They are when implemented using a computer algorithm. There is no inherent difference between, e.g., a computer program that implements one-click shopping and a computer program that compresses data. You can also turn both into dedicated circuit designs, should you want to.

This is incorrect. A compression algorithm has a strict definition for all use cases, a set of input bits mapped via specified transformations to a particular set of output bits. This is no different than a chemical process patent, which specifies the inputs and transformations to generate the output; it says nothing about the specific plant implementation or similar transformations that work on different inputs or generate identical outputs.

What are the logical transforms and the input and output bit set pattern for one-click shopping? What is a universal boolean logic for selling pet food on the Internet? The very reason business method patents are being questioned is because no such specification exists or can exist for a useful implementation; any specification strict enough to be reducible to a machine would also be too narrow to have any value. In short, business method patents lack sufficiently strict specification to be directly mapped to a machine. That is a rather important difference.

Business methods patents are considered bad because in order for them to be useful as patents (i.e. not trivially worked around), they also have to be vague enough that no strict machine specification is possible.

"So I'm skeptical about the usefulness of having patents at all."

I find the consistency of this opinion quite reasonable; I have no strong opinion for a specific outcome, but a logically consistent position would seem to be all or nothing (sans business method patents, which are a different kind of beast). My objections to most arguments is the lack of internal, logical consistency -- they look more like self-serving rationalizations and rent-seeking than reasoned policy.

I've worked in a couple different areas of patentable subject matter, and most of them are functionally indistinguishable from computer algorithm patents in terms of what happens. Business method patents are a whole 'nother kind of mess.

The real software patents have only been acceptable in the US since State Street (Diamond vs. Diehr was about curing rubber, except that the rubber curing was software-controlled; i.e., the patent claimed a process for curing rubber, the fact that it was computer-controlled was just an aside and not central to the patentability). State Street was in 1998.

State Street was a business method patent, not an algorithm patent. While both types of patents are often classified under the rubric of "software patents" they have very different histories. It is only business method patents that were recently allowed in the US. Bilski et al is about business method patents.

True computer algorithm patents have been allowed for many decades just about everywhere. They are (correctly) viewed as strict abstractions of circuit designs, circuit designs being unambiguously patentable subject matter. Business method patents are not an abstraction of a circuit design and therefore have traditionally not been allowed; doing something on a computer does not imply that the process reduces to a particular circuit design. The only kinds of software patents that are vulnerable to a court decision are those that are not equivalent to circuit design.

Things like encryption algorithms are obviously equivalent to specific, novel circuit designs, "one-click shopping" or "selling pet food on the Internet" quite obviously are not.

"Allowing patents on this means giving somebody the ownership of a piece of math."

No, this is only true for a particular patentable algorithm in the same way it is true for *all* patentable subject matter. This is another argument that is not consistently applied (any math argument that applies to a computer algorithm trivially generalizes to all physical machines and material processes).

To use your quicksort example, if someone patents the quicksort algorithm it does not prevent you from sorting data generally. Sure the alternative existing methods might be less efficient (something you could hopefully say about all types of patents), but there are an infinite number of alternative methods to achieve the end result. You are not patenting "sorting", you are patenting one method of doing that transformation. In the same way, you cannot patent a particular chemical, but you *can* patent a process for synthesizing that chemical. There are cases where the patented method is the only known (practical) method for achieving a particular result, but those are precisely the cases patents were intended for.

An argument that does not really follow is like your one above that someone might patent the basic building block algorithms of software. How can an algorithm that does not exist be a "basic building block"? What has everyone been using up until the point where it is invented? The algorithms we consider "basic" are so because they have been around forever and are therefore in the public domain. What were people doing before the existing "basic building block" algorithms were invented? An algorithm that becomes a future "basic building block", those are arguably the algorithms that more than any other deserve patent protection if you are going to bother at all. Pretty much non-obvious, novel, and eminently useful by definition.

There are open algorithm problems in computer science that, if solved, would generate new basic building blocks that don't exist now. This has been true in the chemical process patent field many times as well. But it seems odd to lay claim to them as "basic building blocks" before they exist. If someone invents an anti-gravity device it may become a basic building block of transportation, but does that suddenly make it non-patentable subject matter? I'm just looking for consistent treatment, not any particular outcome. If we exclude all the inventions that are good enough that they become basic building blocks, there is really no point to having patents at all. Arguably the goal of the patent system is to encourage the invention of basic building blocks that do not currently exist.

