Big Ball of Mud Design Pattern

In "Big Ball of Mud", Brian Foote and Joseph Yoder propose that the default (and most common) software architecture in use is the "Big Ball of Mud" pattern and go on to discuss six additional patterns and activities that it gives rise to: "Throwaway Code", "Piecemeal Growth", "Keep it Working", "Shearing Layers", "Sweep it Under the Rug" and "Reconstruction".

Their original article is located here and can be downloaded in PDF form here: Big Ball of Mud by Brian Foote and Joseph Yoder [PDF download]

I have picked out what I think are the highlights of their article ...


Big Ball of Mud ... alias Shantytown or Spaghetti Code

Shantytowns are usually built from common, cheap materials with simple tools and using unskilled labor. The construction and maintenance of the shantytown is labor intensive, and there is little or no labor specialization - each builder must be a jack of all trades. There's no overall planning, or regulation of future growth.

Too many of our software systems are, architecturally, little more than shantytowns. Investment in tools and infrastructure is often inadequate and the tools that are used are primitive. Parts of the system grow unchecked, and the lack of architecture and planning allows problems in one part of the system to erode and pollute adjacent portions. Deadlines loom like monsoons, and architectural elegance seems unattainable.

The time and money to chase perfection are seldom available and there is a survival at all costs attitude, to do what it takes to get the software working and out the door on time. The biggest cost borne by the Big Ball of Mud development is the lack of a decent architecture.


Common features of Big Ball of Mud Code

• Data structures are haphazardly constructed, or non-existent.
• Everything talks to everything else.
• Important state data is global.
• State data is passed around though Byzantine back channels that circumvent the system's original structure.
• Variable and function names are uninformative and misleading.
• Functions use global variables extensively, as well as long lists of poorly defined parameters.
• Functions themselves are lengthy and convoluted, and perform several unrelated tasks.
• Code duplication.
• The flow of control is hard to understand, and difficult to follow.
• The programmer’s intent is next to impossible to discern.
• The code is simply unreadable, and borders on indecipherable.
• The code exhibits the unmistakable signs of patch after patch at the hands of multiple maintainers, each of whom barely understood the consequences of what he or she was doing.
• Did we mention documentation? What documentation?

Working with Big Ball of Mud Code

Some software engineers come to regard life with the Big Ball of Mud as normal and become skilled at learning to navigate these quagmires, and guiding others through them. Over time, this symbiosis between architecture and skills can change the character of the organization itself, as swamp guides become more valuable than architects.

As per Conway's Law, architects depart in futility, while engineers who have mastered the muddy details of the system they have built, prevail. The code becomes a personal fiefdom, since the author care barely understand it anymore, and no one else can come close. Once simple repairs become all day affairs, as the code turns to mud.


Throwaway Code ... alias Quick Hack or Protoduction

While prototyping a system, you're normally unconcerned with how elegant or efficient your code is. You plan that you will only use it to prove a concept and once the prototype is done, the code will be thrown away and written properly. As the time nears to demonstrate the prototype, the temptation to load it with impressive but utterly inefficient realizations of the system’s expected eventual functionality can be hard to resist. Sometimes, this strategy can be a bit too successful. The client, rather than funding the next phase of the project, may slate the prototype itself for release.

This quick-and-dirty coding is often rationalized as being a stopgap measure. More often than not, the time is never found for this follow up work. The code languishes, while the product flourishes. It becomes a protoduction - a prototype that gets used in production.

Once it becomes evident that the throwaway code is going to be around for a while, you can turn your attention to improving its structure, either through an iterative process of Piecemeal Growth, or via a fresh draft, as discussed in the Reconstruction pattern below.


Piecemeal Growth ... alias Refactoring

Successful software attracts a wider audience, which can, in turn, place a broader range of requirements on it.

When designers are faced with a choice between building something elegant from the ground up, or undermining the architecture of the existing system to quickly address a problem, architecture usually loses.

In the software world, we deploy our most skilled, experienced people early in the lifecycle. Later on, maintenance is often relegated to junior staff, and resources can be scarce. The so-called maintenance phase is the part of the lifecycle in which the price of the fiction of master planning is really paid. It is maintenance programmers who are called upon to bear the burden of coping with the ever widening divergence between fixed designs and a continuously changing world.

Piecemeal growth can be undertaken in an opportunistic fashion, starting with the existing, living, breathing system, and working outward, a step at a time, in such a way as to not undermine the system’s viability. You enhance the program as you use it. Massive system-wide changes are avoided - instead, change is broken down into small, manageable chunks.


