HNHacker News
TopNewBestAskShowJobs

ggchappell

2,759 karma · joined February 27, 2009

Glenn G. Chappell

I'm a C.S. prof. at the University of Alaska. If you want contact info, Google finds me pretty easily.

submissionscomments
ggchappell··on A Disadvantaged Start Hurts Boys More Than Girls
Interesting point.

So you're saying that "girls" is not really being used as a noun phrase, but rather as an abbreviated clause. Thus we use the conjunction "than", while we would use the preposition "from" with a real honest-to-goodness noun phrase.

"Than" is a preposition, too, but only when referencing a comparative. So, "A cow has more than one stomach." But, "Cows are different from people." OTOH, I guess you would prefer, "Cows eat different food than people," (?) since "people" in that sentence is really short for "people eat".

I'll have to ponder this. Thanks for a thought-provoking comment.

EDIT. And perhaps the source of all the actually poor uses of "different than", is that many people mentally classify "different" as a comparative.

ggchappell··on How the Internet Has Changed Bullying
I find this article to be very troubling.

Because there is something the article does not mention: that bullying often happens because the victim cannot escape. Bullying is an ineffective strategy when the victim can get away and/or has no good reason to stick around, so it is done mostly when there is no escape. That is why bullying is so prevalent in environments where people are trapped with a fixed group: public schools (in the American sense), prisons, etc.

FTA:

> But getting away from it has become more difficult.

And that means it will happen more often. It will also be more damaging when it happens.

By the way, I originally saw this idea -- that inescapability is what leads to bullying -- in one of those popular economics books (Freakonomics? The Undercover Economist?). It jibes well with my own experience.

ggchappell··on Elsevier, that just freaked me out
I don't worry much about inserting wacky storage devices, as I run Linux, and I don't have any kind of auto-run enabled.

But this would still work on my machine, right? (I mean that the key combo would be entered; it might not actually bring up the web page, depending on my setup.)

ggchappell··on What Americans know and don't know about science
This is interesting and worthwhile, but I do wonder about some of the questions.

For example, the core-mantle-crust structure of the Earth is fundamental. Everyone ought to know about that. But whether the core is the hottest part? This strikes me as highly non-obvious, even to a relatively knowledgeable person. The article says people were shown a labeled picture; I bet most people just guessed, based on the colors used for the various regions.

And then there is the question about uranium & nuclear energy/weapons. What about plutonium? Sure we can make plutonium from uranium, but then we're getting into subtleties of what "needed" means.

Lastly, it is nice to know the role of Jonas Salk in the development of polio vaccines. But that is more history than science (and a lot of people receive the Sabin vaccine ...). Better questions might address what a vaccine is made from, or roughly how it works.

ggchappell··on The Most Misread Poem in America
Yes, rather like the checkout lines at the local grocery store. Perhaps we should change the last line of the poem to "And that has made everything the same."
ggchappell··on The Web Can't Last for Ever
> On a large screen browser plain HTML is barely readable.

It isn't? One of us is missing something here.

ggchappell··on Computer Sciences Courses That Don't Exist, but Should
We have that in our CS program. We call it "Senior Capstone". I'm sure others have it, too.
ggchappell··on World Airports Voronoi
Very nice!
ggchappell··on Parable of the Boy and the Steam Engine (2014)
Can anyone explain the part about "ignorance"?

> McLuhan cites the parable as a productive use of ignorance and indolence, ....

I can understand indolence. This is a pejorative term, but it basically means the desire to avoid doing work. Certainly productive use was made of this.

But ignorance is lack of knowledge. I do not see how the boy made productive use of any lack of knowledge. Rather, he made productive use of his knowledge of what needed to be done, and that it was always the same thing.

ggchappell··on Ants are as Effective as Pesticides
Yes, here I am tossing around the word, and, to be honest, I cannot say exactly what it means. I suspect that a species would be counted as native Hawaiian if it was found on the island chain before the coming of humans.
ggchappell··on What Is the One Fact Humanity Needs to Know?
> Humans naturally seek reasons why they are right, but to understand the universe, you must seek reasons why you are wrong.

