Interesting presentation about the next programming language for 2010

[kryton9]

Here is an interesting presentation for the next programming language from the perspective of game programming.
It is written by Tim Sweeney from Epic Games.

http://www.st.cs.uni-sb.de/edu/seminare/2005/advanced-fp/docs/sweeny.pdf

[PetrSchreiber]

Hi kryton9,

thanks for interesting link.

Good stuff in the presentation, I need to pass it multiple times

There are few things I don't understand - why not use ASM optimized sections in code ( if commented, no problem I think ), and TOTAL omission of Quake I-III, Doom 3 and Source engine in the history list.
I know everybody thinks their engine is the most perfect, but presents Doom 1 as latest worth mentioning engine from Carmack is very nasty and unprofi I think .

Also although it could be possible we will get 20 core CPU in 2009, I'm not sure about using it to maxx.
Many of current applications don't use multithreading at all, so they are OK with one core. I don't presume I will play Unreal X and have 10 others applications running to make the CPU fully used.

On other side, Alan Wake already uses power of 4 core CPUs, each thread handles different things like physics and so on. But it won't make me happy to hear I must have processor with at least hyperthreading to run game on seriously decreased detail level.
Both Max Payne 1 & 2 look awesome and run fast on my sub-gigahertz PC, this is huge jump .

Well, the screens from "Gears of War" in whole presentation look awesome awesome awesome !


Thanks a lot,
Petr

[joske]

Interesting to see some numbers and quantities

[Barney]

If one takes a look at page 20 of that PDF file, one will find the answer to almost all of the problems plaguing the programming side of the game industry. It's the page about performance where he said "...everything is performance sensitive" and then finishes with "We never use assembly language".

This summarizes nicely the whole mind setup of modern programmers. They tend to find all sorts of excuses (not reliable, not productive, etc.) not to use machine code. We did not have such problems back in the 70's and 80's. Machine code programming was something no one worthy of the programmer's title would be running away from. I still believe if machine code programming is more widely used we would not be having all this bloated OS's and applications. The productivity may.. I repeat MAY be a bit smaller but that would be more than compensated for by smaller and faster applications and much smaller requirements for expensive hardware. But that's me speaking. Modern guys will probably say I'm just a dinosaur of the ages long gone...

Barney

[PetrSchreiber]

I agree with Barney,

... I wonder why they refuse asm, as it *could* make the hardware requirements much lower so better accessible to more people. In a "closed source" product I think faster is better, at least try to do some optimizations when game is in pre release stage and all features added.

I think the creator of code should be the "optimizer", not some AI compiler
I like the way PowerBASIC has covered this optimizing problem.
In classic SELECT CASE you can specify 1 word extra to determine how the code will be generated.

I was always curious if there is some statistic regarding game coding. If I presume 3 year development time, how much of this they spend on writing engine, how fast they are able to code it and what % of all dev time takes the "game design" itself. And how fast are the artists too in creating the "visible" stuff

Still, there is lot of interesting information in the slides, will see how big impact it will have on thinBASIC game engine

OT : Did anyone played Gears of War ? Will it be released on PC too ?


Bye,
Petr

[erosolmi]

Petr,

Power Basic could be used to develop such games, no doubt. It is fast, full optimized, easy to master and a lot of other positive sides.
The only problem is that it is not cross platform.

Why people use C/C++ to develop games? Portability, cross platform, standards.

Is ASM portable from one platform to the other? Do not think so.
When you have to produce a game for PC, XBox, Plystation at the same time, C/C++ is the only road I think.

Eros

[Mike Stefanik]

Small, independent game developers use a lot of different languages. For large, commerical games on the Windows platform you'll almost exclusively see C++ being used. The principal reasons are that's what most toolkits support, and doing anything in DX9 using a procedural language would be difficult; it makes extensive use of COM and is definitely geared towards C++.

Of course, the biggest reason that C++ is used is because C++ is used. It has its own kind of inertia. If you want to be a professional game developer, then you have to know C++. If professional game developers know C++, then that's the language they're going to use. It's the same kind of thing that you see in systems development.

[kryton9]

I think no ASM is not from not being able to do it, but from a productivity sense and making it easy for a team to work on it. And if you program in ASM and it makes your code not cross platform then that is a good reason of course right there. He is saying the compiler and languages need to provide these things with what is coming up. I am glad that these kind of talks happen.

