developer.* Blogs - Only the Best? - Comments

Personal mastery

chrishmorris — Sun, 10 Jun 2007 23:05:18 -0700

I agree Donna, that most people can do something useful, even if it isn't the task they would choose. If I can't help someone to be effective then I also try to learn something about selection, so I don't repeat the mistake next time I'm hiring.

It's heartbreaking when you have someone in their early twenties who really wants to be a programmer, and who clearly won't make it. They may be plenty smart, but just not have the right sort of smarts, and it isn't fun to be around as they face up to that fact.

There are others who do fine at that stage but stop growing later, and eventually become fifty year old junior programmers. That seems pretty tragic too.

Anyone who reads the literature on managing programmers is familiar with the claim that the best are ten times more productive than the worst (an underestimate in my experience since it doesn't allow for the negatively productive). But actually most programmers and many of those who manage programmers are not familiar with this discussion.

One thing we can do for our staff is ensure that they have an accurate vision of what a great programmer is. To make a commitment to personal growth you must know that you haven't go there already! I've seen this work.

A note on being the "best"

Edward G Nilges — Tue, 10 Oct 2006 19:13:44 -0700

The Economist this week has an article about "attracting the best talent".

The problem is that the idea of the "best" in corporate and in MIS is unexplored.

Missing is the idea of specialization. You can drink the Kool aid on a platform and rapidly become the "best" at this platform, to the extent it blinds the soul, and makes you into one of those unpleasant MIS personalities who attacks people for not drinking his kool aid.

The Economist means by "the talented employee" a fragmentary man, a one-dimensional man, a time slice with a relationship to the means of production useful for a short term goal.

His designation as "talented" is given to him, or withheld, by men in suits who pride themselves on their ignorance of the technology.

Life for the full person as opposed to the time slice, the "programming resource", the CEO, becomes stunted, as stunted as Darth Vader was when Luke Skywalker took off that stupid plastic helmet.

I'd prefer to have been the best father and husband to being the greatest programmer in the world, because the latter doesn't exist.

It's just a one-shot relationship to the means of production.

Software development, religion and philosophy

Edward G Nilges — Tue, 10 Oct 2006 09:27:26 -0700

The problem with Steve Yegge's column is that to "prove" a software platform "superior" you have to engage in a dedicated and pernicious form of pseudo-science.

The purpose of software is ultimately to control, modify, or at a minimum influence HUMAN behavior, to produce software that humans, who it seems remain necessarily outside the topos or margin of software because, it appears to me, their consciousness is not reducible to that of a computer peripheral.

The problem with a scientifically "proven" tool for ultimately controlling, modifying or at a minimum influencing human behavior is that "nine out of ten doctors agree, Whizzie cigarettes are good for you".

That is: human behavior is the sum of human knowledge over time and as an historical phenomenon, can never be properly accounted for from inside the same historical box. Nine out of ten doctors may have agreed that Whizzie cigarettes are good for you in 1948 but they were later proven wrong.

How on EARTH could you "prove" that Agile or lightweight is good for you? Yet as a human subject engaged in coding, you KNOW, as do I, that lightweight (dunno about Agile) is good for you.

You simply depend less upon empirical facts about the platform.

The ideal, null, limiting case of elegance would be software that built its own computer. This is unattainable. But imagine software running on a space station with absolutely minimal facilities assumed because it is intended to work for millenia on low power.

It would assume almost nothing.

People naturally aspire to lightweight. Shakespeare's Cleopatra says at the death, "I am air and fire: my lesser elements I give to baser life". Old IBM mainframes WERE Satanic, they WERE Saturnine, they ate their human children, because we DON'T want to depend on corporate products.

Your esthetic sense rebels against that-which-is-heavyweight, a large amount of code that is which must be poured into the user's network before your software can start up.

This isn't religion. It is esthetics, and in the absence of a scientific test esthetics does matter.

I am unmoved by any argument that claims that a lotta programmers like the favored platform. Nine out of ten doctors, after all, agreed that Whizzie cigarettes are good for you.

