Category Archives: Technology

The Use/Build Fallacy

Working in the language space, especially in language design, you frequently encounter people who fall victim to what I call the “Use/Build Fallacy.” It goes something like this:

Because I know how to use something, I know how to build it as well.

This fallacy is best illustrated by a story I heard from a friend who’s a teacher (another profession that frequently has to deal with this). She was teaching middle-school when teacher conferences rolled around. Talking to the father of one of her students, she explained to him that his daughter was having a lot of trouble in English class and that, based on her observations of how hard the daughter was working, she was pretty sure that the daughter had some sort of language learning disability. She therefore strongly recommended that he take his daughter to an expert to get tested, and that she be tutored by someone trained to deal with the specific kind of learning disability. The father was nonplussed, mainly because he didn’t like the idea that all this would cost him money. “Can’t you just help her more in class?” he asked. My friend explained that she was helping her all she could, but she wasn’t an expert in diagnosing learning disabilities and his daughter really needed to see someone who had the appropriate training.

After a bit of back-and-forth, the father finally got exasperated and said, “Fine, I’ll just tutor her myself! I mean, how hard could it be? I went to school!” My friend then shot back, “Look, you’re a general contractor, right? What would you think if I came to you and said, ‘I don’t need you to build my house—I’ve lived in a house before, so how hard could it be to build one myself?” This, finally, stumped him. I’m not sure whether he actually got his daughter the help she needed, but the story stuck with me because my friend’s response is the perfect distillation of the Use/Build Fallacy.

Note that I’m not saying that just because you’re not an expert on something you can’t have an opinion. I may not know how to build a house, but that doesn’t mean I have nothing to say to the contractors if I decide to do some renovations on my house. Not falling prey to the fallacy, though, means that I always keep a healthy respect for the expert in a field—as long as they truly seem to know what they’re talking about. (I hear this from friends who are architects all the time—they get hired by someone to build or renovate a house for them, and then their client spends all their time endlessly arguing with everything they do. Why bother to hire an expert if you think you already know how to do it yourself?)

I try to remember this myself every time I encounter some aspect of some programming language that I don’t like. Right now, I’m neck-deep in C++ code and it’s tempting to spend all my time kvetching about how how horrible a job Bjarne has done over the years. And then I try to remember—even as someone who’s actually built a language—that this stuff is hard. A language of any complexity has a huge number of moving parts, all of which interact with each other in an unpredictable manner. Historical choices can come back to bite you in all sorts of unexpected ways. Oftentimes all you have are a bunch of imperfect choices, and you have to simply pick the least bad of them all. And then you get to sit there and listen to everyone on the sidelines complain about how horrible a job you’ve done and how they could do it so much better than you because, hey, they’ve used a programming language before.

So I try to temper my complaints with a little humility, and remember how much different building is from using.

Black Hole Projects

4 Replies

OK, so I may have reset my blog, but there were some interesting posts that probably shouldn’t disappear totally down the memory hole. This is one of them, which I am rescuing from back in 2004 because of it’s continuing relevance. It seems that six months can’t go by without something I hear making me think of this. Edited from the original for clarity and to bring it up-to-date.

Many, many years ago, Steve Maine took the opportunity to reminisce about a project at Microsoft that was being worked on while he was an intern. He says:

[ When I was an intern… ] there was this mythical project codenamed “Netdocs”, and it was a black hole into which entire teams disappeared. I had several intern friends who got transferred to the Netdocs team and were never heard from again. Everyone knew that Netdocs was huge and that there were a ton of people working on it, but nobody had any idea what the project actually did.