Keep It Working ... alias Continuous Integration

Businesses become critically dependent on their software and computing infrastructures. There may be times where taking a system down for a major overhaul can be justified, but usually, doing so is fraught with peril. Therefore, do what it takes to maintain the software and keep it going. Keep it working.

This approach can be used for both minor and major modifications. Large new subsystems might be constructed off to the side, perhaps by separate teams, and integrated with the running system in such a way as to minimize disruption.

A development build of each product can be performed at regular intervals, such as daily or even more often via an automated build tool. Another vital factor in ensuring a system's continued vitality is a commitment to continuous testing, which can be integrated into the automated build process.


Shearing Layers

Software never stands still. It is often called upon to bear the brunt of changing requirements, because, being as that it is made of bits, it can change.

Over time, the software's frameworks, abstract classes, and components come to embody what we've learned about the structure of the domains for which they are built. More enduring insights gravitate towards the primary structural elements of these systems and change rarely. Parts which find themselves in flux are spun out into the data, where users can interact with them. Software evolution becomes like a centrifuge stirred by change. The layers that result, over time, can come to a much truer accommodation with the forces that shaped them than any top-down planning could have devised.


Sweeping It Under The Rug

At first glance, a Big Ball of Mud can inspire terror and despair in the hearts of those who would try to tame it. The first step on the road to architectural integrity can be to identify the disordered parts of the system, and isolate them from the rest of it.

Overgrown, tangled, haphazard spaghetti code is hard to comprehend, repair, or extend, and tends to grow even worse if it is not somehow brought under control. If you can’t easily make a mess go away, at least cordon it off. This restricts the disorder to a fixed area, keeps it out of sight, and can set the stage for additional refactoring.


Reconstruction ... alias Total Rewrite

One reason to start again might be that the previous system was written by people who are long gone. Doing a rewrite provides new personnel with a way to reestablish contact between the architecture and the implementation. Sometimes the only way to understand a system it is to write it yourself. Doing a fresh draft is a way to overcome neglect. Issues are revisited. A fresh draft adds vigor. You draw back to leap. The quagmire vanishes. The swamp is drained.

When a system becomes a Big Ball of Mud, its relative incomprehensibility may hasten its demise, by making it difficult for it to adapt. It can persist, since it resists change, but cannot evolve, for the same reason. Instead, its inscrutability, even when it is to its short-term benefit, sows the seeds of its ultimate demise.


The above are highlights from the original article, which is located here ... Big Ball of Mud by Brian Foote and Joseph Yoder [PDF download]


Some further reading on Programmers.StackExchange ...

I've inherited 200K lines of spaghetti code — what now?
How to convince my boss that quality is a good thing to have in code?
How to keep a big and complex software product maintainable over the years?
Techniques to re-factor garbage and maintain sanity?
When is code “legacy”?
What is negative code?


Some related books ...

Working Effectively With Legacy Code ... by Michael Feathers [Amazon]
Refactoring: Improving the Design of Existing Code ... by M. Fowler, K. Beck, et al. [Amazon]
Design Patterns: Elements of Reusable Object-Oriented Software ... by the Gang of Four [Amazon]
Patterns of Enterprise Application Architecture ... by Martin Fowler [Amazon]
Domain Driven Design: Tackling Complexity in the Heart of Software ... by Eric Evans [Amazon]
Head First Design Patterns by E. Freeman, E. Freeman, et al. [Amazon]

Follow @dodgy_coder

Subscribe to posts via RSS

The first right answer is the only answer

Reading through David B. Stewart's paper entitled "Twenty-Five Most Common Mistakes with Real-Time Software Development" (PDF, 131 KB).

There's an interesting nugget of advice at number 8, "The first right answer is the only answer";

Inexperienced programmers are especially susceptible to assuming that the first right answer they obtain is the only answer. Developing software for embedded systems is often frustrating. It could take days to figure out how to set those registers to get the hardware to do what is wanted. At some point, Eureka! It works. Once it works the programmer removes all the debug code, and puts that code into the module for good. Never shall that code ever change again, because it took so long to debug, nobody wants to break it.

Unfortunately, that first success is often not the best answer for the task at hand. It is definitely an important step, because it is much easier to improve a working system, than to get the system to work in the first place. However, improving the answer once the first answer has been achieved seems to rarely be done, especially for parts of the code that seem to work fine. Indirectly, however, a poor design that stays might have a tremendous effect, like using up too much processor time or memory, or creating an anomaly in the timing of the system if it executes at a high priority.