This particular argument from the article is oft-repeated but weak:

  "Software developers already enjoy strong copyright protections for their work, rendering patent protection largely redundant."

The exact same argument could be made for several classes of patent, such as chemical process patents, that people seem to generally consider legitimate patents in pretty much every country that has patents. If I am to believe that this is a compelling argument against software patents, then it is also a compelling argument against some other patentable areas. (Most arguments against software patents have this feature.)

On the other hand, a much more compelling argument can be made against "business method" patents (a subset of the suitcase called "software" patents) because they do not strictly define a machine. The reason algorithm patents (also part of the "software" patent suitcase) have long been acceptable just about everywhere is that they are strict abstractions of novel circuits (patentable material in virtually every country). As a general observation, most proponents of software patents are thinking of algorithm patents while most opponents of software patents are thinking of business method patents. The ambiguity of the term "software patent" muddies the context and makes intelligent discussions more difficult. It would help if everyone was more precise in their selection of terms.

Interestingly, for some government contracts the lowest bidder is automatically discarded -- it is the *second* lowest bidder that gets the contract. This is a well-known theoretical mechanism for removing bullshit from the bidding process. The end price will be slightly higher, but the price will usually be more accurate for a given contract spec.

USA has the highest rate of incarcerated people per capita of any country other than possibly China.

It is a bipartisan thing, unfortunately. Counter-intuitively, the most drug friendly states in the US tend to be western "red" states, due to their more libertarian perspective on individual liberty. For as nominally progressive as states like California are, their laws are awfully socially conservative even though they vote for people like Obama in a landslide. Honestly, most "blue" states are every bit as socially provincial as the "red" states. I guess that doesn't say much for American culture.

Total decriminalization of drugs has been tried in Portugal since 2001, and by all accounts has been a raging success by just about any metric you care to use. I'm happy to see other countries jumping on board the clue train, not that I expect to see something similar in the US for the foreseeable future.

For more on the Portuguese experience, see: http://www.salon.com/opinion/greenwald/2009/03/14/portugal/

If it was easier to enter the country legally fewer people would do it illegally.

As a point of fact, the US allows more legal immigration than any other country in the world. It is not as though the US is stingy about letting people immigrate compared to the rest of the world. Canada beats the US on a per capita basis, though generally North American immigration policy makes the rest of the industrialized world look like a joke. Compared to the EU, the US is positively libertine with its immigration policy. What, precisely, is so horribly wrong with US immigration policy that does not make most of the rest of the industrialized world look even worse? Exactly how many more millions of immigrants must the US accept every year before the US earns your approval? The entire EU, with twice the population of the US, only accepts a measly 1.8M immigrants.

The US is pretty liberal about who it accepts as well. From Wikipedia: "Of the top ten countries accepting resettled refugees in 2006, the United States accepted more than twice as much as the next nine countries combined..." The US may have policy problems, but its willingness to accept immigrants is not one of them.

And yes, at one point long ago, back probably before you were born, the United States used to pride itself on being the longest average lifespan in the world.

The United States still has the longest average life expectancy if you control for fatal injury rates. The actuarial statistics are misleading for the purposes of determining medical effectiveness. In essence, American medical life expectancy is at the top of the heap, but the OECD averages are brought down because people in the US have a relatively high probability of dying in vehicular accidents or being murdered when young compared to other industrialized countries. If a fatal injury does not end your life prematurely, you'll outlive your counterparts in the rest of the industrialized world. More on that here:

http://angrybear.blogspot.com/2009/07/per-capita-spending-and-life-expectancy.html

Not only mirrors (that occurred to me as well), but have the missile spin so the energy of the laser is spread out over a much larger surface. Spinning would also allow the areas of the missile to cool down somewhat.

All of the "obvious" solutions like you mention generally will not work. The power levels for the lasers are specifically designed to defeat the known technical counter-measures available to a missile designer. This is why laser weapons have a power rating orders of magnitude greater than is strictly required in most conventional circumstances; they are obviating counter-measures before anyone tries to develop them. Among other things, they are designed to ablate the target faster than you can reflect it or physically spin it.

This is also the reason a lot of US military research focuses on hyper-kinetic weapons these days; good ones can defeat all plausible molecular armor and even weak ones can defeat all current armor. The power levels of US weapon systems are getting to the point where any passive counter-measure would have to be very exotic.

