« first day (1252 days earlier)      last day (3742 days later) » 
00:00 - 21:0021:00 - 00:00

Ampt
Ampt
9:00 PM
> Blah?
aha
 
user55340
The '@' form of reply doesn't work, but the :message form does work... and works with oneboxes too.
 
user55340
 
user20683
@MichaelT not sure re: flag. Did not see "evidence" but questions are very suspect cross the network.
 
user55340
@WorldEngineer I'd put money on it. Its the same set of accounts as before.
 
user20683
@MichaelT you'd almost certainly lose money
 
user55340
9:04 PM
 
user55340
 
user55340
He may have gotten better at masking is identity... he's had quite a few failures at that in the past.
 
user55340
But look at the accounts, look at the questions...
 
user20683
not convinced
 
user55340
(he's even provoked a @RobertHarvey answer...)
 
user55340
9:06 PM
0
A: How do we go from assembly to machine code(code generation)

Robert HarveyAssembly opcodes have, for the most part, a one-to-one correspondence with the underlying machine instructions. So all you have to do is identify each opcode in the assembly language, map it to the corresponding machine instruction, and write the machine instruction out to a file, along with its...

 
user20683
we shall see
 
user55340
which is one of my "this is likely to be Andy" warning signs.
 
user20683
fair enough
 
user20683
still taking it slow
 
Ampt
Ampt
he's got a physics account too
 
user20683
9:08 PM
@Ampt so do I
 
user55340
I'd appreciate it if you could pass it up the chain, though I can only ask that you do.
 
Ampt
Ampt
@WorldEngineer I mean that that is a part of his MO. Physics genius.
 
user20683
@MichaelT already done
 
Ampt
Ampt
I just want to know what he's working on that has such a wide range of stuff he needs to know
 
user20683
@Ampt C code
 
user20683
9:13 PM
look at his SO question
 
Ampt
Ampt
-1
Q: How do photons 'connect' during wireless connection

user12979So wireless router broadcasts a signal and then your device searches. So what actually happens when the photons 'meet' it's kind of like saying, 'ah your one of us, so we will follow you, show us the way' It's so bizarre, how do photons connect during wireless connection?

electromagnetism
-1
Q: Why can TV broadcasts send such large amounts of data(photorealism) and a PC cant

user12979Firstly I think I am right in saying that TV broadcast are sent via electromagnetic waves which means they are sent via photons, how is that even possible? And then the main questions, how can you broadcast such large amounts of data over the airwaves and produce photorealistic TV and a PC canno...

electromagnetic-radiation photons
1
Q: Can global Illumination via path tracing replace all other current lighting techniques?

user12979In the sense that you currently have algorithms like HDR, shadows, reflections, caustics, motion blur and so on, does complete path tracing take care of all these effects, or would you still have to implement these effects separately?

rendering lighting global-illumination
 
amon
amon
The key to that person's motivation is probably this question: gamedev.stackexchange.com/q/69522/30882
> As far as I can see, all that can be done in 2D/3D has been done or is know how to do, software algorithms are being replaced by hardware ones, the understanding of game engine subsystems is known, in fact Quake 3 was really the when things changed, we are waiting for more raw hardware power to reach more photo-realistic graphics that's all(well I think).
 
user20683
@amon Quake 3 was not when things changed but I digress.
 
amon
amon
assembler is a path he's investigating to get closer to “raw hardware”, the physics questions are about light, which he's interested in because of ray tracing and other stuff.
 
user20683
@amon His questions are much more focused than Andy's typically are
 
Jimmy Hoffa
Jimmy Hoffa
9:23 PM
@MichaelT You know what the funny thing is? He's like 15 or something - considering the topics and detail of his questions, he's probably going to be some extremely successful very smart software engineer in the future.
When his brain finishes developing the part that analyzes risks properly anyway
 
user20683
@JimmyHoffa I think I've started to grok Lambda Calculus and also Monads in a systemic way.
 
user20683
Monads do currying and partial application by treating state as a function
 
user20683
I think...
 
