Philip Dorrell's Blog

Source code quality is a specific aspect of software quality which needs to be managed by a specialist source code quality controller.

Testing increases the value of ugly code, and prevents the code from being refactored. Is there anything that developers or their managers can do to prevent lock-in to ugly code?

I compare various online music perception tests to what I believe would be the most scientifically interesting music perception test: a musicality perception test.

Amusia is sometimes equated with tone-deafness. But this equates music appreciation with pitch discrimination. Even if we add other components of music perception, such as tempo discrimination, it seems very unlikely that music perception is a simple combination of these individual components. Music scientists studying amusia should make more effort to discover "true" amusics whose amusia is not simply a function of more basic perceptual deficits. "True" amusics may be much rarer than the relatively common "tone-deaf" amusics.

Why do the two big lies about copyright and copy-control get told? Why do people continue to believe them?

Lie 1: Copy-control can be made to work. Lie 2: There is no possible alternative to copyright.

By a probabilistic argument, it may be possible to determine whether or not we live in a parametrized multi-multiverse, or in a non-parametrized multiverse, where the latter case suggests the existence of a God who has chosen the free parameter values of the laws of physics for some particular purpose. Such a purpose might include our own existence.

Using the Python Queue module to perform independent S3 requests in separate threads.

The really basic business web-site is a web-site owned by you and controlled by you. It provides the basic facts about your business, and, it can be found when someone searches on Google for the name of your business.

Can the Dream-Maker theory of dreams be reconciled with the theory of synaptic downscaling proposed by Giulio Tononi and Chiara Cirelli?

How to create an audio CD using the open-source applications SoX and InfraRecorder.

Music theory is theory in the sense that it is distinct from musical practice, but it is not theory in the sense of being a scientific theory.

Music perception reveals to us an otherwise hidden sense. The three ways of knowing are genetic, experiential learning and cultural learning. Cultural learning can be verified or unverified. "Musicality" perception is a hidden sense that increases the reliability of unverified cultural learning.

Both adaptive and non-adaptive theories of music have to deal with the problem of amusia. For adaptive theories, amusia is a deficit in the creation and appreciation of music itself, and the problem is to explain how music helps long-term reproduction. In the case of non-adaptive theories, there has to be some other cognitive deficit associated with amusia, i.e. other than the inability to create and appreciate music.

If music is like a mental virus, then the brain's response to it is analogous to how the immune system responds to germs. Another useful metaphor is Steven Pinker's "auditory cheesecake", but if music is a very powerful super-stimulus, then a better phrase might be "auditory super-cheesecake".

The most secure way I know to do Internet banking is to do it on a PC booted from a LiveCD. However no system is perfect, so here I consider ten ways that LiveCD security could fail.

The development of the human ability to perceive musicality may have triggered one of the most important events in human evolution: the appearance of modern behaviour.

If music is like a drug, then why haven't we evolved resistance to it? Or have we indeed already evolved resistance? In which case, the emotional and pleasurable effect of music may have been much stronger in the past than it is now.

Uncritical listening may play a significant role in worldview construction, simply because there is no other means of acquiring information as efficiently. But, because it is (obviously) vulnerable to manipulation, it must occur in a secret manner which is kept hidden from both the speaker and the listener.

If you are an application security tester, you should be able to answer yes to all three of these questions.

Advice addressed to banking customers, banks, ISP's, developers of operating systems and hardware manufacturers.

Why haven't computers yet replaced books and magazines? Although books have various advantages over all existing "e-books" and other computerised reading devices, I think there is only one advantage that really matters, and what's more, I think that we already have the technology available to eliminate that advantage – someone just needs to put all the right pieces together in the right way.

A commentary on the paper The Distance Geometry of Music in relation to the super-stimulus theory of music.

There are things that only a select few can do. And there are things that almost anyone can do. But as time passes, and technology develops (and develops at an accelerating rate), some things move from the "select few can do it" to the "anyone can do it" category.

A scientific theory should be able to inform the experimental scientist. Here I show how the super-stimulus theory of music can suggest better approaches to musical experimentation.

I recently wrote an article What is Music? explaining my super-stimulus theory of music, which Jack Hunter kindly posted as part of his What is Music Project.

I try out nine different methods for annotating HTML elements, judging each method according to five criteria: proximity, non-kludginess, syntactic correctness, reliability and absence of side-effects.

Song lyrics seem more significant and profound than they really are. Is this an accidental side-effect, or is it due to the operation of an adaptive brain system whose very purpose is to estimate the signficance and profundity of what is being said?

