Monday, April 17, 2006

A Rare Moment of Unadultered Hatred: Shut the hell up.

I get frustrated just like any other person. Anyone who knows me in person can vouch that this is indeed the case. I try to keep it in check, but right now, I have but a few words to say to a great many persons:

Shut the hell up. If you can't think of anything intelligent to say, or at least something that doesn't jepordize the future of everything you might reasonably care about, then just go into your little corner of reality and don't say anything at all. You are now, and always will be, a nuisence, a distraction and a danger. While in general, I don't advocate "censorship," I must at this point admit that I need some isolation from the overwhelming idiocy that pervades what passes for political discussion in this country, and the only way to do that is for the debate to become intelligent, or for me to leave my political awareness behind. Of the two, I prefer that the dangerous idiots leave.

Having said this, I suppose I probably should make clear who I am aiming at, since we as a society have been damn near brainwashed into thinking that we actually have meaningful discourse on political, social, moral and scientific issues. I am refering to the overwhelming and completely unfounded attack on science that seems to have manifested itself this week in the form of "articles" denying the existance of global warming. By the way, don't even dare accuse me of failing to make an argument. I shouldn't have to. Many others have done so, and I don't intend to waste any effort in convincing people so utterly disconnected from reality. At this point, I wish only to encourage anti-global-warming dittoheads to simply lay off the issue until they learn what it means to have a brain and apply this learning in practice.

Let me describe some examples of the sort of thing that prompts this Unadultered Hatred. Attempting to read the opening comments of this recent Slashdot article on a related issue simply sickens me to the point of physical nausea. Reading pseudo-arguments like the following examples make me despair for humanity.

We already have droughts, floods, powerful storms, varying jet streams, famines, and lots of other weather. Why should we expect next century's droughts to be drier than last century's? When was the time when the weather was perfect for everyone? What makes you think that you can have the weather you want?
-- Kohath, comment #15145483.
Marked as 3, Insightful. I guess noting that storms exist without making any sort of actual argument passes as Insightful these days.
Face it. Most people in the US are bored. They on average spend 4 hours a day in front of the tv, 8 hours working, 8 hours sleeping, and 4 hours unexplained.

From what I hear, New Orleans is a blessing since the hurricane. Crime is almost non-existant, and people are focused on rebuilding the city, working, and being nice to each other.

Maybe a shifting environment and real estate changes will be good for us.
-- hackstraw, comment #15145816.
Marked as 4, Interesting. I suppose that I can't deny that it's interesting. Then again, isn't Hitler interesting, too, or did I just lose the debate via Godwin's Law?
Does it bother you that hurricane researchers have said repeatedly that global warming had little or nothing to do with it, and that there was an expected upswell of activity due starting last year, give or take? Or that the US coastline had been dodging the averages for the better part of 20 years, with a far smaller fraction of hurricane strikes than the historic record would otherwise suggest? What will you be saying if the next hurricane season shows lower activity than the last?
-- Martin Blank, comment #15145268.
Marked as 5, Informative. So that means if I make a bunch of baseless and uncited claims that have nothing to do with the argument at hand, it's not only perfectly on topic, but an informative contribution to an intelligent debate?
Let's make a deal:

Global warming caused last year's record number of hurricanes. So this year, when the number of hurricanes is fewer, we'll know it's because global warming has peaked and is no longer a problem. Do we have a deal?
-- Kohath (again), comment #15145883.

Marked as 4, Insightful. I give up on Slashdot moderation for now. I suppose I have to find a new target, like maybe... Digg? At least on Slashdot, the article itself is fine. The comments are what are so scary. Even then, there are some good commenters mixed in, but they spend all their time responding to idiocy like what I just referenced. On Digg, however... well, let's look at the very headlines of some recent front page articles:

"Global-warming alarmists intimidate dissenting scientists into silence."

"Global Warming Reportedly Stopped in 1998."

"Remember Global Cooling?"

Sick. To be fair, some of these have been marked as inaccurate, but again, they never should have made it to the front page. There is no substance to such articles; no arguments, no compelling presentation of new perspectives, no attempt at intelligent thought. To repeat myself, then, please, for the love of whatever you hold dear, shut any orfice from which words may emit, cease to utilize any appendage capable of recording written words, and go take a middle-school science class.

technorati tags: , , , , , ,

Wednesday, April 12, 2006

Semantics and Software: What the hell is a beta?

If this blog has any recurring theme, it is that language doesn't mean the same thing to everyone. Often times, this subjectivity can lead to great amounts of confusion and general miscommunication. Should we not expect, then, that there exist those willing to exploit such difficulties? Of course we should. This effect is seen everywhere, and we are not surprised to see it again in issues relating to software quality.