Fantastic! Is that original?

ggchappell··on Ants are as Effective as Pesticides
Just a wee bit off-topic, but:

The situation with the native species of Hawaii is actually much worse than most people are aware -- even people who live in Hawaii. I read somewhere[1] that, if you walk around a typical coastal town in Hawaii, and look at the lawns & gardens & parks, and watch the birds & whatnot, that you will generally not see any native species at all. The ones big enough to see are all confined to the interior.

EDIT. And, by the way, the rats weren't native, either. Hawaii has exactly two native mammals: the Hawaiian Hoary Bat and the Hawaiian Monk Seal.

[1] I think it was in Remains of a Rainbow, by D. Liittschwager & S. Middleton. https://amazon.com/dp/0792264126/

ggchappell··on San Francisco Firefighters Become Unintended Safety Net for the Homeless
Comments on web forums often make it seem as if this phenomenon is universal. So, I thought I would point out that where I live (Fairbanks, Alaska), we have a healthy, vibrant public library that lots of people go to for books.
ggchappell··on My wife found my email in the Ashley Madison database
Yes, indeed. But, alas, this was 20 years ago, and the messages in question are long gone. At the time I was not thinking of humor value for posterity; I was just annoyed.
ggchappell··on My wife found my email in the Ashley Madison database
Yes, that was eventually the big tip-off as to what had happened. I would get messages like, "Hi, Fred! Glad to see you're settled in at BigU. Who's this 'ggchappell' guy?" (My name isn't "Fred", BTW.)
ggchappell··on A Word Is Worth a Thousand Vectors
This is interesting stuff. I recall that at one time Google seemed to be heading in somewhat similar directions with Google Sets (now sadly gone -- I miss it).

I know that the author is looking squarely at use cases along the lines of a recommendation engine that would replace a human expert. But personally, I think it might be more interesting to examine things the algorithm can do that humans would find difficult or unintuitive. Sure, king - man + woman = queen is a very significant achievement; it's also obvious, to a human. Now, what can this algorithm come up with that is worthwhile, but that I would not find so obvious?

A couple of little comments:

> The algorithm eventually sees so many examples that it can infer the gender of a single word, ....

Do we really want to say that? Perhaps we should say that the algorithm is eventually able to make inferences that people would make based on knowledge of the gender of words -- which is not quite the same thing. (And again, I ask: what useful inferences can the algorithm make that humans would not make so quickly?)

> Despite the impressive results that come with word vectorization, no NLP technique is perfect. Take care that your system is robust to results that a computer deems relevant but an expert human wouldn't.

It should be noted that that "no NLP technique is perfect" idea applies to the NLP techniques used by human brains.

ggchappell··on My wife found my email in the Ashley Madison database
Okay, time for my story:

About two decades ago, while going through new-employee orientation at a university, I sent some e-mail from a computer lab. As part of this, I entered my e-mail address as the reply-to address. At that moment, there was some kind of misconfiguration: the directory on the university-wide server where global defaults were stored happened to be writeable.

A couple of weeks later, when all the students showed up, my e-mail address was the default reply-to address on the standard e-mail client in all university computer labs. And because the configuration problem had been fixed -- the directory was no longer writeable -- my address remained the default address.

It took me a while to figure out what had happened. For several days, all I knew was that suddenly I was subscribed to every mailing list in the universe.

When I did figure it out, I told the IT people, and they fixed it. But then -- I'm not making this up! -- there was some kind of crash, they had to restore from a backup, and it happened all over again.

There was no serious negative impact from this episode; I just got tons of wacky e-mails for a while. Perhaps things might be different if something similar were to happen today.

ggchappell··on Intermediate Python
> If you don't need the laziness (in Python 3 only), you probably shouldn't use them ....