On the first day, God created the perfect exception ...

Is music a puzzle, which requires access to special inside information to be solved, or is it a mystery, where we already have enough information, and we just need to work out how to use that information?

Copyrights and patents link payment for the creation of ideas and content to the ability to control the use of ideas and content. This causes a direct conflict between intellectual property and digital freedoms. The conflict is most intense for high-budget content sold to a mass-market – so called "premium content" – where big money is involved. The conflict could become a war to the death, and I think we will all be better off if we can find an alternative: a way to pay for premium content without sacrificing our digital freedoms.

Two central themes in "Web 2.0" are decentralisation and ownership. Voting sites like Digg and Reddit are considered to be "Web 2.0". But they aren't fully decentralised. True decentralisation would mean ownership of content by the content producer, ownership of voting identity and votes by the voter, and separate ownership of the aggregation process by the aggregator. In each case "ownership" would be by means of association with the owner's domain name.

If, as the genetic evidence suggests, the Neanderthals were a separate species, then they must have been ugly, and they would have looked even uglier to our modern human ancestors than they would look to us if we saw them now. (Of course our ancestors looked ugly to them too.) Despite this mutual repulsion, the genetic evidence suggests that occasional inter-species mating did occur, although such encounters may not necessarily have been consensual for both parties.

Zero divided by zero is a value which according to mathematical tradition does not make any sense. Not only is it undefined, it is also indeterminate. Can one invent a useful "algebra" that includes such a value?

I am adding comments to my blog. This blog entry is a sandbox-style entry for trying it out.

Every introduction to the study of automobiles must start with the question of definition. Using techniques of discourse and analysis borrowed from the academic study of music theory and music philosophy, a complete and satisfying definition of the word "automobile" is derived in three steps. A special challenge to the student's thinking about automobiles is included at the end of the article.

If musical taste is subjective, then perhaps music is not intrinsically musical. But we often talk about our subjective judgements as if they were objective, or at least independent of the beholder. If musicality as a property of musical items is intrinsic in the sense of being consistently inter-subjective, this has important consequences for music science.

Music is defined by what happens within it and what doesn't happen within it, and by the contrast between the things that happen and the things that don't happen. This "things that happen"/"things that don't happen" contrast is the musical universal, which accounts for both pitch-related aspects and rhythm-related aspects of music.

Is there a syntax of music which is analogous to the syntax of language? Some recently published scientific studies explore this analogy, and identify the "rules", "principles" and "regularities" of music with "syntax". However, musical items may be constrained by certain rules because they are local solutions to an optimisation problem. I suggest an experimental protocol which tests this hypothesis by looking for dissociation between musical expectation and musical pleasure.

Combining recent advances in identifying human accelerated regions in the human genome with the possibility that music perception involves non-neuronal circuitry (i.e. glial cells such as astrocytes), it may be sooner rather than later that a strong candidate for a musical gene can be identified.

It's good if you can get a lot of traffic to your website because one of your articles appears on the front page of Digg. It can be even better when an article makes it to the front page twice. Read this article to find out the four steps to follow if you want your article to appear on the front page of Digg twice.

Which is the biological adaptation: making music or listening to music (or both)? What makes some music better than other music? Would what I am saying about music make sense if I said the same thing about something else?

One of the things that music scientists study about music is music perception. But "music perception" is the perception of music, and music is something that we don't know what it is. In fact, music appears to be something created entirely so that we can perceive it. Which is all very circular.

Traditional programming requires the programmer to deal with one kind of source code written in one programming language. DSL-based metaprogramming requires the programmer to handle five different kinds of source code (written in at least three different languages).

Four steps towards creating an online music science community that attracts attention and contributions from both professionals and amateurs. These techniques may also be of interest to anyone else trying to create an online community on a specialist topic.

Most real-world programming is like solving a puzzle. But some programmers see writing programs as a way of expressing meaning.

Some handy tips on writing an animated 2-D video game in Javascript which can run directly in the web browser, based on my experience writing the game PrimeShooter^TM.

The traditional philosophy of "DRY" (Don't Repeat Yourself) considers all repetition in the source code to be "harmful". But the AganeAndAgane programming language is based on a newer philosophy of "DRYDBERIOK".

The Peruvian singer Dina Paucar sings Huayno in a very distinctive style which achieves a strong emotional effect.