Jimmy Hoffa
Jimmy Hoffa
Though he fails to get answers here and is taken to just screwing with us, he has done so with the obsessionist fashion which in software engineering would be called 'rigor', further that same rigor is probably still being applied to his drive to find information which he is undoubtedly getting great heapfuls of from other sources around the internet by reading forums and countless pages of documentation
 
user20683
@JimmyHoffa the right mentor and he'd be fine. On the other hand, I've seen evidence that he's more of a pro troll than anything else.
 
Ampt
Ampt
9:27 PM
@JimmyHoffa If theres one thing I've seen in my college career, it's that it takes an astounding amount of raw talent to overcome even the smallest sliver of discipline.
 
Jimmy Hoffa
Jimmy Hoffa
@WorldEngineer Iduno. I understand monads in a simplistic way, I just think of them as a stack in a list that gets folded to stay flat - this is how things collapse. State is picked up by the monad which lives between the cells of the list and hands them off to accomplish the sequencing.
while your statement may make sense, I can't really make the analogical leap right now.
The lambda calculus bit though is good. Go forth and write JavaScript now ;P
(no not in forth, damnit....)
 
user20683
@JimmyHoffa I mostly need to memorize the DOM
 
user20683
that's really where I'm weakest
 
Jimmy Hoffa
Jimmy Hoffa
@WorldEngineer Bull. The DOM is a bloat of garbage, just use JQuery and when you learn the selectors tada.
no need to know the DOM past the selectors.
 
user20683
@JimmyHoffa well yeah but I'd rather understand the guts
 
user55340
9:31 PM
Jquery is the answer to all the questions.
 
Jimmy Hoffa
Jimmy Hoffa
@WorldEngineer Dont. They're as they have ever been, a mess.
 
user20683
@JimmyHoffa fair enough
 
user20683
gonna go deliver a talk about SE at 5:30
 
Robert Harvey
Robert Harvey
@MichaelT The hand-wavy, vague but essentially correct answer, vs. the precise, detailed, complete answer with graphs, charts and diagrams.
 
user20683
My girlfriend likes the handwaving
 
user55340
9:33 PM
@WorldEngineer Just remember to never look at Andy's area 51 question... wonder if he's gonna go onto that site when it goes public...
 
Jimmy Hoffa
Jimmy Hoffa
You want to know the dom? hit f12 right now, click the open javascript console button, type document.getElementById($('div').first().attr('id')) hit enter and then start inspecting that objects return. It's a pile of properties.
Knowing that pile of properties is not particularly useful beyond just memorizing what's there which is just memorization - always useless, but you already know what's there: events for things like click, keydown, keyup, mouseenter, mouseout, and an attributes collection for the variety of attributes. The DOM is really just an API over XML - that is, it's the JavaScript way of viewing XML. That's why jquery selectors work on arbitrary XML in HTTP responses etc.
 
user20683
I see
 
Jimmy Hoffa
Jimmy Hoffa
9:57 PM
List of time periods
The categorization of time into discrete named blocks is called periodization. This is a list of such named time periods as defined in various fields of study. Major categorization systems include cosmological (concerning the various time periods in the origin and evolution of the Universe), geological (concerning time periods in the origin and evolution of earth) and historical (concerning time periods in the origin and human evolution). Human time periods These can be divided broadly into prehistorical (before history began to be recorded) and historical periods (when written recor...
^--- I wonder what would be come up with if somebody tried to overlay this with the history of electronic computers
would we still fit into a single period? The pace of evolution in the space over it's history makes one think you could see correlaries to many entire periods fit into considerably smaller modern periods where the modern computer evolved in the course of 5 or 10 years equivalently to the amount of technological evolution you would see in an entire century of past periods
please don't break your declarations across lines. you'll give me histrionics again. — Jimmy Hoffa 7 secs ago
 
user55340
10:21 PM
Dedicated devices
	Machinery
		Ancient (abacus)
		...
		Modern (Babbage)
	Electronic (tabulation & calculation)
General Computation
	Governments
		Cryptography
	Instituions
		Mainframes
	Individuals
		Personal computers
		Mobile devices
 
user55340
The general computing days show the accessability of computing. Initially the only ones that had computers were goverements... then universities... then personal computer.
 