As a general rule of thumb, always come up with at least two designs for anything. Quite often, the best design is in fact a compromise of other designs. If a developer can only come up with a single good design, then other experts should be consulted with to obtain alternate designs.

As David suggests, when dealing with complex, mission critical or concurrent sections of code, that first success is often not the best solution. Weeks later, you might find that its not performing as well as it should be in production and you have to revisit the section of code again looking for a better solution. But the best time to develop a better solution was back when the job was fresh in your mind, when you developed the original solution.

So he suggests adapting this into a new practice:

As a general rule of thumb, always come up with at least two designs for anything.

 There's multiple problems with this advice:

• YAGNI, you ain't gonna need it. Develop an extra solution only if and when you need it; don't create extra work where it might not be needed.
• Premature optimization is the root of all evil. Knowing when you need to optimize a solution further is almost impossible to do at design time, before any benchmarking or code profiling have been done.
• Lastly, if your solution passed the unit test, yet fails further down the line (in production for example), then that suggests there was a problem with your unit test, and not with your solution. If required, you should add some specific concurrency and/or performance testing to your unit test. This will mean you can then optimize your code, while maintaining a TDD approach.

I notice the original article is actually from 1999, which I believe is before test driven development came into prominence. I think this particular piece of advice ("come up with two designs for anything") might have been ok for some projects back then, but would now be considered flawed and certainly not advisable.

  

Follow @dodgy_coder

Subscribe to posts via RSS

Q&A With Nine Great Programmers

This Q&A article is originally from 2006, published on the blog of Polish developer Jaroslaw "sztywny" Rzeszótko (AKA "Stiff").  That original blog post is no longer online. Gratefully republished with his permission.

On a hot, boring afternoon I got an idea. With the help of publicly accessible e-mail addresses, I asked 10 questions to a bunch of programmers that I consider very interesting people and I respect them for various things they created. Coming out with the questions was a 5 minute job for me - these are things I would ask about if I could speak with them personally for, let’s say, 10 minutes, and I didn’t have time for thinking too much. The last two question don’t have anything to do with programming, this is simply something I like to know about everyone I talk to, lets say that’s my hobby. Not everyone wanted to answer them, and that’s fine. It was the first "interview" I ever made, so I also made some mistakes, which went out as people started answering. But despite this, I learnt a lot of interesting stuff, so it was definitely a valuable experience.

Not everyone responded to my e-mail, not everyone agreed to answer the questions, maybe I will also get some answers after I published this, I didn’t have the patience to wait longer, so new things may appear here over time (Update: Bjarne Stroustrup was added on 03.08.2006).
-- Jaroslaw

Starring

Linus Torvalds - The Linux kernel author.

Dave Thomas - Author of the "Pragmatic Programmer", "Programming Ruby" and other great books about programming. One can read his mainly programming-related thoughts here.

David Heinemeier Hansson - Author of the Rails Framework - the new hot web development framework. He has a blog here.

Steve Yegge - Probably one of the least known guys here, but he also sent most interesting answers. He has a popular blog about programming. He is also the author of a game called "Wyvern".

Peter Norvig - Research Director at Google, a well known Lisper, author of famous (in some circles at least) books about AI. See his homepage here.

Guido Van Rossum - The Python language creator.

Bjarne Stroustrup - C++ language creator, has a homepage here.

James Gosling - The Java language creator.

Tim Bray - One of the XML and Atom specifications author and a blogger too.


Q 1: How did you learn programming? Were any schools of any use? Or maybe you didn’t even bother with ending any schools :) ?

Steve Yegge
I taught myself to program on an HP calculator using their RPN stack language when I was 17 years old. I’d tried to learn programming a few times before that but never really "got" it. The HP 28c and 48g scientific calculators were pretty powerful and had great docs. I wrote a 3D wireframe viewer for the 48g — I got a book on 3D graphics and painstakingly translated an example program in Pascal into the RPN stack language. It was pretty sweet when I got it running. After that I bought a PC and Turbo Pascal, and started studying programming in earnest. I was a decently good programmer by the time I went into the CS program in college.

I went to the University of Washington and got an undergrad degree in CS. It was definitely worthwhile, and I recommend that all programmers should try to get a CS degree if possible.

Linus Torvalds
I didn’t learn programming in school, but mostly on my own reading books and just doing it (initially on a Commodore VIC-20, later on a Sinclair QL).