There are some people out there investigating the mathematics of music using quite sophisticated concepts from group theory and category theory. But is this mathematics really telling us anything interesting about what music is? I highlight some questions that the music mathematicians forgot to ask.

A common mistake, sometimes made even by mathematics professors in the appendices of their textbooks, is to assume that the Second Law prohibits any decrease in the entropy of a closed system. But it doesn't. It only prohibits "macroscopic" decreases in entropy. Which is a good thing, because evolution by natural selection requires the occurrence of small entropy decreases in closed systems.

If bird "songs" are not bird music, then why do they sound musical to us?

I present another attempt to find a way in which website hosts can make some effort to increase the credibility of advertising that they host on their website, thus increasing the value of the advertising space that they sell, without staking too much of their own reputation on the results.

Tetris Dreams are dreams you have when you have been playing Tetris too much. Something similar happens to me when I spot unfamiliar bird species in my dreams.

Traditional science excludes miracles altogether. But quantum mechanics and thermodynamics make miracles possible, the Multiverse makes them likely, and the Weak Anthropic Principle makes it possible for us to actually observe them.

The technological singularity can only transition to pure machine intelligence if it can find a way to maintain purpose. But purpose depends on reproduction and natural selection, which may not apply to a machine-based super-intelligence.

Schroedinger's Cat does not remain in a mixed state (relative to macroscopic observers outside of the box) for any measurable length of time, because even the tiniest bit of information about the cat's state leaking out of the box is enough to correlate the wave functions of the cat and any external observer.

Whether or not Euclidean geometry is true started out as a mathematical question, but it eventually evolved into a physical question, i.e. is Euclidean geometry the geometry of the universe? Nowadays we believe in both Euclidean and non-Euclidean geometries, because they can both be defined numerically. But we can ask the same kind of question about number systems. For example, is natural number arithmetic the arithmetic of the universe? If it isn't, one plausible alternative is arithmetic modulo N, where N is some very large number. If arithmetic modulo N is the arithmetic of the universe, it follows that the number N is physically equal to zero.

Combine my Voted Compensation system with a suggestion from Hannibal at ArsTechnica, and you end up with a practical scheme for worldwide funding of free entertainment content, where consumers pick the winners after they are produced, and not before.

FeedTree is the latest entrant into the "micronews" business, with a proposed solution to the RSS Bandwidth problem. But I invented a system with very similar properties in 2000.

In Web 1.0, some scientific journals put their content on the web. And now there are pre-print archives, PLOS and open peer-review. But a fully Web 2.0 peer-reviewed journal goes further than this: the journal cedes control and responsibility to authors and reviewers, and if there is any journal at all, it takes the form of a post-publication aggregation of the output of selected authors and reviewers. Web 2.0 peer review promises to achieve something that Web 1.0 peer review can't: the revival of amateur science as a contributor to mainstream science.

Advice on how to create and promote your non-mainstream theoretical science website, based on my own personal experience. Since I haven't been very successful in getting anyone anywhere to take any notice of my ideas on a wide range of subjects, which include thermodynamics and evolution, consciousness, dreams, neuroscience and (most recently) music, it could also be read as advice on how not to create and promote your non-mainstream theoretical science website.

Reddit has a Hot page and a New page. Each of these has its advantages and disadvantages, and there is a case for allowing users to select pages with in-between characteristics.

The Internet is great for those who wish to educate themselves informally and cheaply. But what if you want to achieve a formal qualification? This article describes an architecture for an Internet-based computerised examination system.

In "The Brain – An Orchestra Without A Conductor", Wolf Singer argues that the brain lacks any coordinating supervisor, despite our strong intuition that our mind does have some central point of control. The only thing is, orchestras do have conductors, and there are good reasons why they have conductors. And these reasons apply equally to the human brain.

The defining aspect of consciousness is that it is a centre of responsility and accountability in the brain.

An alternative to advertising is paid reviews. But if the reviews are to be credible, the process by which they are paid for must be totally transparent, and the reviewer must be prepared to put their own reputation on the line.

Can the Internet think? Is it a conscious entity? I suggest that attention, belief and knowledge are relevant attributes that the Internet needs to develop before it can become conscious. Various collaborative and voting-based social websites appear to be the start of a process which will unify these attributes across all or most of the Internet, turning it into a recognisable thinking "being".

It's important to insulate your code from changes to external identifiers used to communicate with external systems.
Or is it?

Statistical analysis of music promises to deliver algorithms that can distinguish "good" music from "bad" music. But analyses disconnected from biology are unlikely to lead to any major improvement in our understanding of what music is, even if they result in the discovery of significant correlations between observed characteristics of music and its popularity.