Jimmy Hoffa
Jimmy Hoffa
I'm just thinking, you have to fit public networking in there somehow but it has to exist across two - maybe three segments, uncommon place and common place, and the uncommon place could be separated by exclusive and inclusive - there was an uncommon period of internet existence where only unis and agencies had connections, there was an uncommon period where anyone could have connections but the vast majority didn't (like '85-'93?).
it's been one of the large backbones of technological evolution, the use of the PC alone in your PC section there - say '80 - '05 was absolutely massive, and largely differentiated across the growth of the internet, mobile devices largely came to be due to networking
 
Ampt
Ampt
10:38 PM
Love it when compilers specify float and double to be the same precision and size
why give me a double if it's not really a double
 
Robert Harvey
Robert Harvey
@MichaelT:
1
Q: asp.net MVC design pattern for consuming data from asp.net Web API

ataravatiI'm working on an application that will have both web and mobile interfaces. That's why I want to use Web API. I don't have any problem with the Web API part, but I don't know what to do with the MVC application. As far as I understand, my data access layer and Entity Framework will be in Web A...

asp.net-mvc asp.net-mvc-4 asp.net-web-api
 
Ampt
Ampt
> //Tickle the watch dog
 
Robert Harvey
Robert Harvey
Another "which pattern should I use" question.
 
user55340
10:59 PM
@RobertHarvey migrate it here and dup it to the one we just answered?
 
Jimmy Hoffa
Jimmy Hoffa
@RobertHarvey c'mon, that indecipherable page is a great traveling path for that indecipherable question...
Though to be fair, the question makes sense - a "smalltalk programming environment" is a very particular thing but it's still a crappy question
 
Robert Harvey
Robert Harvey
@JimmyHoffa: I'd like to talk to you about Lisp for a moment. Specifically Paul Graham's claims that he built a whole company around it because he was willing to wear Birkenstocks while everyone else was wearing dress shoes.
Question: What are the obstacles that prevent companies from using some sort of functional language as the core for everything they produce, rather than some more conventional (and popular) language such as Java or even C#?
I should really ask that question on the main site, but I think it would be closed as too broad.
 
user55340
@RobertHarvey Educational background.
 
user55340
It doesn't take as much mental gymnastics to understand procedural or OO based code.
 
Robert Harvey
Robert Harvey
Is it the Functional paradigm that is the big conceptual barrier, then? I expected someone to say "well, they're the shiz if you want to do something mathematical, but they're shit for GUI's." Or, something like that.
 
user55340
11:10 PM
I remember trying to learn ML for theory of computing class. It hurt. Lisp was a bit better...
 
user55340
but it was more of a "this is another way" but the school was focused on teaching C, compilers, databases, operating systems... and lisp didn't fit too well in there.
 
user55340
its even worse today with the... more vocationally based schools that are just teaching the raw language and libraries rather than the science behind them.
 
Robert Harvey
Robert Harvey
MIT used to teach their programming classes with SICP/Common Lisp, but stopped doing that a few years ago.
 
Jimmy Hoffa
Jimmy Hoffa
@RobertHarvey It's just the background. People just don't have any background or experience thinking in that paradigm.
 
user55340
So... you want to hire a new lisp programmer... you can hire a guy from MIT (from the days of old) or you can hire top notch polyglot, or you hire a programmer and spend N months teaching them FP.
 
Jimmy Hoffa
Jimmy Hoffa
11:13 PM
Hiring employees is the obstacle
 
user55340
Either way, you're spending quite a few $$$.
 