That said, I think especially University was very useful. Rather than go to an engineering school, I went to Helsinki University, which is pretty theoretical, so there the teaching concentrated not so much on programming (which was just a small part, and which I ended up doing more of "on the side" anyway), but most of the courses tended to be on fundamental concepts and things like complexity analysis. Which can seem boring and even a waste of effort at times, but I think it was useful, and I mostly enjoyed it. And I think I’m probably a better programmer for it.

David Heinemeier Hansson
I learned programming by starting to put together my first web page in HTML. Then I wanted to make some dynamic pieces and picked up first ASP then PHP. After I already knew how to program, I then started on a joint computer science and business administration degree.

Peter Norvig
I took courses in high school and college, but always felt I learned more on my own.

Dave Thomas
During my secondary schooling I took a class in a local technical college on computers. It got me totally hooked: I fell in love with programming, and looked around for colleges offering courses in software. Eventually I went to Imperial College, part of London University. It was only the second year they’d offered a course in software, and it was absolutely marvelous: the staff and students worked together to make the materials better, and everyone learned a lot. The undergraduate course there gave me an incredibly strong background in software development. I stayed on to start a PhD, but got lured away by a startup.

But the overall question is "how did you learn programming?" The real answer to that is "I’m still learning programming." I think any good developer continues to learn throughout their careers. It isn’t just a question of picking up new languages and libraries: good developers also refine their techniques and practices over the years.

Guido Van Rossum
I went to university where they had a big mainframe and there were various computer courses. This was very important for me.

James Gosling
Initially, I was self-taught. I got my first programming job before I went to college. But I’m glad I did. I had a lot of fun. I kept going until I had a PhD.

Bjarne Stroustrup
In university (Aarhus and later Cambridge). The universities taught me much that was useful, incl. most of the bases for my future work. In addition, I learned a fair bit from programming for money - where understanding of real-world problems, correctness, maintainability, on-time delivery, etc. were more pressing than in an educational setting.

Tim Bray
I thought I was going to be a math teacher. The math program at University required a few computer science courses.


Q 2: What do you think is the most important skill every programmer should posses?

Steve Yegge
Written and verbal communication skills. You’ll never make it very far as a programmer in any field unless you can get your ideas across to people effectively. Programmers should read voraciously, practice writing, take writing courses, and even practice at public speaking.

Linus Torvalds
It’s a thing I call "taste".

I tend to judge the people I work with not by how proficient they are: some people can churn out a lot of code, but more by how they react to other peoples code, and then obviously by what their own code looks like, and what approaches they chose. That tells me whether they have "good taste" or not, and the thing is, a person without "good taste" often is not very good at judging other peoples code, but his own code often ends up not being wonderfully good.

But hey, it’s not the only thing. One thing that is very useful, especially in an open source project, is simply the ability to communicate well what you want to do, and how you are going to do it. The ability to explain to others why you do something a certain way is very important, and not everybody has that ability.

That said, in the end there are also the people who just churn out good code. They may not be good at explaining it, and they may not even have great taste, but the code works well. Sometimes you need another person (one that does have that hard-to-define "taste") to maybe massage the code into a form where it’s useful in the bigger picture, but just the ability to write clear code for difficult problems is obviously a fairly fundamental part of any programmer.

David Heinemeier Hansson
A strong sense of value. The ability to ask yourself the question: Is it worth doing what I’m doing right now? So many programmers seem to waste oceans of time on stuff that just doesn’t matter. And not enough on the stuff that does.

Peter Norvig
I don’t think there’s one, but let’s say concentration.

Dave Thomas
Passion.

Guido Van Rossum
Your questions are rather general and hard to answer. :-) I guess being able to cook an egg for breakfast is invaluable.

James Gosling
To be self motivated. To be really good, you have to be in love with what you do.

Bjarne Stroustrup
The ability to think clearly: A programmer has to understand problems and express solutions.

Tim Bray
Ability to prefer evidence to intuition.


Q 3: Do you think mathematics and/or physics are an important skill for a programmer? Why?

Steve Yegge
There is a large branch of mathematics that’s very important for programmers called "discrete math" or "concrete math". It includes disciplines such as probability, combinatorics, graph theory, induction proofs, and other useful tools. I would encourage all programmers to study discrete mathematics to whatever extent they can. Even a little is better than none at all.

As for more traditional math, well, I don’t use it as often, but it comes in very handy when I need it. For instance, I’ve only used calculus once in the past year as part of my job. I had to estimate loads for the peak traffic hour of the day for a service whose load "follows the sun" in an approximate sine curve. The simplest way to make the estimate was to integrate over 1/24th of the curve at a specific time. If I hadn’t known calculus, I would not have known how to make a reasonably accurate estimate.