I also knew a few people who invested quite a few years of their lives into “Netdocs” and it got me thinking about the phenomenon of “black hole projects” at Microsoft (and elsewhere, I’ll wager). There was one I was very close to very early in my career that I managed to avoid, many others that I just watched from afar, and one or two that I got dragged into despite my best intentions. I can’t really talk about most of them since most never saw the light of day, but it did get me thinking about the peculiarly immutable traits of a black hole project. They seem to be:

They must have absurdly grandiose goals. Something like “fundamentally reimagine the way that people work with computers.” Nobody, including the people who originate the goals, has a clear idea what the goals actually mean.
They must involve throwing out some large existing codebase and rewriting everything from scratch, “the right way, this time.”
They must have completely unrealistic deadlines. Often this is because they believe that they can rewrite the original codebase in much, much less time than it took to write that codebase in the first place.
They must have completely unrealistic beliefs about compatibility. Usually this takes the form of believing you can rewrite a huge codebase and preserve all of its little quirks without a massive amount of extra effort.
They are always “six months” from from major deadlines that never seem to arrive. Or, if they do arrive, another milestone is added on to the end of the project to compensate.
They must consume huge amounts of resources, sucking the lifeblood out of one or more established products that make significant amounts of money or have significant market share.
They must take over any group that does anything that relates to their absurdly broad goals, especially if that group is small, focused, has modest goals, and actually has a hope of shipping in a reasonable timeframe.
They must be prominently featured as demos in public settings such as company meetings, all-hands, conferences, etc. to the point where people groan “Oh, god, not another demo of this thing. When is it ever going to ship?”
They usually are prominently talked up publicly by high level executives for years before dying a quiet death.
They usually involve “componentizing” some monolithic application or system. This means that not only are you rewriting a huge amount of code, you’re also splitting it up across one or more teams that have to all seamlessly work together.
As a result of the previous point, they also usually involve absolutely massive integration problems as different teams try madly to get their components working with each other.
They usually involve rewriting the application or system on top of brand-new technology that has not been proven at a large scale yet. As such, they get to flush out all the scalability problems with the new technology.
They are usually led by one or more Captain Ahabs, madly pursuing the white whale with absolute conviction, while the deckhands stand around saying “Gee, that whale looks awfully big. I’m not sure we can really take him down.”
Finally, 90% of the time, they must fail and die a flaming death, taking down other products with it (or at least severely damaging them). If they do ship, they must have taken at least 4-5 years to ship and be at least 2 years overdue.

It’s kind of frightening how easy it is to come up with this list – it all kind of just poured out. Looking back over 19 years at Microsoft, I’m also kind of frightened at how many projects this describes. Including some projects that are ongoing at the moment…

You should also follow me on Twitter here.

Seven Rules for Beginning Programmers

59 Replies

A little while ago Phil Wadler posted “VS Naipaul’s Rules for Beginners,” listing the famous author’s seven rules for beginning writers. Upon reading them it occurred to me that, with a little adaptation, they could equally apply to beginning programmers. So, with apologies to Mr. Naipaul, here are my “Rules for Beginners:”

Do not write long procedures. A procedure should not have more than ten or twelve lines.
Each procedure should have a clear purpose. It should not overlap in purpose with the procedures that went before or come after. A good program is a series of clear, non-overlapping procedures.
Do not use fancy language features. If you’re using something more than variable declarations, procedure calls, control flow statements and arithmetic operators, there is something wrong. The use of simple language features compels you to think about what you are writing. Even difficult algorithms can be broken down into simple language features.
Never use language features whose meaning you are not sure of. If you break this rule you should look for other work.
The beginner should avoid using copy and paste, except when copying code from one program they have written to a new one they are writing. Use as few files as possible.
Avoid the abstract. Always go for the concrete. [Ed. note: This one applies unchanged.]
Every day, for six months at least, practice programming in this way. Short statements; short, clear, concrete procedures. It may be awkward, but it’s training you in the use of a programming language. It may even be getting rid of the bad programming language habits you picked up at the university. You may go beyond these rules after you have thoroughly understood and mastered them.

You should also follow me on Twitter here.

