Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What Is Code? (2015)

bloomberg.com

Post reply on HN
Print view View original

41–50 of 54 posts

Re: What Is Code? (2015)

#41

Post by ge96 »

ldh0011 wrote: post #7
periram wrote: post #4

Earlier quoted context omitted.

I was a little skeptical. Repeated the steps. Indeed it asks for camera access.

Looks like it is to take a photo for the certificate at the bottom of the page https://github.com/BloombergMedia/whatiscode/blob/master/mod...

Oh I see. Yeah when you do a ctrl+f it doesn't say something like "let us take a picture of you" but I do see the cert at the bottom/placeholder when you decline.

Suppose it could be a wow factor/adds to the interactive elements of the page.

Re: What Is Code? (2015)

#42

Post by bmitc »

the_af wrote: post #29
bmitc wrote: post #26

This is a somewhat myopic and outdated description of what "code" or "software" is. I'm writing an article on this, but software is a much broader idea and domain than just getting computers to do something. At a minimum, it includes getting a computer to do something, communicating between humans and not humans and computers, and encoding a domain. All this adds up to software being "executable knowledge", which mak…

Are you equating "computer science" with "coding"? Because it's really not. CS studies systems, complexity, requirements, domain knowledge, etc. I'm not arguing against people from other fields bringing valuable insight; I'm disputing that CS doesn't also involve the study of complex heterogeneous systems or that it is about "coding". Seems like you're describing a bootcamp or coding school instead.

No. Why?

Re: What Is Code? (2015)

#43

Post by the_af »

dagw wrote: post #40
the_af wrote: post #38

Earlier quoted context omitted.

> "They get some token attention but they're always subservient to business values" Hard disagree. CS education doesn't have business at its core, but science . It's all about systems design, theory, formal verification, requirements gathering, math, logic, etc. It's neither about (primarily) business nor coding concerns. You probably don't need CS for those.

Isn't the core problem that universities have started offering far too much under the "Computer Science" umbrella. If they where to split it up into 5 or 6 different degrees, like Computer Science, Computer Engineering, Software Engineering, Human Computer Interaction, Business Systems design etc. then many of these problems of definition and meaning would disappear.

Definitely, multiple unrelated things are being "sold" as CS.

Re: What Is Code? (2015)

#44

Post by ldh0011 »

ge96 wrote: post #41
ldh0011 wrote: post #7

Earlier quoted context omitted.

Looks like it is to take a photo for the certificate at the bottom of the page https://github.com/BloombergMedia/whatiscode/blob/master/mod...

Oh I see. Yeah when you do a ctrl+f it doesn't say something like "let us take a picture of you" but I do see the cert at the bottom/placeholder when you decline. Suppose it could be a wow factor/adds to the interactive elements of the page.

Yeah it wasn't as immediately obvious to me what was going on as I think the author expected, seeing something ask for my camera unexpectedly is alarming! But the article was otherwise pretty great.

Re: What Is Code? (2015)

#45

Post by bmitc »

the_af wrote: post #38
TheOtherHobbes wrote: post #33

Earlier quoted context omitted.

CS doesn't. At least not in any useful way. Systems Analysis and Knowledge Engineering typically assume corporate/industrial philosophies and goals. Broader human considerations - political, social, cultural - are very peripheral in CS. They get some token attention but they're always subservient to business values. The idea that software is primarily about human relations of all kinds is novel and rather subversive.

> "They get some token attention but they're always subservient to business values" Hard disagree. CS education doesn't have business at its core, but science . It's all about systems design, theory, formal verification, requirements gathering, math, logic, etc. It's neither about (primarily) business nor coding concerns. You probably don't need CS for those.

Computer science is not a science. Computer science typically bounces between mathematics and engineering, with some unique aspects of its own. But it definitely isn't a science, which typically needs a loop of hypothesis, observations, and theory.

Re: What Is Code? (2015)

#46

Post by galaxyLogic »

tejuis wrote: post #34
galaxyLogic wrote: post #10

A great all around introduction to programming. It poses an interesting question which I think it doesn't really answer well: "What Is the Relationship Between Code and Data?" I can easily understand what a "program" is. It is a set of instructions we give to the computer for it to do something we want it to do. But what is "data"?

Data is information about things and code is information about how to transform data. Every program transforms input data to output data. Code is interpreted by the computer. Data and code are encoded to bits for the digital computer.

That sounds enlightening to me.

Code is interpreted by the computer, Data is interpreted by Code !!!

But when we say "code is interpreted by computer" we are really saying "Code is interpreted by Code" right? Meaning the interpreter or compiler.

So how come this snake doesn't eat its own tail?

The answer, the MAGIC of computers, is that at the lowest level somehow Hardware is able to interpret Code .

Re: What Is Code? (2015)

#47

Post by galaxyLogic »

Jtsummers wrote: post #15
galaxyLogic wrote: post #10

A great all around introduction to programming. It poses an interesting question which I think it doesn't really answer well: "What Is the Relationship Between Code and Data?" I can easily understand what a "program" is. It is a set of instructions we give to the computer for it to do something we want it to do. But what is "data"?

Data are instructions for the computer interpreted in the context of a program. Or no program in the case of instructions which can be interpreted directly by the computer. That is, data are programs, programs are data. Just need to establish the environment in which to execute/process them.

Right, except what got me thinking is that "programs" can be conceptually seen as a sequence of "instructions" to the computer. Whereas it is more difficult to think of data as "instructions". Data is more like a model of the world, than a set of "instructions".

Re: What Is Code? (2015)

#48

Post by galaxyLogic »

akomtu wrote: post #23
galaxyLogic wrote: post #10

A great all around introduction to programming. It poses an interesting question which I think it doesn't really answer well: "What Is the Relationship Between Code and Data?" I can easily understand what a "program" is. It is a set of instructions we give to the computer for it to do something we want it to do. But what is "data"?

If code is a relationship between the next state and the current state, then this state is data.

Well said. Code stays as it is, typically. Data changes.

Of course code can modify its own 'code". But that makes it much more difficult to reason about its behavior. It is good to have some "invariants". If we can assume that code does not change except when the human author changes it, then it becomes much easier to reason about the changes the program may cause to the data.

Re: What Is Code? (2015)

#49

Post by galaxyLogic »

js2 wrote: post #16
galaxyLogic wrote: post #10

A great all around introduction to programming. It poses an interesting question which I think it doesn't really answer well: "What Is the Relationship Between Code and Data?" I can easily understand what a "program" is. It is a set of instructions we give to the computer for it to do something we want it to do. But what is "data"?

The Harvard architecture makes the separation clearer: https://en.wikipedia.org/wiki/Harvard_architecture

And Turing machine? Can it modify its own tape?

Re: What Is Code? (2015)

#50

Post by galaxyLogic »

atoav wrote: post #22
Tomis02 wrote: post #18

Earlier quoted context omitted.

> But what is "data"? It's a sequence of bits. You take reality (or whatever you want) and represent it as bits. That sequence of bits is called "data".

In the computer science world yes. But the word also exists outside of that context (and did so long before computers). Data is the plural of datum. A datum is a piece of information. Of course not any piece information is automatically a datum, as that noun implies a certain structured nature or a normative perspective on that piece of information. Certain information only became commonplace because it became a datu…

I would say that data is "encoded information". But then again is there any other kind?
Post reply on HN