When I was writing my game, Wyvern, having a solid working knowledge of basic planar geometry was incredibly helpful. And it’s quite common to use algebra and linear algebra on a regular basis. But I rarely use trigonometry or differential equations on the job, and not much calculus either.

I’d say my basic math foundation has made me maybe 5% to 10% better as a programmer. If I knew a lot more math, I’d undoubtedly be a much better programmer than I am today, so I study and practice math several hours a week.

I love physics and I have an ongoing, lifelong quest to try to understand the underpinnings of quantum mechanics. But I’ve never personally found any physics very useful towards my job as a programmer. That would, of course, be diffferent if I were doing something in a physics domain, such as 3D game programming, or certain types of simulation.

Linus Torvalds
I personally think a fairly strong math background is a good thing. I’m not as sure about the physics side, but I’m convinced that understanding math and having a good background in it helps you to be a better programmer. If only because the mental models are similar - you can build up any kind of set of rules you want, but it should be self-consistent.

David Heinemeier Hansson
Not at all. At least not for the kind of business programming needed for web applications. I consider it much more important that someone is a good writer.

Peter Norvig
Yes. Many ideas are inherently mathematical: induction, recursion, logic, etc.

Dave Thomas
Maybe. But, to be honest, I haven’t seen much of a correlation either way between these types of discipline and good software developers.

However, I have seen a strong correlation between people who have some music in their background and programming skills. I have no idea why, but I suspect that some of the areas of the brain that make someone musical also make them good at software development.

Guido Van Rossum
Math, yes (for some parts; I don’t care for differential equations, but algebra and logic are important). Physics, I don’t think so except it’s always useful to be interested in many different things.

James Gosling
Yes! They teach you logic & deduction... To have an analytical eye. And there’s no replacement for mathematics when it comes to analyzing algorithms.

Bjarne Stroustrup
Depends on the programmer and the programming tasks. Some forms of math are frequently useful; physics less often so, but then learning physics is one of the best ways of learning practical math.

Tim Bray
In my case, I’ve almost never used my university-level math to support my programming.


Q 4: What do you think will be the next big thing in computer programming? X-oriented programming, y language, quantum computers, what?

Steve Yegge
I think web application programming is gradually going to become the most important client-side programming out there. I think it will mostly obsolete all other client-side toolkits: GTK, Java Swing/SWT, Qt, and of course all the platform-specific ones like Cocoa and Win32/MFC/etc.

It’s not going to happen overnight. It’s very slowly been going that direction for ten years, and it could well be another ten years before web apps "win". The tools, languages, APIs, protocols, and browser technology will all have to improve far beyond what you can accomplish with them today. But each year they get a little closer, and I’ve finally decided to switch all my own app development over to browser-based programming from now on.

Microsoft and Apple definitely don’t want this to happen, so a necessary first step will be for an open-source browser such as Firefox to achieve a dominant market position, which will in turn require some sort of Firefox-only killer app. (A killer app would be something like iTunes, something that everyone in the world wants to use, badly enough to download Firefox for it.)

Linus Torvalds
I don’t think we’ll see a "big jump". We’ve seen a lot of tools to help make all the everyday drudgery easier - with high-level languages and perhaps the integration of simple databases into the language being the main ones. But most of the buzz-words have been of pretty limited use.

For example, I personally believe that "Visual Basic" did more for programming than "Object-Oriented Languages" did. Yet people laugh at VB and say it’s a bad language, and they’ve been talking about OO languages for decades.

And no, Visual Basic wasn’t a great language, but I think the easy DB interfaces in VB were fundmantally more important than object orientation is, for example.

So I think there will be a lot of incremental improvements, and the hardware improvements will make programming easier, but I don’t expect any huge productivity help or revolutions in how people do things.

At least not until you start approaching real AI, and I don’t think real AI is going to be anything you will ever "program".

David Heinemeier Hansson
I try not to predict the future. I’m not a big believer in fortune telling. The best way to predict the future is to implement it.

Peter Norvig
Large-scale distributed processing.

Dave Thomas
The next big thing in computer programming will be eclipsed by the next–next big thing in programming, and so on, and so on. I’m kinda tired of the endless search for the big things, because while doing it people tend to forget about the real issues: getting the fundamentals right. We need to get a whole lot better at talking with our customers, focussing on delivering value, and taking pride in what we do. A developer who can do these things can deliver great software with any tool set, and won’t need to worry about tracking the fads and fashions.