Murphy’s Computer Law

5 Replies

A long time ago, my family took a trip to Expo `86 in Vancouver, with stop offs in San Francisco and Los Angeles. In LA, we went on the Universal studio tour, something which I basically have no memory of. I did get a memento, though-a poster entitled “Murphy’s Computer Law” with a bunch of humorous computing “laws” on it. This poster went up in my room, accompanied me to college and has been in most of my offices at Microsoft. However, a few years ago, a corner ripped off in a move. Then while it was sitting around waiting to be repaired, it got a bit stained. And then I realized just how dated and ratty the thing looked. So, I figured it’s time to retire it. However, I would like to hang on to the “laws” since some of them are are still quite pertinent, even if some are quite outdated. So here they are, on my “permanent record:”

Murphy’s Computer Law:

Murphy never would have used one.
Murphy would have loved them.

Bove’s Theorem: The remaining work to finish in order to reach your goal increases as the deadline approaches.

Brooks’ Law: Adding manpower to a late software project makes it later.

Canada Bill Jones’ Motto: It’s morally wrong to allow na‹ve end users to keep their money.

Cann’s Axiom: When all else fails, read the instructions.

Clarke’s Third Law: Any sufficiently advanced technology is indistinguishable from magic.

Deadline-Dan’s Demo Demonstration: The higher the “higher-ups” are who’ve come to see your demo, the lower your chances are of giving a successful one.

Deadline-Dan’s Demon: Every task takes twice as long as you think it will take. If you double the time you think it will take, it will actually take four times as long.

Demian’s Observation: There is always one item on the screen menu that is mislabeled and should read “ABANDON HOPE ALL YE WHO ENTER HERE.”

Dr. Caligari’s Come-back: A bad sector disk error occurs only after you’ve done several hours of work without performing a backup.

Estridge’s Law: No matter how large and standardized the marketplace is, IBM can redefine it. [ed, later "Microsoft", now "Apple," I guess]

Finagle’s Rules:

To study an application best, understand it thoroughly before you start.
Always keep a record of data. It indicates you’ve been working.
Always draw your curves, then plot the reading.
In case of doubt, make it sound convincing.
Program results should always be reproducible. They should all fail in the same way.
Do not believe in miracles. Rely on them.

Franklin’s Rule: Blessed is the end user who expects nothing, for he/she will not be disappointed.

Gilb’s Laws of Unreliability:

At the source of every error which is blamed on the computer you will find at least two human errors, including the error of blaming it on the computer.
Any system which depends on human reliability is unreliable.
Undetectable errors are infinite in variety, in contrast to detectable errors, which by definition are limited.
Investment in reliability will increase until it exceeds the probable cost of errors, or until someone insists on getting some useful work done.

Gummidge’s Law: The amount of expertise varies in inverse proportion to the number of statements understood by the general public.

Harp’s Corollary to Estridge’s Law: Your “IBM PC-compatible” computer grows more incompatible with every passing moment.

Heller’s Law: The first myth of management is that it exists.

Hinds’ Law of Computer Programming:

Any given program, when running, is obsolete.
If a program is useful, it will have to changed.
If a program is useless, it will have to be documented.
Any given program will expand to fill all available memory.
The value of a program is proportional to the weight of its output.
Program complexity grows until it exceeds the capability of the programmer who must maintain it.
Make it possible for programmers to write programs in English, and you will find that programmers cannot write English.

Hoare’s Law of Large Programs: Inside every large program is a small program struggling to get out.

The Last One’s Law of Program Generators: A program generator creates programs that are more “buggy” than the program generator.

Meskimen’s Law: There’s never time to do it right, but always time to do it over.

Murphy’s Fourth Law: If there is a possibility of several things going wrong, the one that will cause the most damage with be the one to go wrong.

Murphy’s Law of Thermodynamics: Things get worse under pressure.

Ninety-Ninety Rule of Project Schedules: The first ninety percent of the task takes ninety percent of the time, and the last ten percent takes the other ninety percent. [ed: words to live by]

Nixon’s Theorem: The man who can smile when things go wrong has thought of someone he can blame it on.

Nolan’s Placebo: An ounce of image is worth a pound of performance.

Osborn’s Law: Variables won’t, constants aren’t.

