kdtop

The problem with chatbots is one of an unfulfilled promise, an unrealistic setting of user expectations. An empty Google search window doesn't pretend to be a human being -- although it is actually very powerful. Clippy, on the other hand, has a human face and pretends to be an intelligent agent, only to fail. So chatbots, when seeming to act like a human via natural language, are held in the user's mind to a higher standard. Siri, in my iPhone, seems to strike a better balance. It takes my spoken input and gives replies (albeit not great replies). But I never get the feeling that it is trying to carry on an extended conversation with me.

That's not completely true. MacOS has started this process where all apps are checked for safety prior to launching. It makes app launch seem slow, which feels like an app problem.

Outsider here. What is Strava? How widespread is it's use?

I love a clear explanation by someone knowledgeable. :-)

I'm just glad Gilligan didn't run through the markings and mess it up!

As a non-computer science person, I feel very dumb right now. :-(

I couldn't help but think of this XKCD comic: https://xkcd.com/1217/

This study, not to be minimized, works in a cell culture. That is a long way from being an effective treatment in actual humans.

I agree that this was a bit confusing. Higher up in the article, it shows that a linear function is one that doesn't change when scaled:

F(a * x) = a * F(x)

This is showing the relationship between two uses of the same function.

Then, further along, we find:

"So, what types of functions are actually linear? Plain-old scaling by a constant, or functions that look like: F(x)=ax In our roof example, a=1/3"

I think in this second situation, F(x)=ax is not a relationship but rather a DEFINITION of the function F(x).

In programming terms:

function F(x: real) : real;

begin

  Result := x * (1/3);

end;

Are you my twin brother from a parallel universe? My experience has mirrored yours.

I spoke with a blind woman this week who loves talking with her smart speaker, Alexa. "She's quite a character." I imagine that some lonely people might get comfort talking to their house plants, so a smart speaker that can actually talk back, however limited, would be a great improvement. So a chatbot can be useful even if zany at times. The big issue is how user expectations are set up. I'll put up with a digital assistant trying its best to help process my query with much more tolerance than a chatbot trying to pretend it is a human. Researchers are trying to narrow the gap between these two. But for now I will enjoy the former and be annoyed by the later.

Eternal truths:

* New things are typically more popular than older things.

* People not using ObjectPascal / Delphi love to hate on it.

* People using it feel outraged at the injustice, pointing out it's merits.

I have been using Delphi for about 30 yrs it seems. I have a love / hate relationship with it. Love the language and the ability to rapidly get the program I want created. Hate that version after version, the IDE STILL has stupid bugs; that the built-in help seems to get worse and worse with time (I don't want a slow web browser showing me Microsoft C calls and all sorts of irrelevant nonsense), and get very annoyed when having to get required custom components installed before being able to open a project using the components. And I REALLY dislike that they charge so bloody much for it. For business reasons I had to purchase a license that cost a few thousand dollars. Like 2 weeks later, sales people were calling me encouraging me to spend a similar amount to upgrade to the newest version that had just come out. They must have only a handful of paying customers and depend on the few to keep their entire business afloat.

Isn't this a bit like saying the last release of Ford's Model T was back in the 1930's? It is now called C++ Builder, and it is still regularly released. https://www.embarcadero.com/products/cbuilder

As I recall, c# was developed by one of the primary designers of object pascal, Anders Jejlsberg, after being poached away by Microsoft. (https://en.wikipedia.org/wiki/Anders_Hejlsberg)

Fascinating read. I wonder if the remaining wrecks will ever be raised?

I figure the old adage still holds: "If you have to ask, it's too much."

For the newbie here, can someone tell me if this is open-sourced? Many are fearful of building on a MS platform in case they change their minds in the future etc?

Has anyone been able to find if there is any further progress on this site in the decades that have since passed after this article was published?

I came to say how horrible the formatting of the original post is. But this site fixes that. Much better! Thanks!

I was taught that salt (sodium) unmasks hypertension, but doesn't cause hypertension. In other words, sodium is the tool that the body's regulatory systems use when trying to raise blood pressure. So a healthy younger person can probably not worry about sodium intake. But for an older person with congestive heart failure, a bag of potato chips might end them up in the hospital.

This looks like an empty github repository. Am I missing something? ... Oh, is this a joke? I'm slow on the uptake.

Outstanding article

Do you have a link to such a comparison? I would be surprised if similar tasks on similar hardware are different. Seems like I have heard that head to head comparisons are against the terms of service of Cache'.

I work with M/MUMPS on a daily basis. It's no surprise that it is a primitive language, as it was developed in 1966 to run on PDP-7 computers, less powerful than a raspberry pi. But M is still very relevant today because it is the power behind many (most?) of the electronic medical health record (EHR) systems in the world: Epic, Meditech, VistA (from the VA), Allscripts (more here: https://en.wikipedia.org/wiki/MUMPS#Current_users_of_MUMPS_a...). Note that it is also extensively used in the banking industry. I have heard it said that at least 50% of people have their money run through a Mumps financial system during electronic fund transfers. Mumps is very fast and very stable, IMHO because of it's lack of bells and whistles. GT.M's version of Mumps is serious about it's moto of "Rock Solid. Lightning fast". (https://www.osehra.org/content/fis-gtm).

The most important part about any computing system/language is the resulting application. Yes, it is nicer for programmers when they have advanced language features. But ultimately, the test of how "good" a language is should really be about how useful it is, and how widely it is used.

The wheel is very old technology and yet still very important.