Guido Van Rossum
Sorry, I’m not much of a crystal ball person. I predicted CGI about 5 years after it had been invented. :-)

James Gosling
The two issues I’m most concerned about now are coping with parallelism and complexity.

Bjarne Stroustrup
I don’t know, and I don’t like to guess.

Tim Bray
No idea.


Q 5: If you had three months to learn one relatively new technology, which one would you choose?

Steve Yegge
I do happen to have 3 months (part-time), and I’m spending it learning Dojo (http://dojotoolkit.org) and advanced AJAX and DHTML. I’m learning it by writing a fairly ambitious web application. Dojo’s really cool, and I’m sure it will only improve with time.

Linus Torvalds
Hmm. I’d really love to do FPGA’s, but I’ve always been too busy to really sit down and start learning. I love the notion of playing with hardware: it’s obviously one of the reasons I ended up doing operating systems, since that (along with compilers) is about as close as you can get to playing with the hardware, without actually designing or building it yourself.

David Heinemeier Hansson
Cocoa programming for the Mac.

Peter Norvig
I’d like to know Javascript better. Also flash.

Dave Thomas
If "new" means "new to Dave Thomas" then I think I’d take intensive piano lessons.

If "new" means technology stuff, then I guess I’d choose technologies related to accessibility for people with disabilities.

Guido Van Rossum
Snowboarding.

James Gosling
For fun, I’d catch up on the latest in 3D rendering. I’d probably write a photon-map renderer.

Bjarne Stroustrup
There are very few things of any importance that you can learn in three month. I think you must be thinking of training in a well established field.

Tim Bray
Security, encryption, digital signatures, identity, etc. It’s a big problem for me that I’ve never learned this stuff.


Q 6: What do you think makes some programmers 10 or 100 times more productive than others?

Steve Yegge
I think if you pause to consider why not all atheletes are equally good, you’ll have your answer(s). Thomas Edison has a relevant quote about genius that might also provide you some clues.

Linus Torvalds
I really have no idea. I think some people are just better able to concentrate on the things that matter, and I think a lot of it is just doing it. Most of the really good programmers I know started doing it fairly young.

David Heinemeier Hansson
The ability to restate hard problems as easy ones.

Peter Norvig
The ability to fit the whole problem into their heads at one time.

Dave Thomas
They care about what they do.

Guido Van Rossum
Genetically different brain structure.

James Gosling
They think about what they do. They don’t rush in and slap things together. They have a holistic picture of what is to be built.

Bjarne Stroustrup
First a general lack of professionalism and adequate training. that sets the base level too low. Secondly, some people have a combination of "smarts" (ability to think clearly and get to the heart of things), experience, and knowledge of tools. Programming leaves more scope for that because it is a combination of theory and practice - neither of which is much use without domain knowledge.

Tim Bray
The surprising variability of the human mind.


Q 7: What are your favourite tools (operating system, programming/scripting language, text editor, version control system, shell, database engine, other tools you can’t live without) and why do you like them more than others?

Steve Yegge
OS: Unix! I use linux, cygwin, and darwin all about equally often now. You just can’t beat it for productivity tools. Every programmer should learn how to use every tool in /bin and /usr/bin.

Scripting language: Ruby. I’m proficient with just about every major scripting language out there: Perl, Python, Tcl, Lua, Awk, Bash, and others I’m forgetting. But I’m really lazy, and Ruby’s by far the easiest, so it’s a match made in heaven.

Programming language: I don’t have a favorite; I think they all suck. I tend to prefer Java because it’s a strong, portable platform with good tools and good libraries. But the Java language will evolve or die; it’s not good enough as-is to hold the lead indefinitely.

Text editor: Emacs, because it’s the best thing out there today.

Version control: SVN. Perforce is better, but it’s very expensive.

Shell: Bash, because I’m too lazy to learn a better one.

Database engine: MySQL, of course. Nothing else comes close.

Others: I find the GIMP invaluable, and also maddeningly unintuitive. I’ve been using it for years and can still barely do anything with it. But I couldn’t live without it, ironically enough.

Firefox is becoming an increasingly critical part of my tools lineup. I feel suffocated when I’m forced to use IE or Safari.

Note that all these tools (Unix, Emacs, Firefox, GIMP, MySQL, Bash, SVN, Perforce) have something in common: they’re extensible; i.e., they all have programming APIs. Great programmers learn how to program their tools, not just use them.