O’Toole’s Commentary on Murphy’s Law: Murphy was an optimist.

Peer’s Law: The solution to a problem changes the problem.

Rhode’s’ Corollary to Hoare’s Law: Inside every complex and unworkable program is a useful routine struggling to be free.

Robert E. Lee’s Truce: Judgment comes from experience; experience comes from poor judgment.

Sattinger’s Law: It works better if you plug it in.

Shaw’s Principle: Build a system that even a fool can use, and only a fool will want to use it. [ed: also known as "Bob's Law"]

SNAFU Equations:

Given an problem containing N equations, there will be N+1 unknowns.
An object or bit or information most needed will be least available.
Any device requiring service or adjustment will be least accessible.
Interchangeable devices won’t.
In any human endeavor, once you have exhausted all possibilities and fail, there will be one solution, simple and obvious, highly visible to everyone else.
Badness comes in waves.

Thoreau’s Theories of Adaptation:

After months of training and you finally understand all of a program’s commands, a revised version of the program arrives with an all-new command structure. [ed: also known the "Office Principle"]
After designing a useful routine that gets around a familiar “bug” in the system, the system is revised, the “bug” is taken away, and you’re left with a useless routine.
Efforts in improving a program’s “user friendliness” invariably lead to work in improving user’s “computer literacy.”
That’s not a “bug”, that’s a feature!

Weinberg’s Corollary: An expert is a person who avoids the small errors while sweeping on to the grand fallacy.

Weinberg’s Law: If builders built buildings the way programmers write programs, then the first woodpecker that came along would destroy civilization.

Zymurgy’s First Law of Evolving System Dynamics: Once you open a can of worms, the only way to recan them is to use a larger can.

Wood’s Axiom: As soon as a still-to-be-finished computer task becomes a life-or-death situation, the power fails.

The Five Levels of Incompetence

2 Replies

In my “Learning and Teaching” post last week, I talked about the different stages of learning, from “unconscious incompetence” up to “unconscious competence.” It occurred to me today, though, that there really are different levels within those levels and, in particular, there are some very distinct levels of incompetence that I’ve encountered in my nearly (yikes!) two decades of working in the industry. The reason why the levels of incompetence are somewhat more important than the various levels of competence, it seems to me, is that incompetent people are often a very real threat to the stability of teams that they work in, while competent people usually aren’t.

The five levels of incompetence are, in increasing order of danger:

Level 1: The N00b.

There’s not much to say about the n00b since, let’s face it, we’ve all been there. Hiring n00bs is unavoidable in most situations. Being a n00b is a basic fact of life.

Danger: Low, assuming that they are properly sandboxed or are experienced enough to sandbox themselves. (Otherwise, they’re likely to hit “launch” instead of “lunch” and then you’re really in trouble.)

Level 2: Out of their depth.

Typically, this is someone who’s not really incompetent in general but who’s just been pushed up to a level of responsibility beyond their capabilities (a.k.a. a victim of the Peter Principle.). I’ve seen this happen most often in situations where a senior, experienced person leaves the team and the leadership decides that they have to put someone equally senior in their place regardless of whether that person can, you know, actually do the job. So they pluck someone who’s senior but not as experienced and plops them into the departing person’s chair.

Danger: Moderate. Because the person isn’t completely incompetent, they tend to be able to give the appearance of competence and avoid leading the team totally off the rails but usually end up leading the team in circles. So the team doesn’t make any forward progress and people eventually wise up and leave.

Level 3: Dumb and Dumber.

Now we start getting into the fun levels. This person is just plain incompetent, someone placed into a totally inappropriate position for them (which, for the most part, is going to be any position). I’ve actually encountered very few instances of this in my career, and it usually happens when someone transfers between two wildly different kinds of jobs. That tends to mask, for a little while anyway, their completely lack of ability to actually do anything under the guise of just being a n00b.