I wish I kept a link about the next C++ coming out from the creator of C++, sounds like a nightmare, it is due out around 2009. Let me see if I can find it in my history, brb....
http://beust.com/weblog/archives/000374.html

[Barney]

I think no ASM is not from not being able to do it, but from a productivity sense and making it easy for a team to work on it. And if you program in ASM and it makes your code not cross platform then that is a good reason of course right there.

As I've said... those are just the usual (and completely wrong IMHO) excuses for not using ASM. And talking about productivity... great game houses are not exactly known for their speed in bringing the goods out to the market. Writing a game for two or three years and ending up with a product that immediately needs a 100MB+ patch is not productivity. At least not in my vocabulary.

Barney

[Mike Stefanik]

I wish I kept a link about the next C++ coming out from the creator of C++, sounds like a nightmare, it is due out around 2009. Let me see if I can find it in my history, brb....
http://beust.com/weblog/archives/000374.html

As a C++ programmer, I don't see why they're a "nightmare" at all. It looks like a fairly modest, incremental change to the language. The author of that blog looks like he's primarily bemoaning the fact that it isn't C#. Well, if you want to program in C#, then guess what? Program in C#.

There's no question that C# is easier to program in, and you can be more productive "out of the box", so to speak. On the other hand, there's also a lot of black boxes that you have to work with. C++ is decoupled from any framework, while C# is intricately tied to .NET. For example, you don't see any C# compilers out there generating standalone, native x86 code because a lot of what is in the language (reflection, attributes, generics, etc.) has an inherent dependency on the .NET framework.

The imminent demise of C++ is much exagerated, I think.

[Blue Steel]

What gets me is that in over 20 years no one has successfully buit an every day language.. eg: where one simply tells the compiler / interperator .. and say something like

Code: [Select]

open screen   ' opens a screen the same as the desktop settings
ask question from keyboard "What is your name"
ask question from Microphone "What is your name"

display "Hi " answer to question ' Displays responce on te screen
say ""Hi " answer to question   ' says reply over sound card
print "Hi " answer to question  ' prints responce to printer

notice "ok to continue" ' displays a notice dialog with oprion OK (the first word as the button)

or

startrek  like

Computer
I'd like a game where the player is object "xxxxx" and they can mover hotizontaly at the bottom of the screen. the enemy AI to use object "yyyyy" will move at the top end of the screen and on each pass will move closer to the player shooting at random intivils. .... etc.. etc"

Not holodeck type .. simply on the screen.. Today they have translators that can convert the spoken word .. so why not have a translater type of programming language where you tell it what you want

I used to like mucking around with "Forth" and getting it to learn stuff through multriple choice.. ok.. its not really learning.. but building uyp a data base of choices then computing which matches best by the process of ilimination..

there are many ways and aspects of programming that need to be conmsidered in my opinion.. eg: crossplatform.. there is no or little reason that programming languafges of any sort need to be platform dependant.. they should be able to use micro kernal techno9logy.. where they get compiled to a universal token language.. when then in turn gets interperated by each platform that its run on.. so all the graphics and api , sound , input , output. is handled by the microkernal fore that system. this would mean that it would free the programmers from having to know all the api calls etc.. but untill the computer market becomes more involved with crossplatform (which it won't untill one compane owns or has a great vested interest in the other processors.. thats right processors nor individual machines) then it wqould be in their interest to have software being able to run on multiple platforms , and cxhip arcitecture.

this IS a real possability with the speed of machines today.. unlike in the old days when one had to write something in machine code to get every ounce of speed that they could. these days unless you really really need speed for doing massice calculations like in  a physics lab one can use an interperated / tokonised language.. then have like a library that gets called on the end users machine to covnert or use those tokens to how the end users machine needs it..
This would split programming into 2 major fields .. one to write the micro kernals and the other to write, design the products.. . some may say that this is way in the future if at all.. I sayu WHY.. its not so different from today .. hardle a windows program gets written that doesn't call API's , dll's on windows. so whats wrong with writting to a universal library that then gets concerted by the end machine.. this way technologies can come and go.. and the software no matter what language they are written in will still work because they are converted by the end mackings kernals into what they need...

With all that in place then the language writters would have their hands / minds freed to write better easier interfaces for with the programmers could then impliment their thoughts through ie: syntax or the lack there of.. through the use of like fuzzy logic.