Linus Torvalds
I actually don’t end up having that many tools I work with, and for the many of them I have spent some time of my own to just make them work for me. The OS part is clearly the biggest one, but I’ve obviously also written my own version control system (git), and the text editor I use (micro-emacs) I’ve ended up customizing and extending upon too.

Other than those three parts, the only thing I care deeply about is my email reader. I use "pine" - not because it’s necessarily the greatest email reader ever, but because I’m used to it, and it does what I need it to do with a minimum of fuss.

David Heinemeier Hansson
OS X, TextMate, Ruby, Subversion, MySQL. That’s the combo currently keeping me happy. I like tools that exhibit good taste and a focus on the stuff that matters.

Peter Norvig
I dislike all three major OS - Windows, Mac, Linux. I like Python and Lisp. Emacs.

Dave Thomas
I switched to Macs a couple of years ago after being a Linux person for more than 10 years. The tools are not necessarily better, but they don’t have to be sharpened or maintained as often, which lets be concentrate on just using them.

I'm not a great believer in single tools: I tend to switch around quite frequently just so I can get experience with as many tools as possible. Right now I’m using OSX, Emacs, TextMate, Rails, Ruby, SVN, CVS, Rake, make, xsltproc, TeX, MySQL, Postgres, and a whole lot of small productivity aids. Who knows what I’ll be using next year.

Guido Van Rossum
Unix/Linux, Python, vi+emacs, Firefox.

James Gosling
These days I live in NetBeans. It does everything I want, very cleanly simply and efficiently. It’s the nicest environment I’ve ever lived in.

Bjarne Stroustrup
Unix, sam (a very simple text editor), and a good C++ compiler, of course.

Tim Bray
I like Unix-like operating systems, dynamic languages like Python and Ruby and statically-typed languages like Java (in particular the Java APIs), Emacs, whatever, bash, whatever, NetBeans.


Q 8: What is your favourite book related to computer programming?

Steve Yegge
Man, that’s a tough one. Maybe "Gödel, Escher, Bach: an Eternal Golden Braid" (Hofstadter)? Although it’s not strictly about programming. If you specifically mean "favorite book about programming", then maybe SICP (mitpress.mit.edu).

Linus Torvalds
Heh. When I read these days, I tend to either read fiction, or non-computer-related stuff (oldie but goodie: "The Selfish Gene" by Richard Dawkins).

When it comes to programming, the only real programming book that comes to mind is actually the classic Kernighan & Ritchie "The C Programming Language" book, because it’s such an incredibly useful book while being so very readable and short. Considering that you can basically learn one of the most important programming languages of our times from it, the fact that it’s thin and readable is just a wonder.

That said, many other books I enjoyed a lot were not about programming per se, but about computer architecture and hardware. There’s obviously Patterson & Hennessy’s computer architecture book, but for me personally perhaps even more Crawford & Gelsinger’s "Programming the 80386?, which was what I used when I started with Linux.

For similar reasons, I have a soft spot for Andrew Tanenbaum’s "Operating Systems: Design and Implementation".

David Heinemeier Hansson
I like Extreme Programming Explained for its rejection of common thinking about programming practices and Patterns of Enterprise Application Architecture for striking the right balance of abstract and concrete.

Peter Norvig
Structure and Interpretation of Computer Programs

Dave Thomas
It depends on what you mean by "favorite." Probably the best written book I’ve read in the area is IBM’s "IBM/360 Principles of Operation."

Guido Van Rossum
Neil Stephenson’s Quicksilver.

James Gosling
Programming Pearls by Jon Bentley.

Bjarne Stroustrup
K&R.

Tim Bray
Bentley’s Programming Pearls


Q 9: What is Your favourite book NOT related to computer programming?

Steve Yegge
Just one book? You’re asking for the impossible. There are too many great books out there to choose just one.

My favorite books that I’ve read this month are "Stardust" (Neil Gaiman) and "The Mind’s I" (Hofstadter/Dennet).

My favorite writers are Kurt Vonnegut, Jr. and Jack Vance.

Linus Torvalds
Well, I already mentioned the Selfish Gene by Dawkins. On the fictional side, there’s just a lot of books I’ve read and enjoyed, but few I’d say were my "favourite" one. I tend to not often re-read the books, and the selection would change over time. It’s mostly science fiction and fantasy, eg "Stranger in a Strange Land" by Heinlein was my favourite one as a teenager, but it’s a bit less clear-cut for me these days..

David Heinemeier Hansson
1984, George Orwell.

Guido Van Rossum
Neil Stephenson’s Quicksilver.