Danger: High to moderate. It really depends on how fast everyone figures out just how incompetent the person is-usually, truly incompetent people get shunted aside as soon as everyone figures out what’s going on. If that takes too long, competent people tend to get pissed off and leave.

Level 4: Bozo.

The difference between a Bozo and a Dumb and Dumber is that a Bozo is Dumb and Dumber who thinks he is competent. Bozo’s tend to believe that they are as good or better than everyone else, deserve special treatment, and that their genius is being under-rewarded. And they ignore the fact that they have absolutely no idea what they are doing.

My best Bozo story is a Program Manager that I worked with a long time ago. I implemented a feature he specified. He entered a bug saying that the feature didn’t work correctly. I resolved the bug “by design” after I verified that the feature worked exactly the way he specified it. He then came to my office and started to argue with me that I shouldn’t have resolved the bug “by design,” because the feature didn’t work correctly. Finally, I pulled out a copy of his specification and pointed at the paragraph that said exactly how the feature should work. He then got totally exasperated with me and started ranting that I was supposed to implement the feature “the way he wanted the feature to work, not the way he specified it!”

Danger: High. Bozos are always on the lookout to get ahead (in line with their great abilities), so they often manage to worm their way in to management positions. They then tend to lead the team the way Mr. Toad drives motorcars: careening all over the road until they finally end up in the ditch. Bozo’s are adept at bringing down even the most experienced team in a surprisingly short amount of time.

Level 5: Evil Genius.

I was debating whether this is even a level of incompetence at all, because in many ways Evil Geniuses are not incompetent people. Quite the contrary, they are often quite adept at many things, including manipulation, spin, intimidation, self-aggrandizement, and sucking up. But I think in a deep sense, Evil Geniuses are just a more highly evolved form of Bozos because the end result tends to be the same: the team blows up in a very spectacular way. However, while a Bozo usually does this in a totally oblivious way (“What happened?”), it’s often all part of an Evil Genius’s plan to use the force of the explosion to propel them ever higher. These are the kind of guys who end up running major corporations and then running them totally into the ground. And then jumping ship to run an even bigger corporation. But, at the core, I think that Evil Geniuses act this way because they couldn’t actually figure out how to do things in an above-board manner. Thankfully, I’ve met very few Evil Geniuses in my day. And those I have met, I’ve been able to largely avoid.

There are only three types of programmers in the world…

10 Replies

..and they are:

Programmers who want to write an operating system
Programmers who want to write a compiler
Programmers who want to write a database

It’s not that every programmer ever actually works on one of these, just that every programmer seems to dream of doing one of these things. It’s the primary reason why things like Linux exist. Yes, open source, blah, blah, blah, OS choices, blah, blah, blah, evil Microsoft, blah, blah, blah. But I would bet my bottom dollar that 9 out of 10 of the people donating their valuable time to the Linux project do so not because they want an alternative to Windows but because they always dreamed of being OS hackers. It’s also why there are so many damn programming languages out there, all the people who sit around dreaming of being, I don’t know, James Gosling or something.

(I think with the advent of the Internet, it’s likely that there’s now a fourth kind of programmer who wants to write websites, but I’m not totally sure about that yet.)

The interesting thing about these categories is that the Venn diagram tends, in my experience, to be pretty distinct-most “data” guys aren’t also “language” guys, and most “language” guys aren’t also “OS” guys, and so on. My theory is that it’s like the parable of the blind men and the elephant: although we all grapple with basically the same set of problems, each kind of programmer grapples with a different aspect of it.

I say all this because although I started out working in databases, it’s clear to me that I’ve always been a “language” guy. In college, I did so-so in the OS course and never touched a database course (I’m not even sure they were offered), but my compiler course netted me a special letter of commendation from the professor (the only one I ever got). Anyway, now I’m back in the “data” world as an even more confirmed “language” guy and the most interesting thing is how many of the problems are the same, but the way they’re conceptualized, handled, or even talked about, are different from what I’ve been used to working on programming languages. It’s kind of. refreshing to see things in a different light. More on that soon.

I Heart Beagle Brothers