I really *really* like the look of "FlowCode"! :-D

[iamwhoiam]

FlowCode for PIC, AVR etc. Dave Jones, your "mail bag" video got me looking at this; THANKS!

Why did noone tell me before? LOL!



Your first program:



Just when I said I was sick of programming and trying to learn... haha. It's all HIS fault --->

[amspire]

The problem I have with using flowcharts to program is that they are fine if you are toggling a single bit, but I find a whole program written in flowcharts to be totally incomprehensible.

Text-based code ends up being much easier to follow, as you can add as much extra documentation to each routine as you like.

Flow chart programming could be useful as an intermediate stage to bridge people from the Novice stage to coding, but after that, I think it would be a waste of time.

I guess if could be good as well for people who do not want to learn programming and just want to do simple control tasks with their micro. Flowcode could be a good option for some people.

Richard.

[IanB]

Although it has to be said that graphical programming tools are very common in the automation world for programming PLCs and the like. Ladder logic, state machines and sequence programs are naturally expressed graphically. I think it depends a little on the domain.

[Psi]

Eventually someone probably will invent a graphical object based programming language that's as powerful as a text language and faster to create even the most complex programs.
It will probably use some combination of touchscreen and/or gestures to manipulate a 3D view/design of the program and all it's functions.

Going to be an interesting day to be a programmer when it happens.

My guess is it's not as far away as people think.
Phones are getting more and more powerful and it's only a matter of time before people will need to write/maintain programs using a phone touchscreen.

[EEVblog]

I can't believe the number of Youtube comments where people cannot understand the value (and cost/effort) of software and good documentation...

Dave.

[amspire]

It looks to me that a really huge amount of work has gone into developing this suite of tools and I do not find the cost surprising at all.

To have the one graphical programming tool that seemlessly works with 4 totally different microprocessors families, and a programmable logic chip (even if it is old) is an impressive effort.

One problem I can see for the company is that many of the PCB's are so simple, I can see colleges making their own copies of the peripheral boards.

[Hypernova]

Graphical programming is a complete cluster fuck if you need to implement signal processing that need loops and if/else/switch function. If your function can be nicely pipe-lined or can be cleanly described by FSM's it's great.

[EEVblog]

Would you expand a little, Dave?

As amspire said.
They have developed this flowcode software presumably from scratch, and it seems like they have put a LOT of effort into producing learning materials for school.
Developing all that learning material can take an order of magnitude more work than producing the hardware!
If that material is sufficient for school lessons, then the teachers don't have to waste their time writing and that's worth a LOT.
Yet many commenters just can't seem to grasp why a PIC development board is worth $600. They just don't get it, it's got nothing to do with the hardware.
Are they supposed to just give away the software and learning materials? Of course not, that's not their business model, and that's what took them the most amount of time to develop, so they are going to charge for that.

Dave.

[Bored@Work]

Such graphical programming systems only bring you so far  ----><----. Which isn't very far. You have to give up on a lot of things you need for non-trivial programs. Such systems are not new, but their time might finally have come, taking the current trend of dumbing down everything to the point even the most lazy doofus can claim to be a "programmer". Cf. "web programmers" and Arduino "programmers.

[Kremmen]

I can't believe that the majority of those dabbling in application development using (only) Arduino would consider themselves "programmers". Certainly no professional would. The general level of skill demonstrated is pretty nonexistent.
Such users could benefit from visual flowchart "programming" but for "real" programs, naah. I dabbled wit C++ code generation already in the 90s and then it became obvious there is such a vast sea of problems and design patterns and whatnot, that automatic code generation quickly becomes totally lost. AI research has progressed, sure, but there is still no alternative in sight for human understanding.
The expressive power of flowcharts is so low that SW pros have dismissed them a long time ago. Now if someone comes up with a working generator to process UML notation into real source code tha would be something. There have been attempts but so far those have been less than convincing.

P.S. Just for your amusement and to show that a lowly Arduino can be moderately useful, attached is an application for Arduino Uno, written in true C++, with dynamic class creation and deletion. It is the control program for a camera dolly for time lapse photography with controlled x/y movements. Unfortunately i wrote it for domestic hobbyists so all comments are in Finnish, but then SW pros don't need comments because the source is self commenting, right.
This is easily achievable using an Arduino, but just try that with some kind of flowchart presentation

[McMonster]

I can't believe the number of Youtube comments where people cannot understand the value (and cost/effort) of software and good documentation...

Dave.

