in the cradle and all was fine.

So, if I didn't want to lose the notes I had taken on the trip, I would have to set it aside, but check it regularly, and hope to spot the window after the battery was low enough to trigger the warning that uncrashes it, but before it actually ran out of energy and erased itself. And hope that that was the problem, not something more serious.

In the meantime, I was going crazy without my Palm, so I got the cheapest new model, the Z22. It does everything that the Vx does, costs $100 new (my original V was about $300 in 1999; after I lost it 2001, I replaced it with a Vx that I bought used in Hong Kong for US$130), and has 32 Mb instead of 8, a color screen, and USB instead of serial. I considered the other, more expensive Palms, but they are all bigger and heavier, and do things I don't want my Palm to do, like take phone calls (no thanks), connect to the internet over wireless (too, too tempting), or play music (got a Shuffle for that).

After a few days, I reached my conclusion: I don't care for the Z22. The color screen is too bright in the dark and too dim in direct sunlight. There's no cover, so when you put it in your pocket it likes to wake itself up; keylock is a pain in the ass. Palm lost a lawsuit over Graffiti, and the new Graffiti 2 sucks in comparison. I followed some instructions to hack the old Graffiti on, and I like how it shows your strokes when you write in the main area, but it crashed whenever I wrote anything in the special writing area. The battery seemed to run down a lot faster than the Vx, as well. The Vx's screen is perfectly readable in direct sunlight and the Indiglo mode is sexy in the dark. Dropping the Vx into the cradle is much more satisfying than putting the Z22 on the desktop and plugging in a loose cable. And the Vx's thin, shiny aluminum shell is simply far more attractive than the Z22's mundane glossy plastic.

A week later, I put the Vx in the cradle for a few minutes, and it came back to life. I lost a few days of data from the trip, but I had my lovely machine back. The Z22 went back to Best Buy, unlamented.

Bruce Schneier on the new airport rules:

And a long-term prohibition against liquid carry-ons won't make us safer, either. It's not just that there are ways around the rules, it's that focusing on tactics is a losing proposition.

It's easy to defend against what the terrorists planned last time, but it's shortsighted. If we spend billions fielding liquid-analysis machines in airports and the terrorists use solid explosives, we've wasted our money. If they target shopping malls, we've wasted our money. Focusing on tactics simply forces the terrorists to make a minor modification in their plans. There are too many targets -- stadiums, schools, theaters, churches, the long line of densely packed people before airport security -- and too many ways to kill people.

Security measures that require us to guess correctly don't work, because invariably we will guess wrong. It's not security, it's security theater: measures designed to make us feel safer but not actually safer.

Airport security is the last line of defense, and not a very good one at that. Sure, it'll catch the sloppy and the stupid -- and that's a good enough reason not to do away with it entirely -- but it won't catch a well-planned plot. We can't keep weapons out of prisons; we can't possibly keep them off airplanes. —Schneier

And a cartoon

Today's good news, a twofer of adjacent articles in the Times. First, PepsiCo corporation selected an Indian-born woman as CEO. Second, parts of India are banning Coke and Pepsi. Unfortunately, they're banning them because they think they have pesticides, not because they are unhealthy sugar water that shouldn't be sold in schools. But I guess that's now Ms. Nooyi's problem.

Mammoth, John Varley.
Another competent, pedestrian science fiction novel from former sf god John Varley.

Pretender, C.J. Cherryh
"The second book of the third FOREIGNER sequence." That would make it book eight in this series about the politics of various factions of humans and aliens on planet far from Earth. The cover art still sucks; the prose is as good as ever if you like her thing, which I do.

I love my Thinkpad. My first was a used A20 (which got brain-transplanted into somebody else's disused A21 when the A20's screen went wonky). Last year I bought a new X40, and demoted the A21 the dining nook for reading while snacking. (Its battery life is below 30 minutes anyway.) The X40 is a superb laptop; small, light, good screen, great keyboard, hardware which is adequately supported on linux. (There's even a linux module to read the hard drive impact sensor, so you can tap signals on the case to run commands.)

When I heard a few months ago that " Lenovo will not install or support the Linux operating system on any of its PCs ... Lenovo is positioning itself as an exclusive partner of Microsoft..." (), I was very dismayed at the thought that my third laptop would have to be something else, although maybe in a few years wearables will be cheap and standard. Happily, Lenovo later disavowed that marketing position.

Anyway, the subject of this post is the docking station. I used to think that laptops and docking stations were silly, until I got more nomadic a few years ago. I still do most of my work in the productivity pod, but I take the laptop on trains and to client sites and to coffee shops and up to the top floor lounge in my apartment building. The laptop came with Windows, and I decided to leave it on (after installing Linux and a dual-boot setup) for games. When I want to play Civ IV on the big monitors, I have to unplug and replug the keyboard and laptop, plug in the external DVD drive, unpack the power cable and plug it in (both ends), etc. I'll readily admit that it's less onerous than walking miles every day to collect drinking water, but after a year of this I finally got a USB KVM switch and, after a month of hunting, a used docking station that included the power adapter.

And they work (the Belkin KVM switch is, astonishingly, working as advertised, making it far and away the least defective piece of Belkin equipment I have ever used). But I do want to complain about one detail of the docking station, which is the docking. (And undocking). To undock it, you tug a lever, which feels both stiff and flimsy, as if it might break off at any moment, and this causes the laptop to disengage and hop up a few millimeters. It also makes a loud beep.

To put the laptop into the docking station, you line it up and push down. It has a very mushy feel, so you push down some more. It never actually clicks or in any other way provides tactile feedback that the electronics are engages, but the screen does brighten when it goes back on AC. Of course, you can't see that because you've closed the lid in order to push. Even after it's fully seated, there is still some wiggle and wobble room between the Thinkpad and the dock.

So obviously an ergonomic improvement would be the best fix - a nice smooth push into place with gradually increasing but still mild resistance, terminated with a solid click. But failing that, it would be nice if the beep was used to tell you that the machine is now seated, rather than telling you the machine that was seated is now unseated. I can generally tell that it's unseated by the fact that it's in my hands and I'm walking away from the dock, thank you very much.

Two thirds of the way into the season, each of the five teams in the NL West (baseball) has been in both first place and last place. With 106 games played, the total distance between first and last is 4.5 games. Last year the Padres won the division with a record of 82 and 80, the worst winning record for any team in the history of baseball. This year, maybe the whole division can finish three or four games over .500, in a five-way tie. Then we can go to goal differential to determine the winner.