Hope that ppl don't think that I'm a crackpot .. lol.. its just the way that i see programming developing .. if not now r soon.. it will one day go down that track unless all compurts start using the same chipsets which is highly unlikely

[erosolmi]

Code: (thinbasic) [Select]

uses "SAPI"
uses "console"

DIM YourName AS STRING
SayAndPrint("What is your name?")
YourName = inputbox$("What is your name?", "Tell me your name, please", "")
if YourName = "" then
  SayAndPrint("I'm sorry but I cannot play without a name! Bye, bye.")
  stop
end if

SayAndPrint("Hi " & YourName & ".")

DIM YourInput     AS STRING
DIM YourCommand   AS STRING
DIM YourParams    AS STRING
dim AskUser       as string

randomize
while ucase$(trim$(YourCommand)) <> "BYE"
  AskUser = choose$(rnd(1,3), "What do you want to do!", _
                              "Tell me what I can do for you.", _
                              "Tell me.")

  SayAndPrint(AskUser)

  YourInput = inputbox$(AskUser, "Tell me what you want to do. Type bye or cancel to exit.", "")
  YourCommand = parse$(YourInput, $spc, 1)
  YourParams  = trim$(mid$(YourInput, len(YourCommand) + 1))
 
  if function_exists(YourCommand) then
    call YourCommand(YourParams)   '---This is the trick to call a script function using a dynamic string
  else
    if YourCommand <> "" then
      SayAndPrint("I'm sorry but I cannot understand: " & YourCommand & ".")
    end if
  end if
  if trim$(YourCommand) = "" then exit while
 
wend
SayAndPrint("Bye, bye " & YourName & "! See you next time.")


'---Console output plus say
function SayAndPrint(sText as string)
  console_writeline(sText)
  if sapi_moduleloaded = %TRUE then sapi_speak(sText)
end function

'---Some functions to play with
function Play(sParam as string)

  select case sParam
    case ""
      SayAndPrint ("Ok, you want to play but at what game?")
    case else
      SayAndPrint ("Ok, " & YourName & ". We can play: " & sParam & ".")
  end select

end function

function Bye(sParam as string)
  SayAndPrint ("I hope you had good time with me.")
end function

function Nothing(sParams as string)
  local Count as long
 
  SayAndPrint ("Ok, " & YourName & ". We can have some rest for the moment.")
  for Count = 1 to 4
    SayAndPrint (choose$(Count, "One.", "Two.", "Three.", "Four."))
    sleep 1000
  next
  SayAndPrint ("Ok, " & YourName & ". Time to play again.")
 
end function
  Just for joking of course.


Multiplatform? Final target of .Net technology will be nothing more/less than having platform independant languages running on a common interface. And recent Microsoft/Novell agreement is a fist (or second or third ...) sign of this move.

Eros

[Barney]

What gets me is that in over 20 years no one has successfully buit an every day language..

Err... actually there is one (sort of) and it's called Cobol. If you use it in it's original form it pretty much resembles everyday language. That was the intention behind it, anyway... 

Barney

[Blue Steel]

I meant .. that the compile / interperator sets up all the variables etc.. and the p[rogrammer only has to tell it what they want it to do .. notice there were'nt any variables in my sample codes .. that is unless one wants too stipulate variables..
I'd like to see a programming language like .. telling someone how to get from a to b .. eg: kind of like elaborated turtle graphics.. pen up, move rto middle of screen , change pen to red square 2px, pen down, move up 20px, move left 30px, change pen green round 1pc , draw filled circle radius 40px. etc..etc

game example:

Code: [Select]

choose a number 1-10, ask player to guess number if number is less than computers number tell player nember is less. if number is greater than computers number then tell player number is greater do all this until number is same then say you got it.
thats what i mean by everyday language no programmers jargon , no variables no structure.. the person simply tells in their language the computer what they want done.  the computer then goes away and generates the variables needed the loops needed and comes back with a sample when then can be modified with further instructions to make the game better or change it entirely.. eg: change your number from 1-100 and just tell the player if their guess was right or wrong.

just like that in pure english like someone would explain a game to someone else.

I know this is still only a pipe dream.. but i really thought we'd be closer now than befgore .. but we arn't .. programming is becomming more complex instead of easier.. eg: back in the C64 i wrote a game in less than 4k of memory.. the same game today would take at least double.. you used to be able to use the screen locatins as your array .. now you can't .. you have to build an array then tell iy how to display that array.. as opposed to simply placing data into the systems array for the screen directly.. if you do that today your data may be overwritten (say when you minimise the window etc... so its more complicated..