And if you do need the laziness, then you can use generator comprehensions.

ggchappell··on Brains Sweep Themselves Clean of Toxins During Sleep (2013)
Indeed. But there is also an easy test that can be applied to people that talk about "toxins": see if they can name a toxin. And both the article and the original paper[1] discuss beta-amyloid.

In any case, it appears that the blame for the suspect terminology lies primarily with the NPR writer, not the researchers. For example, the word "toxin" does not actually appear in the paper[1], which is titled "Sleep Drives Metabolite Clearance from the Adult Brain".

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3880190/

ggchappell··on The Hamburger Menu Doesn't Work
Good point.
ggchappell··on We Still Let Harassers Participate in Our Community
A worthy post.
ggchappell··on For only the second time in our history the ownership of The Economist changes
> The Economist is a prime example of how for-pay curated content can work and can work well to serve a market... a wide market.

It certainly is. But I wonder how much their strategy can really serve as a model for others.

On the one hand, the Economist might be seen as saying, "Here is how to do it." Others can follow and similarly prosper, with increasingly large markets being served.

Or, on the other hand, the Economist might simply be meeting all of the demand for a publication like itself. And the only way for another company to do as well using the same strategy, would be to produce a publication that is convincingly better than the Economist, thus putting the original out of business.

ggchappell··on Commuting Kills
I don't think there's any problem with citing an official Census Bureau report. Certainly they are the definitive source for statistics on the U.S. population. And I imagine the numbers in the report give a correct overall impression of commute times. However, I suspect the conclusions in the report as to the accuracy of the data are a bit overblown.
ggchappell··on Commuting Kills
Long commutes are of course a problem.

However, I am not sure the analysis method used in this article is really the way to look at this issue. The fact that lots of people spend 15 minutes sitting on their rear is not a serious problem, despite the fact that you can multiply that 15 minutes by millions of people to get a large number.

The problem is that some people have very long commutes. 30 minutes or more is a problem. An hour or more is a serious problem. 90 minutes or more is, for most jobs, ridiculous.

Actually, I see rather more good news than bad news in the referenced statistics. According to the source ([1], see the table on page 2) only 2.3% of Americans have a one-way commute of 90 or more minutes. Only 7.5% have an hour or more. But 33.5% -- more than a third -- have 30 minutes or more; that is troubling.

By the way, the statistics in the source[1] strike me as a little suspect. In particular, the big bump at "30 to 34 minutes" suggests that a lot of people were thinking, "Oh, about 30 minutes." That kind of estimation is not mentioned among the possible sources of error. Someone was not thinking clearly enough when that Census Bureau report was written.

[1] https://www.census.gov/prod/2011pubs/acs-15.pdf

ggchappell··on Mathematical software wish list
> The coolest piece of mathematical software I’ve come across in the last few years is Chebfun, http://www.chebfun.org

Oooh, nifty! I've never heard of this. Wonderful idea.

I'm thinking this really needs to be ported to Python.

ggchappell··on CS for All
Sounds good.

In fact, I think we are trying to solve them simultaneously.

ggchappell··on CS for All
This looks like it would make for a very nice course -- a good middle ground between the hardcore practical course ("let's learn C") and the hardcore theoretical course ("let's do denotational semantics with Scheme" -- or whatever).
ggchappell··on The Hamburger Menu Doesn't Work
Dare I suggest that the "gear" menu, so ubiquitous on Google pages these days, suffers much the same problems as the "hamburger"?
ggchappell··on CS for All
How about we try to solve them simultaneously?
ggchappell··on Ask HN: The Internet is getting lame. What's next?
> But Usenet had threads, ....

Yes, using the "References:" and/or "In-Reply-To:" headers. Later clients tended to do this, I think.

> ... to get the benefits.

Yes, and the drawbacks: in-depth discussions involving multiple people were harder to come by.

In any case, in both Usenet -- late in the game -- and many mailing lists, there are relatively few participants, so scaling is not really an issue.

← PreviousPage 4 of 34Next →