I never implemented my invention, and I never patented it. All that is left now is a copy of a web page on Wayback.

In an ideal world, website passwords with a "guess factor" of, for instance, 10000, would provide sufficient security for most users, and so-called "strong" passwords would be less necessary. But you, as website developer, have to make some effort to provide this level of security to your users.

We've all been conned. The term "identify theft" describes an interaction between the fraudster and the victim, but it fails to accuse the intermediary in the transaction of any wrong-doing. A better term would be "negligent identification".

The super-stimulus theory of music predicts the existence of a specific type of brain cell whose sole purpose is to detect musicality. It should be possible to detect these cells anatomically, and maybe someone already has. Can you help me to find them?

How scientifically can we answer questions about the purpose of life? Is it even scientifically meaningful to talk about "purpose"?

An explanation of the Fundamental Theorem of Arithmetic, which states that every natural number has a unique factorisation into prime numbers.

Yes, with a bit of practice, you too can calculate the cube root of a 9 digit number, in your head.

The Technological Singularity is a predicted future event in human history caused by the ever-increasing ability of new technology to speed up the rate at which new technology is developed. There is one type of technology that has an enormous potential for self-hosted speedup in the very short term: open source software on the Internet.

Is there a way to develop and distribute software such that any contribution from anyone can be added to a software project without any explicit co-ordination between any of the parties involved? What if there was only one way to write any software component? You could give everyone "commit" rights. You wouldn't even need version control!

Mind Energy is a useful concept that helps to explain the interaction between volition (or "willpower") and feelings.

It's hard enough for amateur scientists to do any kind of science, and be taken seriously. But theoretical science appears to be 100% the domain of professionals. Is there any way that amateur scientists can make a useful contribution to the development of scientific theories? Can we invent new Internet/Web technology to solve this problem?

DRM requires someone else to have power over your computer. Absolute DRM means absolute power, and as Lord Acton famously said, absolute power is not a good thing.

Do New Zealanders really understand their own electoral system? If they do, why aren't they all doubling their effective party votes by voting for overhang candidates?

Can we scientifically explain morality? What is the purpose of morality? What is the difference between "moral truth" and "scientific truth"? Does this difference mean that science has no right to make claims about the nature of morality? Is morality absolute or is it relative?

The notion of "The Common Good" implies that humanity has common goals. But a bit of biology tells us that there are (almost) no common goals of humanity. The only possible common goals are survival of the human race and expansion into space. All other goals relate to relative success, and it is logically impossible for everyone to be relatively successful.

How do you load programs into a computer? You use a program-loading program. But how do you load that program? That's the bootstrap problem. It's a problem for those designing computers, and, when you think about it, it's also a problem for the "design" of the human brain/mind.

The "meaning" of a neuron is not something that is precisely pre-determined, and it can easily change over time. Which makes it difficult to describe exactly what the "job" of a neuron is. But we can apply an economic analogy to define the individual neuron as an "entrepeneur", tasked with producing the most valuable possible outputs from the available inputs.

The phenomenon of absolute pitch may not be completely irrelevant to the scientific study of music, but it is much less relevant than some people think it is.

Music acts on our emotions and feelings. Drugs act on our emotions and feelings. We generally recognise that the feelings created by drugs are not "real". Does the same apply to music? Is music a drug?

According to my super-stimulus theory of music, music perception is actually the perception of an aspect of speech, and this aspect provides information about the consciousness of the speaker. It follows that if you are creating a text-to-speech system, which does not take account of my theory, then your system will probably talk like a zombie.

Part of John Searle's Chinese Room Argument is that a simulation of something is different from the thing itself. But is this always true? Can a simulation of an object from category X itself be an object from category X? In particular, can a simulation via computer software of an object from category X itself be an object from category X? The surprising answer is, yes, it can be, if category X is a category of information processing system.

The difficulty of formally defining a particular algorithm is related to the problem of achieving re-use, because if we could write an algorithm down in a definitive form, then software developers would never have to rewrite that algorithm ever again.

How cheap can a computer be, using current technology? If we accept that the most important thing about your computer that makes it "yours" is the bits written onto the hard disk, then using LiveCD technology, you can own a "virtual computer" for just US$1. The only additional cost is that of hiring a real computer to host your virtual computer while you are using it – and this should be no more and no less than the cost of hiring a computer in an Internet café.