I know its wishfull thinking.. as the highly paid programmers then would think themselves out of work.. far from it.. they'd then be making the computers interact with humans and not the other way around..

[kryton9]

I hear you Blue. There are so many areas for improvment in programming and operating systems. I am surprised by the direction the world has gone from the great start we had in the early days. I hope some young genius out there is working on something that is a revolutionary language or OS that will shake up the establishment and help us move ahead and out of this sort of rut of the last almost 20 years.

I watched a video on youtube about the video toaster launch on the amiga, that was a revolution, that was almost 20 years ago, I think 1987. Who would have thought 20 years later we would still have almost the same type interface, the only revolution was the internet and video graphics, but the OS hasn't really improved that much and we just had popularity changes in languages, but no real revolutionary languages have appeared.

[Mike Stefanik]

What exactly do you see as a general area where operating systems could improve? The core functions of any OS is managing memory, processes, devices and (usually) providing some kind of file store. The things people generally talk about with regards to "future operating systems" usually has to do with things beyond those core functions, like voice recognition, natural language commands and so forth. The problem is, while talking to your computer or wandering around in a virtual 3D environment that represents your filesystem sounds cool, it's also not very practical or efficient. Lots of sizzle there, but not a whole lot of steak.

As to programming languages, they tend to evolve towards complexity because the problems that developers are required to solve have become more complex. Cameron talks about "no variables no structure.. the person simply tells in their language the computer what they want done", but my guess is he's never seen the process (and associated documentation) that businesses have to go through for things like ISO9000 certification. If you think that writing code is complicated, wait until you have to describe your internal business policies in the formalized language of a certification standard in order to implement the core business logic that drives your company. Computers can only do what we tell them to do. There is no option to "do what you think I mean, not what I say" and hopefully there never will be. Once computers start making subjective decisions about what code they think they should execute, we've got a problem on our hands.

[Blue Steel]

but my guess is he's never seen the process (and associated documentation) that businesses have to go through for things like ISO9000 certification. If you think that writing code is complicated, wait until you have to describe your internal business policies in the formalized language of a certification standard in order to implement the core business logic that drives your company.

that has nothing to do with programming.

Computers can only do what we tell them to do. There is no option to "do what you think I mean, not what I say" and hopefully there never will be. Once computers start making subjective decisions about what code they think they should execute, we've got a problem on our hands.

I gree.. what i'm getting at is that the writters of the programming languages themselves could be taking more of the leg work out.. eg: assign variable types and values according to what the user is askinbg for .. so that the endprogrammer dioesn't need to know wether its an int , byte or what ever.. also make syntaxes more human readable .. that the programming languages need to do mre.. eg: back in the old days we the everyday programmer had to keep track of heap, registers , accumulators.. these are now all handled by the compiler.. so what i'm saying is that the languages of tommorow (down the track) should be able to tajke up even more.. eg: if your inputting a string of letters then it should know that a variable that contains a string is neded and be able to assign it on compilation and be ableto associate it with a routine. the compiler / parser should be able to understand which variables need to be global and which should need to be local and alike.. NOT THE COMPUTER .. THE PROGRAMMING LANGUAGE.. OR PARSER after all its acting as a tyranslater from what ever syntax it is using into what the computer can understand.. if this wasn't the case then we'd all be writting al our code in binary still.. ALL COMPUTER CODDING LANGUAGES are like translaters.. that talk to the Cmputers processor or other DLL's / libraries ..

Basicly it boils down to this .. yes computer programming has come a long way.. but it has a long way to go to make it so that most people out there can simply sit in front of a computer and get it to do what they want..

As far as OS's .. to me they are taking away the flexability of programmers.. ou can no longer talk directly to hardware.. you have to go through their libraries.. and IF the OS or Graphic driver allows something to be done then you simply can't do it.. eg: on the older computers one could send low level commands to the sound cards/ chips  to generate sounds on the fly.. now days from what i've seen thats no longer pssible.. eg: to get a soundcard to play a waveform for a certian length of time with simple ADSR settings.. to get the sound effects you want you now have to use external devices and record thewm then load them in either through a resourse or as a stand alone file....

Please f i'm wrong tell me how to do it without spending thousands on extra plugins or software.. hell on the C64 all this and more was possible and more..