Programmers with esthetic sense are NOT religious fanatics, and I fear that Yegge concludes that they are only because owing to the First amendment, about the only people left in America, whose subjectivity has not been destroyed by workplace moronization and infantilization, drugs and computer games, are religious people.

As Dijkstra said, "elegance is not a luxury". And, the experience of having to wait for an ActiveX control to install so that one can perform a trivial task, is an UGLY experience. There's quite enough UGLINESS in modern life and to not want to add to it isn't religion, it is humanity.

Jeff Atwood on Software Development as Religion

Daniel Read — Tue, 10 Oct 2006 08:57:55 -0700

I'm sure there are many good posts out there responding to Steve Yegge this week, but this one by Jeff Atwood at CodingHorror.com struck me today as being particularly good.

Dan

Yegge Agile Follow-Up

Daniel Read — Mon, 09 Oct 2006 10:45:50 -0700

For those who developed an interest in the Agilism/Googlism part of the threads linked above, Steve Yegge has posted a follow-up to his original post. Given that Steve has plenty of thunder to offer, I won't steal much of his by quoting a few paragraphs from his post:

Agile is a niche, a market minority, almost an aberration, really. It just happens to have a lot of marketing, because it opened up a hitherto entirely untapped new market for snake oil in the tech sector. Consultants are making money hand over fist by extending their contracts with credulous clients who are told they don't have the process "quite right" yet, but they're almost there!

Agile wouldn't be a big deal, except the Agile camp is really loud. Annoyingly so. Loud enough to start interfering with regular developers' work. Which is why I had to speak up. Agile was the Mystery Topic that gave me Blogger's Block for nearly 2 months. At some point, though, I just couldn't take it anymore. There were critics here and there, sure, but none of them were as loud as the Agile folks.

So I yelled as loud as I could: loud enough to make it onto Slashdot, even! Which is exactly what I wanted. I only had one high-level takeaway, one marketing message I wanted to get out to developers everywhere: It's OK to say No to Agile. That's it. Nothing more complicated than that. But the Church of Agile was getting so powerful that it was becoming increasingly acceptable to criticize people in the workplace for not being Agile.

Thanks for reading,
Dan

Donna, I agree...

Edward G Nilges — Fri, 06 Oct 2006 10:53:42 -0700

It is ABSURD to fire people until you have a great team, it's being a monkey with a typewriter, and the ease of termination in America results in a complete lack of due dilgence at the other end, in hiring.

Yet this is the operating mode of many software firms, and has become the philosophy of the Bush administration. Coalition Provisional Authority head in 2003, Paul Bremer, a product of business school, decided to "lay off" the entire Iraqi army and civil service, and this created today's Sunni insurgency.

A software manager isn't going to create such a massive problem except in the case of the "disgruntled employee" (and workplace violence is a growing problem for the simple reason that doing things such as telling a man he's a "diseased limb" is inhuman).

The entire philosophy of firing people is cockeyed. It rests upon a primitive view of personality in which personality is fixed over time and unaffected by inputs, or a view that is only slightly more sophisticated: that a job is a fixed thing for which a person is, or is not, a "good fit".

Japanese managers have discovered that if an employee is underperforming her performance can often be improved by promoting him or her! This finding is similar to the "Hawthorne effect" discovered in the USA in the 1920s, where performance improved when management gave a sign that it cared about the physical environment of Western Electric workers.

Now, consider that the Japanese practice is completely unheard of in the USA. The fact that it is a complete non-starter in American management circles means that a practice, with rational results, is not considered...and this means that we need to examine the IRRATIONAL grounds of the American system of employment at will, not allowing its defenders to equate rationality with American management practice.

The irrationality is that in the absence of an agreed upon religion or philosophy, decision procedures in some societies converge to sadistic and masochistic impulses. The Marquis de Sade, an aristocrat driven mad by the events of the French Revolution, felt that decisions would ultimately be based on sado-masochistic impulses in a mass orgy.

Some time ago, a Rutgers university sociologist wrote a book which I read, but can't find tonight on Amazon, and in that she analyzed sado-masochistic impulses in business.