Robert Harvey
Robert Harvey
But, according to Graham, if you can find the people, you have a huge competitive advantage.
How does one make a GUI in Lisp, Scheme or Haskell? Are they all command-line programs?
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT It hurt the way learning to program in junior high hurt
@RobertHarvey I would completely agree with this. Think about it, who have you run into who knows FP?
 
user55340
@JimmyHoffa Actually, in Jr. High I was self taught basic (better than the teacher) and learning Pascal.
 
Jimmy Hoffa
Jimmy Hoffa
You don't run into the brilliant paula bean when trying to find lispers
 
user55340
11:15 PM
No... pascal was HS...
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT I know, and it was hard at the time I suspect. My point is learning FP is hard at the beginning yes, just like learning imperative programming is hard at the beginning. The problem is there's not a whole ton of carryover from imperative to FP - though there is luckily carryover in the other direction
 
user55340
Though, if Logo stuck around as a 'not toy' language (it wasn't, but people didn't see it that way) it is a nice FP lang... FP might be more popular.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT You just say that because you're in love with stack based languages
 
user55340
@JimmyHoffa I don't recall logo being stack based.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT thought I remembered it being similar (not completely)
 
Robert Harvey
Robert Harvey
11:18 PM
So how do you make a GUI? Are there FP GUI libraries? Do you have to interop to something more conventional?
 
user55340
to spiral :size
   if  :size > 30 [stop] ; a condition stop
   fd :size rt 15        ; many lines of action
   spiral :size *1.02    ; the tailend recursive call
end
 
Jimmy Hoffa
Jimmy Hoffa
@RobertHarvey Haskell has great bindings for open GL, some not spectacular ones for QT, but partially it's a matter of nobody has come in to do GUI apps with Haskell. You do largely web apps because desktop GUIs are a dying breed among the enterprise software which is what Haskell is aimed for
Haskell has some notoriously great web app frameworks
 
user55340
Trying to find where it was showing how to do closures in logo.
 
Jimmy Hoffa
Jimmy Hoffa
LISP has some as well, LISP probably has bindings for other UI stuff
 
user55340
> The following program draws a picture displayed in figure (captured from the screen). Note that the turtle state is not visible, because the program is completely based on turtle library functions, which are curried in their last argument. Therefore, we always call library functions without their last argument (the turtle state). The turtle state is "carried around" by library functions, and we do not care about it in the main program.
 
Jimmy Hoffa
Jimmy Hoffa
11:21 PM
@RobertHarvey It's all interop - just like in any language, what's QT written in? C/C++, but you can create applications using QT in how many languages? What about winforms? You're talking mostly about Win32 MFC internals stuff mostly in C/C++ but Delphi calls all of that stuff. UI/Graphics is always interop in high level languages because UI/Graphics stuff is done by lower-level languages
 
user55340
{ spider where rec
spider(n) = turtle([repeat(10, [five(n), turn(36)])]) and
five(n) = repeat(5, [side(n)]) and
side(n) = once([forward(n), turn(72)]) and
forward from turtle and turn from turtle and
turtle from turtle and once from turtle and
repeat from turtle }
 
user55340
 
Jimmy Hoffa
Jimmy Hoffa
@RobertHarvey look at leksah.org <-- Haskell IDE written in Haskell using Gtk, works in linux, windows, and mac
 
user55340
> The Logo Programming Language, a dialect of Lisp, was designed as a tool for learning. Its features - interactivity, modularity, extensibility, flexibility of data types - follow from this goal.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT That's awesome (and logical) that it was derived from LISP. Never knew..
 
user55340
11:24 PM
The thing is, people saw it as a 'toy only' language for drawing shapes on the screen.
 
user55340
And they "progressed" to BASIC and Pascal back in the day.
 
user55340
But if the programming had continued with logo, FP concepts could be built on top of that... and well... we'd have a world full of lisp weenies.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT they'd be all "imperative? Pffft I guess if you want to write like a driver or something... but who wants to do that..."
 
user55340
to reverse :stuff
ifelse equal? count :stuff 1
[output first :stuff]
[output sentence reverse butfirst :stuff first :stuff]
end

print reverse [apples and pears]
pears and apples
 
user55340
Note that there are no mutable variables there.
 