I know that these is a need for stability and conformaty but things are going too far that you can't do things that you used to be able to.. IF one wants to.. the commands and instructions that used to makje such sounds were only 2-3 commands .. now you need a sound file of frim x00bytes to do it.. and have to load it into memory..
 
I'm sorry if people think i'm talking rubbish .. everyone has their own opinion as to what a programming language should / shouldn't do.. thats why there are so many out there .. Pity there arn't as many OS's maybe then there would be more diversity and peoples ideas rather than having them dictate to US humans the way THEY want us to use them.. WE are supposedly the masters not THEM.. at the moment THEY are dictating to US how we can use them instead of the OTHER way around..

I'm not saying that Computers should make assumptions.. the programming languages themselves should be more intuative.. and tollerent / self correcting.. in some aspects..

The day will come that one only needs to tell a computer what they want to do and it will do it.. like the computers making cars .. they have been programmed to do it.. the end operater only tellsa it what part to make and the program does the rest.. thats howe i see programming ending up.. where the programming language has a lot of the workings inbuilt and the end user / programmer doesn't need to worry about them.. much like dll's are today but on a grander scale..

[rob]

Hi,

I hope that next programming language was not only for programmers but for almost every people.
Moreover programmes have already  today a great choice of language for their stuff, of course every new improvement or innovation is welcome, and personally I need it.
But if we watch around ourselves we find a lot of people that work with a computer and although they have the need to automate or simplify some tasks, they aren’t able to do it.
Often their first attempt break up with the Excel macro and VBA.
Well for me next generation language should be more possible natural, language neutral, and if not requested by the user it should hide the complexity of type, vars, structure and son on (the details),
leaving the user to focused itself only on the problem and its resolution.
It’s for that I’m following the .NET revolution with greater interest.

Regards,
Roberto

[Mike Stefanik]

I hope that next programming language was not only for programmers but for almost every people.

Why? Most people out there don't care about programming, they just care about the end tasks. They couldn't care less about how the computer does what it does, only that it does it correctly. The people who do care, well they will become interested in programming.

What you're saying would be the same as saying "I hope that in the future, microfluidics would be for everyone". As society becomes more complex, and the tools that it depends on become more complex, the need for specialization increases ... it does not decrease. And, of course, we see that happening already. It used to be that 40 years ago, you were a "programmer". Today, it's a multidisciplinary field that is only getting more and more specialized.

[kryton9]

Mike it is being more specialized exactly because of the problems with the languages and os's. You say having a 3d interface is not useful, how do you know? Who knows what use will come when someone has it as a tool, it could be used in remote surgery, exploration, just as we use video now to conference and telecast lectures and educational material, the 3d world could put us there. This is just one idea.

All of us have interests in many things, we are turned off and migrate to those things we can grasp and work with, but if we had the power we would interact and work on those other things that might be too difficult now. That is the power of what can happen when languages and os's free us.

The real power in pc's is coming and is already here, but for true voice and stuff maybe another 3 to 5 years. Our languages and os's need to be ready for the explosition that is about to come. We need the tools to make it happen and to keep pace. The fact that the languages aren't meeting the need has been evident, the last couple of years the video cards have been way ahead of the curve in what the game developers could code. The reason is the complexity of the languages as they are now.

The OS should do more than what you listed. It should be able to talk to any device, there needs to be a language where any device out there now to ones unknown of can say what they are, what they do, how you interface with it and it should present info that allows the OS to work with it dynamically. That is once the OS gets this info, it creates a driver on the fly for that device. That is just one idea.

Will stop here as could go on all day

[Blue Steel]

people want to be able to change things.. to better suit their needs .. If i had $1.00 for every time i've heard "Why can't it just do this" .. or " I wish this could to that" or "I want to change how that does this" ..

It shows that people Do want more from their software.. and what better way than to let them alter it themselves ..

Your right the average end user arn't interested in programming. Who can blame them the way things are .. its like learning a foreign language .. If it could be turned into every day phrases and words then a lot more people could then get their computers to do what they want. instead of having to run software designed by people who don't have their individual needs. It is after all what programming is all about .. to be able to get the computer to do what you want/need.

thats why Graphical desktps are around.. can you imagine the average user today having to type every command in from the command line.. Of course not some people are simply point and click. They still have the same need to have what they want done how and when they want it done. thats why thre are lts of GUI's out there that take away the need to know how to use all these terms that we use in programming today.. The programming of the future will evolve way passed the easy of use that it is today as more and more language programmers and interface programmers take heed of what the end user needs..