James Gosling
Guns, Germs & Steel by Jared Diamond

Bjarne Stroustrup
It changes over time. Currently O’Brian’s Aubrey/Maturin series.

Tim Bray
One Day in the Life of Ivan Denisovich


Q 10: What are your favourite music bands/performers/composers?

Steve Yegge
Favourite genres: classical, anime soundtracks, video-game music
Favourite composers: Rachmaninoff, Chopin, Bach
Favourite performers: David Russell (classical guitar), Sviatoslav Richter (piano)
Favourite anime OSTs: Last Exile, Haibane Renmei

Linus Torvalds
I'm actually not very much into music, but when I listen to it, I tend to listen to various classic-rockish things, ranging from Pink Floyd to the Beatles to Queen and The Who.

David Heinemeier Hansson
I like a lot of genres. Beth Orton, Aimee Mann, Jewel, Lauryn Hill. Actually, all those examples would fit under Girls with Guitars ;).

Guido Van Rossum
Philip Glass.

James Gosling
I tend to like folk musicians: Christine Lavin, Woody Guthrie, Pete Seeger...

Bjarne Stroustrup
Band: The Dixie Chicks. Composer: Beethoven.

Tim Bray
Read my blog.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Postscript

How I found this: I was reading this post by Jeff Atwood, titled "Linus Torvalds, Visual Basic Fan" where the source article he linked to was: "STIFF ASKS, GREAT PROGRAMMERS ANSWER". But that link was dead, and doing a search didn't find any trace of it anywhere on the net. So I resorted to an archive.org search and sure enough, up popped this incredible blog post by Jaroslaw Rzeszótko.

Due to it no longer being available on the net, I thought it should reposted again so others can enjoy reading it. I requested permission from the original author to do this, and he happily obliged - thanks again Jaroslaw!

Follow @dodgy_coder on Twitter

Read a Chinese translation of this blog post  << Thanks to the guys at CoolShell.cn for this!

Read a Turkish translation of this blog post  << Thanks to Seval for this!

Old school developers - achieving a lot with little

Ken Thompson

I was reading through bits of the excellent Coders at Work book by Peter Seibel and found it interesting that the (what I'd call respectfully) old school developers, such as Ken Thompson, Joe Armstrong and Jamie Zawinski use hardly any modern tools and techniques while developing software.

Ken Thompson was the designer of the B programming language (the precursor to C) and also the Go programming language (with Rob Pike). He also worked alongside Dennis Ritchie to develop the C programming language and the original Unix operating system while working at Bell Labs. After designing the UTF-8 method of character encoding on a placemat in a New Jersey diner, he worked overnight to implement full UTF-8 support in Bell Lab's Plan 9 operating system. Regarding his programming style, he debugs only via printf statements, hardly ever uses unit tests, starts his projects by designing the data structures and then works bottom up, with throwaway test stubs.

Joe Armstrong

Joe Armstrong developed the Erlang programming language and the open source Open Telecom Platform (OTP) framework for Ericsson. When designing software he prefers to rigorously document as much as possible up front before starting to write code, especially for difficult projects such as those involving real-time networking protocols. He uses prototypes to solve the hard problems first, and for debugging, just uses print statements. He is a critic of Object Oriented Programming, and favours functional programming languages like Haskell. He never uses an IDE, preferring just Emacs and the command line (no mouse required).

Jamie Zawinski worked in MIT's AI lab with Lisp, then moved onto leading the development of Lucid Emacs, also in Lisp, which later became known as XEmacs. After this he joined Netscape and worked on the front-end for the Unix version of Netscape Navigator, and later led the development of Netscape Mail (both projects were in C). He too prefers to just use print statements to debug his code. His process is either to write code top down or bottom up, let the code evolve naturally, then refactor it when necessary. During development he hardly ever uses unit tests, believing it slows things down - he thinks there's a lot to be said for getting the code right first time. In his view, its a matter of priorities, "do you want this to be good software or do you want it to be done next week - pick one because you can't have both".

Jamie Zawinski

So what tools and techniques do you actually need to be a great programmer? Are these modern techniques of TDD, BDD, Scrum, Agile, Design Patterns, XP etc. just window dressing or do they serve a higher purpose? Is it just a fact that a gifted programmer will always find a way to be a great programmer, no matter what tools he or she works with?

My own view is that some level of TDD and a reliance upon a decent set of unit tests is invaluable. Maybe then these modern techniques should just be seen as advances to make our job a little easier and more productive, especially for those who are not a genius-level developer, like these guys.

Follow @dodgy_coder

Subscribe to posts via RSS