Jimmy Hoffa
Jimmy Hoffa
11:27 PM
@RobertHarvey also xmonad is a very large well known windowing system written in Haskell that works across OS's where it's all pure code underneath and has a very thin veneer of code working with the IO monad over top that deals with the interop. It's the ideal of how a real world FP app should be done - thin top layer of interaction with the impure outside world, and the entire rest of the codebase maintaining referential transparency, being purely stateless.
 
user55340
Why Johnny Can't Code by David Brin
 
Robert Harvey
Robert Harvey
@JimmyHoffa Thanks.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT when showing F# to one of my colleagues he commented "oh this is just like Perl, a write-only language"
 
Robert Harvey
Robert Harvey
@MichaelT PHP is the new BASIC.
And F# is nothing like Perl.
 
user55340
@RobertHarvey php isn't easy and wasn't designed for beginners.... and is very confusing quite quickly.
 
Jimmy Hoffa
Jimmy Hoffa
11:30 PM
@RobertHarvey It was just the immediate response to seeing FP - just speaks to how unfamiliar it is to most in industry. It's a people problem FP has, not a technical one.
 
Robert Harvey
Robert Harvey
@MichaelT But it seems deceptively simple on the surface. So many fledgling programmers write their first web application in it, and get hacked in 5 minutes because they didn't parameterize their SQL statements.
 
user55340
@RobertHarvey Yep. Thats why they need to work on something even safer and easier so they can learn the fundamentals.
 
Robert Harvey
Robert Harvey
Just like BASIC. You think you're making progress, and before you know it you have a plate of spaghetti.
 
user55340
Basic wasn't meant to be a production language (there are instances of things written in it though)
 
user55340
The thing is, there isn't a basic out there anymore. Javascript and php are non-trivial to write a good program.
 
Jimmy Hoffa
Jimmy Hoffa
11:32 PM
@RobertHarvey most experienced OO programmers would look at a line of code like foldl (+) $ inc <$> [1,2,3] and immediately laugh - but a little attention to the concept of putting multiple operations on a single line and reducing them by the stack would make it totally sound to them.
 
Robert Harvey
Robert Harvey
I used to work for a company that had a BASIC FOUR machine. It's native language was Business Basic (a dialect that had a couple hundred 3-letter functions in it, everything from binary indexing to file records). They ran this core application called IDOL.
The terminal had six keys on it (roman numerals I through IV), corresponding to ADD, CHANGE, DELETE, INQUIRE, PRINT, and END.
 
user55340
(give that Brin article a read... its a good one)
 
Robert Harvey
Robert Harvey
Yeah, I have it up in another tab.
The disk drive looked like a washing machine. The disk pack looked like this:
It held 20 megabytes. Your hand could fit in the blue handle on top.
It took about 90 seconds to spin up or spin down the platters.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT I think processing.js is the name of a javascript library that has it's own language it interprets to visually display things where it's supposed to be great for children and beginners learning to just hack some line-by-line sequential steps together and see them displayed
Or rather processing.js is a javascript interpreter for the processing language I guess
 
Robert Harvey
Robert Harvey
It's times like these that I wish we still had the "Not a Real Question" close reason. — Robert Harvey 20 secs ago
 
Jimmy Hoffa
Jimmy Hoffa
11:45 PM
Maybe I should write a language. It could be fun.
 
Robert Harvey
Robert Harvey
 
00:00 - 21:0021:00 - 00:00

« first day (1252 days earlier)      last day (3742 days later) »