Which is why I always enphosize the need for maximum user configurability in what i write.. To enable the End user the choice in how things happen.. what they can see and do and when they can see or do it.. and it also must be as bug free as it can be.. Nothing is more frustrating for an end user than a program that simple don't work right.. which unfortunatly is about 80% of the software thats out there today .. I can pick up just about any software and find bugs and qwerks in it that weren't a designed feature as some lame programmers put it.. It seems more and more that if you want a program to do what YOU want YOU need to do it yourself. So Yes the End user DOES care about programming .. even if they don't do it themselves.. the programming languages of the future WILL auto correct human errors just like when onme visits over sees into a foreign land.. where they sdpeek a new language and you try to mmunicate with them.. they forgive and can repair the words that you speek into something that they can understand..  It won't happen over night .. but it WILL happen .. unless the human race ceases to exist in the mean time.

[rob]

Hi Mike,

of course users should not take care of programming, they should take care of their end task, but a great end depends on all previous tasks, since task are never ending and more complex, user needed more power to resolve them.
A computer offers power, but often this power is not applicable due a gap between the user knowledge and the system, so I hope next programming language (and OS) could reduce this gap.

Regards,
Roberto

[Blue Steel]

of course users should not take care of programming,

LOL ...

Why not ?

Thats why programming languages have to get easier to use , require less knowledge of the systems and devices. Then they'll get exactly what they want.

In the old days EVERY COMPUTER came with a programming language. and now days that school of knowledge has been taken away from the computers user.. and all they have been left with is an operating system that they point and click with.

If programming languages did more of the leg work for the user then they could then be freed from the shackles of people who say "That can't be done" or "its to hard to do that" and be charged big biccies for that honour.. at the moment the software market TELLS the USER  what they Can and Can't do and as for their intelligence there are some smart users out there .. Why shouldn't they be able (if the want to) to create their own game or utility an end programmer shouldn't have to know about the OS .. or the devices they should;d be able to tell the computer what they want and have the computer do it.

If languages didn't develop and become easier as they have been attempting to rather than pandering to those who know all the ins and outs off the OS's or devices then even todays programmers would still be using punch cards.. and binary input .. can you imagine writting games in binary because there were no other languages .. If no one made them easier to use.. and more relevant to their tasks..
a lot of programming today is all about one thing .. making money... not providing what the customer, wants.. sure they come close .. but they never give the user exactly what they want .. so that they have to come back next release to buy an upgrade that costs even more money.. has more features and even more hindrance's in usage. so that they have to come back for the latest patches .. and it goes on and on.. What a lot of the Big Software Writers are always afraid of (and that includes Microsoft) that someone one day will come and do away with all of that and finally empower the end user with the tools (be it a programming language , OS , or GUI) they need to do things for themselves.. because then they wouldn't make any money.

Look at the windows interface for a start... they give you only the bare essentials to what THEY think the end user needs.. Just look at all the interface improvements like windows blinds , and desktopx , objectdesktop, shell styles, etc.. that Microsoft could have incorporated but haven't .. even to this day with windows vista .. windowsblinds can do more to the desktop than microsoft allows the end user access to and for what .. an extra measly $59.00 (thats commercial value not development costs which in the life span of the OS would be way less per unit sold). Just Lock at all the word processors out there today or spreadsheeds.. none offer much more than they did 2 years ago..  Its the Big software companies and alike that are holding things back. and why professional paid programmers gave a sigh of relief when the programming languages stopped being provided FREE with the OS. and the end users have been brain washed into thinking that they could never write their own software and don't even go looking for all the Resources available to them on the net. Hobby languages and hobby programmers are on the increase now with the availability on the net.. prior to that  it was extremely difficult to find affordable languages to use..

and if you really believe what you said above which i quoted then my friend your in for a big shock... because as i said the end user won't need to know how or why things work.. they'll simply be able to program just as easily as using a word processor and or a GUI.. to point and click to tell the programming language or interface to a programming language to provide them with just what they want

[Mike Stefanik]

Cameron, you do realize that what you're talking about has been prognosticated for (at least) the last 20+ years, right? Heck, I can remember technologists claiming that programming, as a profession, would cease to exist... well, right about now.

The argument that everyone should be able to easily program is like arguing that everyone should be able to build their own cars, design their own microprocessors... it's just not going to happen. People, by and large, do not care that there's no simple BASIC interpreter that they can boot their system to. Let me repeat that:

Most people do not care about programming their computers.