One of her examples was the job search. We all know well that the employee who pretends, or actually has, multiple offers, or who doesn't need the job and/or expresses reluctance to accept the offer, can get the offer, or a better offer.

The usual relationship between the hiring manager and the candidate is mild sadism going from the manager to the masochistic candidate, who acts in a somewhat servile fashion while the manager questions, in many real-life instances, not only the employee's skills but even his sense of self-worth.

A typical game: the employee has a stellar degree in computer science, but is told that the degree that he's spent thousands of dollars on, and on which he's worked so hard, is worthless because "we" know about the "real" world. Upon investigation, "reality" consists in fooling the user, cutting corners, and maintaining old code...including the black art of keeping the bugs in to avoid disrupting the user's business process, in which those bugs are features.

Therefore, the employee's game has to compensate for this sadistic distortion of a reality, wherein the algorithms and tools used in good university are usually superior to the algorithms and the tools usually in use at the typical "shop", with a sadistic counter-reality in which the employee is actually encouraged, by the job-finding gurus, to exagerrate his abilities.

Well, this would theoretically predict that there's two types of MIS employees in the majority: overqualified people bored outa there skull to the extent that they can't meet expectations, and underqualified people, who can't meet expectations either.

But this is exactly what we have. Very few MIS people, in my observation over thirty years, are happy campers, and their skills are either completely out of scale (the PhD maintains a compiler) or completely insufficient (the Visual Basic programmer can't maintain a special-purpose compiler for medical billing, so he bypasses it and adds "hard" code, circumventing the user interface to the compiler).

Both wind up to be "diseased limbs".

Better GOVERNMENT oversight of the employment relationship would not force people, who can't predict the performance of the hire, or their own performance as hires, to play sado-masochistic head games. This GOVERNMENT oversight of hiring exists in many of the countries to which software development is being outsourced for the same reason costs in many of these countries are lower because they have basic health care!

It really shouldn't be legal to fire heads of families without considerably more due diligence because the "paperwork" would force the manager to confront why the employee is being fired.

Too often, employees are terminated because they are not "a good fit" with an incompetent team.

My views are more radical than yours, Donna. But as it is, termination is a reality in programming (many more people are fired than admit it because of shame). It's the "elephant in the living room" and in other societies, a person's sense of self-worth and his or her ability to support his family is not under continual threat.

Mentoring Developers

Donna L Davis — Fri, 06 Oct 2006 06:06:22 -0700

I found this track fascinating...especially reading the comments of various strongly opinionated readers. What made it even more interesting to me is that I have been agonizing about some of the same issues myself. On the one hand, I feel that as a manager of a group of developers, if one of them is not performing, it is a sign of my failure. Surely there is something more I should be able to do to encourage/prod/lead. I don't want to give up on someone. Yet, clearly there are those who consume 90% of your energy and deliver 10% and those who consume 10% and deliver 90%. What makes it harder for everyone, is that often the performers and non-performers make about the same amount of money. Granted, the contributor has room for upward movement. Some are quick to say that the answer is to let the person go who is not performing...cut them off like a diseased limb. On one level coworkers may appreciate that...no longer having to pull the extra dead weight. On the other hand, often these non-performers are not slack, sorry people. Often they are very nice people who clearly try very hard, often working late, and are dependable and dedicated...even friends with the other performers and worked with the organization for a number of years. They *want* to achieve and *try* but may just not have it in them. To see someone like this severed may cause the rest of the team to feel (like the postings indicated) like a commodity that is bought and sold like cattle. My greatest hope is to continue trying to find the right type of work for someone like this where they can be successful. As we know, software development isn't all about coding.

Only the Best?

Daniel Read — Thu, 05 Oct 2006 16:20:42 -0700

On his popular blog, The Mindset, Lidor Wyssocky has written some worthwhile posts recently that riff on the intersection of two current hot topics in blogland, one relating to methodology, the other to software team composition--both of which have an underlying theme of "What is the best way to enable success in building software?"