Cake
  • Log In
  • Sign Up
    • I heard a short debate about this on NPR the other day and it just felt right.

      My view is when we use a very important and very loaded term, everyone should be able to understand what it means whether it's in science or any other field.

      For example, what's a carb? Is an apple a carb? Broccoli? A donut? Does anybody really know what people mean when they use the word?

      The term statistically significant has come to be one of those unclear terms few people understand that carries so much weight. There is a very thin line that separates statistically significant and not. Yet one means true and the other false in most minds.

      If nothing else, talking about it to understand that it doesn't mean what most people think seems like a great thing. 🙂

    • Whether the topic was D&D in the late 70s, or "hacking" in the early 70s or "end of the century" in the 1990s, it has been my observation that most of the news media fails to learn the inside story before they start reporting. Just as "hacking" became confused in the minds of the public with "cracking" due to the typical reporting and just as Y2K and "end of the century" were also misreported (Y2K began on January 1, 2000, the twentieth century ended on December 31, 2000) so it is also with this situation. Scientists often complain that after they release a scientific report that the news media completely botches the story and then the scientists have to explain that the report did not say what the media claimed that it said.

    • Second comment

      Why does the average person think that a kilobyte is 1000 bytes? America and its media is less about getting the facts straight and more about how Americans feel about things.

      Sergeant Friday, where are you?

    • When I was in the book industry, the most popular authors shared a dictionary of about 7,000 words that they tried to limit their writing to. They were the words that American English speakers clearly understood without confusion.

      Sometimes I think words are chosen to create confusion. My family in Southern Utah loves the Affordable Care Act but despises Obamacare.

    • I'm not sure if the "average person" really thinks that - simply because many people don't really care about that "complicated physics business" where kilo- is a prefix to denote "times thousand". It would be an interesting idea to ask a thousand people on the street how long a kilosecond is (or, similarly, name a city that is a megameter away or an event that happened a gigasecond ago).

      Second, the fact that kilo- is used to mean both "times thousand" (10³) and "roughly times thousand" (2¹⁰) doesn't really help.

      With that in mind, I wonder how much of the initial problem of "significance" is really something that should be solved by no longer using the mathematical concept in scientific contexts.

    • With that in mind, I wonder how much of the initial problem of "significance" is really something that should be solved by no longer using the mathematical concept in scientific contexts.

      Both the Nature commentary and The American Statistician issue were directed at scientists, not the general public. @Shewmaker is correct that science journalism is frequently misleading and sometimes simply incorrect, but that is a separate question. The sad fact is that outside of physics, many otherwise perfectly competent scientists have an imperfect understanding of statistics--what they mean and what they don't. I think it's healthy that a correction may be forthcoming.

    • Both the Nature commentary and The American Statistician issue were directed at scientists, not the general public. @Shewmaker is correct that science journalism is frequently misleading and sometimes simply incorrect, but that is a separate question.

      Yeah, but I feel that this whole thing is running into the exact trap it is complaining about. Just have a look at the illustration that is used to head the Nature article that supposedly is "directed at scientists".

      This illustration seems to imply that "statistical significance" is a concept comparable in absurdity to one-eyed mammoth:

      or the existence of a "fire element":

      or some "quintessence" filling the universe:

      That just isn't true, though. "Statistical significance" is a very useful tool, just one that has been used (or abused) for things that it hasn't been created for. The problem is not that the outcome of a scientific study is considered to be more or less likely to show a "real difference" - the problem is that a magic number exists, and that some people blindly claim that any study not reaching that threshold must mean that no actual difference exists.

      It would be wrong to "abolish significance", as the title of this conversation states - it just shouldn't be used out of context.

    • Two to the power of ten will always be 1024 no matter what the majority think.

      There is an old story in which the question was asked "If we call a calf's tail a leg, how many legs does the calf have?"

      The respondent said "Five"

      To which the questioner responded "The calf will have only four legs. Just because we call a tail a leg does not make it a leg."

      Just as King Cnut could not stop the tide by royal decree so also "the people" cannot make binary into decimal no matter how much more convenient it might seem.

    • I agree that the usage of the prefixes "kilo" and "mega" prior to the age of the personal computer was a poor choice. But just as a byte cannot be defined as 10 bits, so also a kilobyte is 2¹³ bits and the fact that it is not easy for the public to process that concept does not change the facts.

    • Conventions change and it’s important to stay current. I asked Google Search and it said 1,000 bytes. That wasn’t a qualified answer but instead it’s the official answer of a company that makes it its business to stay current. (See pic below.)

      Ultimately, it comes down to what is useful to survive and succeed in your environment. For example, If you look at national political polls, you will see that one candidate received 28% of the votes. In reality, it may actually have been 28.35% of the votes, but that level of precision is unnecessary and makes it more difficult to quickly digest statistics in the news. So perhaps the better question is,

      How does ignorance of the number of bytes in a kilobyte negativity impact your life?

      ⬇️

    • I'll just post a little old-fart rant about how people seem to forget that it was ignorance that killed the cat, and curiosity was framed. We live in a day when vast amounts of knowledge humanity has amassed is literally at our fingertips, almost instantly. And yet an average dictionary seems to be diminishing, and the very idea of looking something up seems baffling.

      Years back, I stumbled on a word "arcology" in one of Gibson's novels. None of dictionaries available to me helped (I think I might have tried an Altavista search, or this might be a false memory - if I did, it didn't yield anything). I couldn't find Mr. Gibson's email (apparently, he didn't have one then) so I emailed his comrade-in-arms, Mr. Sterling, at his W.E.L.L. address, and he politely and succinctly explained the word in a reply.

      English isn't my mother tongue, I just enjoy it and Gibson's prose as a great example of it, and was genuinely curious. Today, it would have taken me maybe 3 minutes tops to get the explanation. Where's the people curiosity gone to?

    • it was ignorance that killed the cat, and curiosity was framed.

      I loved this. Had we been in a bar, I would have bought you a beer to celebrate your mastery of English. But since I'm just in my living room and on the Internet I was able to find that it came from C.J. Cherryh, a writer I was not familiar with. It also took less than 10 seconds to find out what arcology meant, which I also didn't know. I sure am glad to have Google--it certainly satisfies curiosity and used with caution dispels ignorance as well. OTOH, corresponding with Sterling to ask about something in Gibson is pretty cool, not to mention resourceful.

    • Regarding C.J. Cherryh: In my opinion her three best series are 1. The Chanur series (This is the earliest of these) 2. The Fortress series (This is in the Fantasy Genre.) and the on-going Foreigner series.

      Considering the level of writing which she has achieved in the past, I am occasionally disappointed in some of the novels within the Foreigner series, but as a series, it is very good.

      You may notice in reading these (and some of her other writings) that difficulties in communication between people of different cultures and differing paradigms is a recurring theme in her writing.

    • You certainly fooled me that English wasn't your first language. You write English sentences than most of us who are native speakers.

      I have to admit that having Alexa near where I work has caused me to shout across the room things like "Alexa, what does arcology mean?" That even though I'm sitting at my keyboard. 😳