You're arguing for something that is completely irrelevant to 99% of the people who buy and use computers every day. Why in the world go through all of the effort to make some "everyman's" programming language that is going to be ignored by almost everyone?

As I see it, languages should be designed and written for the people who actually use them, not for some hypothetical group of end-users who have no education or skills in the design and implementation of software. I have a standing joke with a doctor friend of mine, where I told him "You don't worry about messing with your computer systems, and I won't try performing surgery on your patients." That about sums it all up.

[rob]

Hi Mike,

only not build their own cars but drive them where they want. Unfortunatly still today a computer likes more a train instead than a car, for the rest you are right.

Ciao,
Roberto

[kryton9]

Remember that language Logo or was it Turtle, but something like that and it had nice simple instructions and kids in elementary school or middle school were writing programs for it and then there was a little robot or something that could actually mimic what the onscreen thingy was doing, maybe the thing on the screen was the turtle and the language was Logo. That as back in the mid to late 1980's. I thought we would have had so many more cool things like that afterwards, there were a lot of dead years there when the programming world fell in love with C and if you notice all the cool things sort of dissappeared.

Finally microsoft has a language for writing to mindstorm robots. I haven't looked at it to see if it looks as fun to code in as Logo was.

My Mac Plus as all the Mac's had that really cool feature of you dragging a floppy icon to the trash and it would eject the floppy, no eject button to push. I remember all my friends with PC's would see that and be just as amazed by how cool that was, we went from those kind of OS's and cool languages just the wrong way in the 1990's, the 90's to programming was like the 70's to our space program. Dead or a wasted decade.

Have you guys ever played with the virtual reality units that were out in the early 1990's, those were so cool, even with the simple graphics of back then. That just collapsed and never flourished. I was lucky to live in an area with an amusement park at the time and they had various arcades and virtual reality booths, everyone who came to visit me I would take them there and everyone loved it from young to old, never understood why it died, but the 90's like I said killed a lot of things in computing.

[stevenp]

Krypton, you made some good points.  Logo was interesting in that it was easy to pick-up and use right from the start.  It may have only been a toy language but it was fun.  Creative, fun languages are rare while the world of programmers marching on from embracing C\C++ to .NET.  I personally think there is room for some play languages like Logo or the like for average person of all ages.  There is a very simple language that just about anyone can pick up that's actively being developed and it's called Leopard (http://leopardprogramming.wordpress.com/).  Here's the application page with examples: http://www.weatherbug.com/leopard/leopardApplicationCatalogPage.asp.  The interesting thing is it has Logo built-in.

[Mike Stefanik]

... we went from those kind of OS's and cool languages just the wrong way in the 1990's, the 90's to programming was like the 70's to our space program. Dead or a wasted decade.

I think that's just a natural consequence of the computer and software industries growing up and moving from the "gee whiz, that's cool" phase and into the "business" phase where the focus is on utility and what they offer to the bottom line in terms of productivity.

I think what some of you guys are really trying to say is that you miss the "good old days" when there were a lot of hobbiest programmers where writing software wasn't a means to an end, it was just something fun and challenging to do. That was when "commercial" software meant that it was a floppy disk and some photocopied documentation in a ziploc baggie. Like the old saying goes, "You can never go home again".

[kryton9]

@Steven, thanks for the link, never heard about that language. I bookmarked it and sent it off to my friends who want to get into programming but can never get into it. I think this language could be the magic one. I know I will have fun tinkering with it.

@Mike, those days were cool, but that is not what I or I think Steven were saying. Based on those cool things that were happening, we should have moved onto virtual reality units, then maybe some sort of projection system, I don;t know if you guys ever saw this game at the arcade, but it looked like a holographic image. It wasn't one of course, but it sure looked like it was projected like one. You could walk around and see the characers, I recorded a video of it while playing it, I need to find my video tape and digitize it. I was sure something like that would be available for home use by now.

I think another fun era is coming though, when we have these powerful chips, and what that genius is doing with image recognition by using humans to put together incredible amounts of data, is going to bring a boom back into tech like we haven't seen before.

Robotics, portable devices and really fun full featured media rich interactivity is going to usher in this stuff.
If that new c++ is what the next gen is going to look like, then c++ will not continue being the dominate language anymore, my opinion of course.

Ok found a link for that awesome game:  http://www.klov.com/game_detail.php?letter=T&game_id=10124