In all military advances in offense, the defense will find a way around it (and vice-versa). It's a cat and mouse game. Look at how Iraq tried to foil GPS guided ordinance, they jammed the GPS signals. I don't know how successful they were but given time they might have been successful.

This is based on a media-created myth. The US has never had GPS-guided weaponry, precisely because GPS can be jammed. Therefore, it would not have done much good to have a GPS jammer beyond attracting the attention of missiles designed to destroy RF emitters.

The primary guidance mechanism usually mislabeled as "GPS guided" is ultra-precise inertial guidance, which can't be jammed at all short of altering the physics of the universe. These inertial guidance systems can optionally accept micro-corrections from a GPS input, but only within the (classified) error bounds of the inertial system which are already known to be very small. If the GPS signal deviates from the inertial guidance, the GPS is assumed to be compromised and ignored.

The "GPS-guided weapon" thing is one of many myths about US weapon systems perpetuated by the media. The US never has and never will produce a GPS-guided weapon.

The great irony of the glacier retreat being the harbinger of doom for humanity is that on most continents the glacial retreat is uncovering substantial quantities of archaeological evidence. I wonder what the people whose archaeological evidence we are finding thought about the glaciers when they encroached on their lives thousands of years ago. It is an interesting juxtaposition.

Empirical models of thermodynamic properties are definitely protected by copyright. There is a high-value market for these models, and different models of the same thermodynamic process will evaluate differently so it is a valuable creative product rather than a mere description of reality. For fields where tiny improvements in efficiency generate big cost savings, you want to use the most accurate model available where "most accurate" will be a function of the use case.

Thermodynamic property models are not measurements of reality, they are mathematical models of a physical process derived from empirical data. They are what you use to predict reality when it is not possible or practical to measure it. Turning the empirical data points into continuous functions is a creative step and the value of the creative step is in minimizing the divergence between the model and reality over as broad a range as possible. There are companies that specialize in producing and selling ultra-accurate thermodynamic property models.

I'd point out that software is covered under copyright. If I make a great new program and you copy it, I can sue under existing copyright laws. Giving that program patent protection "protects" it twice which isn't needed and merely reduces competition.

This is incorrect, in that the same argument applies to all patentable material. Patents apply to abstract processes ("algorithms"), copyrights apply to implementations (e.g. "software"). The abstract chemistry for producing a chemical output is an algorithm with no physical implementation and eminently patentable if novel, being a traditional subject of patents for centuries. The creation of an implementation of that algorithm is protected by copyright, and worth quite a bit of money in fact. This is indistinguishable from the case of algorithm patents in all important ways, up to and including the fact that the chemical process algorithm is a purely mathematical construct that only produces effects in implementation.

SQL is not a database, it is a standard interface to a feature set commonly associated with relational models. Before everyone standardized on SQL, there were other relational query languages. The "No" part of "NoSQL" refers to the fact that some basic elements of relational implementations cannot be usefully expressed using a much simpler distributed hash table model.

All the "NoSQL" does is eliminate all the parts of traditional relational databases that do no scale -- discarding the bottleneck rather than fixing it. These are things like joins and external indexing. Unfortunately, discarding those things means you discard a lot of very important functionality as a practical matter, notably the ability to do fast, complex analytics. Adopting the NoSQL architecture runs contrary to the trend toward more real-time, contextual analytical processing. There are a great many analytical applications that are not amenable to batch-mode pattern-matching, and the NoSQL model is a lot less applicable than I think some people want to acknowledge. In its domain, it is a great tool but it has many, many prohibitive limits. We are essentially trading power for scale.

That said, do not take this as an endorsement of traditional SQL relational databases either, as they have a number of serious limitations themselves. As just mentioned, a number of the core analytical operations those models support are based on algorithms that scale poorly. The SQL language itself has mediocre support for many abstract data types (e.g. spatial) and data models (e.g. graph), which in part reflects the inadequacies of the assumed underlying database algorithms (e.g. B-trees) that are implicit in SQL. The inability to efficiently do event-driven/real-time applications is also more a reflection of the access methods used in databases than any intrinsic weakness in SQL; SQL may be clunky for that purpose, but that is not the real limiter.

A truly revolutionary deviation from SQL would usefully implement a superset of the features SQL supports, not take them away. Of course, we would need access methods more capable than hash tables and B-trees to useful implement those features, which is a lot more work than discarding features that scale poorly. NoSQL is a stopgap technical measure for that small subset of applications where the serious tradeoffs are acceptable.