Chances are, if you haven't been asleep for the past fifteen years, you've heard the word "beta" used in a sense other than radioactivity or Greek literature. Particularly, you've probably heard it in reference to the development status of software. Historically, the progress of software progress has been described in terms of a progression from planning, feature-incomplete implementation, feature-complete implementation, tested implementation and maintained implementation. Though the precise terms used to describe these phases depends heavily upon whom you ask, there is little disagreement on the five-phase system. Some may split or divide, but these five seem to be common across categorization systems. One of the more common codification schemes involves referring to the second and third phases (call them Phase 1 and Phase 2 for now) as "alpha" and "beta," respectively. Thus, "beta" is quite often intended to refer explicity software which does everything it should, but is likely to be buggy, and in need of through testing.

At this point, many readers would do well to stop and think of what they've seen the term "beta" applied to. Gmail, which has gained no small amount of features, has been in "beta" since its inception. Internet Explorer 7 Beta 2 is quite obviously not feature-complete, nor is the documentation even close to adaquate. Thus, we see that in both cases, the term "beta" has been shifted from meaning "feature-complete but buggy" to "people won't use it if its called alpha" in the first case, and to "we don't care enough about quality to have a bugfixing phase" in the second. Of course, the idea of distinct phases gets a bit blurry with a feature-incomplete production application. Google can't afford to have major security flaws in a production application, whether they call it beta or not. Thus, I would propose that Google fix the problem by simply choosing another word to use other than "beta." For Microsoft's part, calling Internet Explorer 7 Beta 2 is nothing short of a boldfaced attempt at deception. The documentation and visual styles are far from complete, and most dialogs are completely different than most others, leading to the realization that Microsoft is not really even trying at making the beta phase a quality control phase.

In summary, please be aware of what the word "beta" means, and don't let it sucker you into ignoring large and important faults of a product.

technorati tags: , ,

Friday, April 07, 2006

More Trust Models: To Trust Telecos and Governments.

As discussed in the article "AT&T Forwarding All Internet Traffic to NSA?," the EFF alleges that AT&T has been forwarding that traffic which passes over their lines to the NSA. In keeping with my recent obsession with trust models, I shall raise an important question: to what degree should one's telecommunications provider and one's government be trusted? The most obvious answer seems to be to not trust either at all.

Dealing with each in turn, let us consider the role of a teleco in a trust model. A teleco sells a very specific service: to connect you to the Internet. Nowhere in this is the guarantee that they have the human decency to keep the data which you trust to their networks reasonably secure or private. Though some telecos may give you this decency, there is no compelling reason to assume that they will prevent unauthorized access to your data. Rather, the very people you least desire to have access to your data will seek to integrate themselves with a teleco, just as a pedophile might find access to their victims through a position in a police organization (as seen in the recent Department of Homeland Security child sex scandal). Thus, in one of those many ironies which permeate throughout information security, a teleco should be distrusted by default. How do you deal, then, with securing your data over what is, fundamentally, an untrusted network? For that, cryptography again comes to the rescue. A trust model which assumes a base distrust of the network itself will promote the use of end-to-end encryption. Oh, would that this were the case in practice.

Moving on to trusting a government, let us reflect upon words of wisdom from the Federalist Papers, No. 51, written by Alexander Hamilton:

But what is government itself, but the greatest of all reflections on human nature? If men were angels, no government would be necessary. If angels were to govern men, neither external nor internal controls on government would be necessary. In framing a government which is to be administered by men over men, the great difficulty lies in this: you must first enable the government to control the governed; and in the next place oblige it to control itself. A dependence on the people is, no doubt, the primary control on the government; but experience has taught mankind the necessity of auxiliary precautions.

A careful examinations of these words reminds us that at its most fundamental, a government is a response to imperfections in the human condition. Unfortunately, however, that response is in itself forged from the same flawed humanity. At a practical level, we are again reminded of a very basic axiom of trust:

The positions most requiring of trustworthiness are sought out by those most apt to abuse that trust.

Put differently, the positions that we create to deal with issues of trust and crime are the most desirable to those intent on violating that trust. As I have already mentioned, a position in a police organization is highly desirable for a criminal, so is not a position in lawmaking most desirable for a lawbreaker? How, then, can we ever trust our own government to be responsible with our data? We cannot if we wish to have any expectation of security. Government can secure us from each other, but it can never secure us from itself.

It is thus seen that the recent allegations by the EFF represent yet another failure to apply sane trust models to every aspect of our lives. Instead of harboring a base distrust of our communications providers and our governments, we explicitly place large amounts of trust in them. Though this by no more excuses the alleged crimes than leaving an expensive car unlocked excuses its subsequent theft, we should likewise not be at all surprised that, when we are so naive as to trust our governments and telecos, our trust will be violated in the most profound sense.


technorati tags: , , ,