Yes, I handle software and hardware orders for computer and PLC laboratories at work and quoted prices threw me off my chair a few times, but as a programmer myself I really understand that writing a complicated piece of software that does it job well is a daunting task sometimes.

But I've seen some hobby PLC projects that used graphical programming with ladder logic and FBD that compiled down to AVR assembly.

[mikeselectricstuff]

Graphical programming like this is usually a dead end - fine for simple noddy applications but just doesn't have the power, flexibility or expressiveness for most real-world applications. Stuff like interrupts & handling realtime events is not easy to express in a flowchart type environment. By the time you make the environment complex enough to cope with the majority of applications, the  learning time is no less than a real language, so you're wasting your time learning your way into a non-transferrable dead-end when you could be learning a proper language and programming techniques.
In terms of developing real apps,  you're also at the mercy of any bugs in their software - imagine needing to have something working tomorrow and their software suddenly decides to fall over. At least with a conventional compiler you can generally work round problems like this, and look at the code it generates with a fairly close correspondence between the source and object.
 
ISTR a while ago attending a seminar by ST where  they were pushing some similar graphic based system (?ST-realiser or similar?) , and it was clear to everyone there it was a dead end, When asked how many real product had been developed with it, they could only come up with one. 

It's the same with FPGA software - schematic entry looks attractive on the surface, but apart from reimplementing existing designs it quickly becomes more complex and messy than using HDL as soon as you get any useful complexity.

[Mechatrommer]

i teached student programming lego nxt robot in graphical mode because its the rule in the contest. well, i was in deep suffering! for simpler stuff, this thing can do, but when the code gets complex, even sketching the flowchart with hand is almost impossible.

[tinhead]

... a PIC development board is worth $600. They just don't get it, it's got nothing to do with the hardware.

because they see primary a dev board - they don't know that FlowCode itself works with any DIY board and pickit 2/3.

[mikeselectricstuff]

Here's an analogy - symbols and diagrams are OK for simple instructions - warning signs, furniture assembly instructions etc., but for expressing anything more complex, language works a lot better.

[FreeThinker]

I've Tried Flowcode code today and it's very nice, as stated above it has it's limitations but is very simple to use, has a built in simulator and interfaces to your PICKIT 2/3 (I only tried the PIC version ) so is great for quick and dirty circuit testing. Home version costs £50 Pro £200 not sure of the differences may just be the no of users, but all in all it seems quite nice. You could design your code run it then burn it to hardware (via your pickit) and never see a line of code.

[casinada]

National Instruments LabView is not a toy. It took them a while but now are a standard for graphical programing of instrumentation and control systems.

[Psi]

Here's an analogy - symbols and diagrams are OK for simple instructions - warning signs, furniture assembly instructions etc., but for expressing anything more complex, language works a lot better.

At the moment i agree. but i think sooner or later someone will invent an entirely new way to express logic graphically and at that point things will change.

[IanB]

At the moment i agree. but i think sooner or later someone will invent an entirely new way to express logic graphically and at that point things will change.

In the Star Trek universe they control and program computers by tapping a few different abstract symbols on a keypad.

Now either people will become more intelligent in the future (unlikely), or this holy grail will be within the scope of today's mind. Given the lack of progress to date, I sincerely doubt that the future envisioned by Star Trek will come to pass.

[Psi]

At the moment i agree. but i think sooner or later someone will invent an entirely new way to express logic graphically and at that point things will change.

In the Star Trek universe they control and program computers by tapping a few different abstract symbols on a keypad.

Now either people will become more intelligent in the future (unlikely), or this holy grail will be within the scope of today's mind. Given the lack of progress to date, I sincerely doubt that the future envisioned by Star Trek will come to pass.

I don't think it's a question of people "becoming" more intelligent. It's a matter of teaching currently unknown concepts at an early age so we can develop abilities that we don't know we have.

The whole skill set of computer programming was unknown until computers were invented and the need for it arose.

I think most programmers would agree that if computer programming was taught as a major subject (like English or Maths) from a primary school level and up that... by the time they left high school the majority of people would be able to do semi-complex computer programming without any trouble at all.  (not everyone would enjoy it, but that's a separate issue)

It's not that programming is hard, it's more that people 'think' it's hard.
We limit ourselves a lot by thinking "I'm not intelligent enough for X"

[Mechatrommer]

At the moment i agree. but i think sooner or later someone will invent an entirely new way to express logic graphically and at that point things will change.

logic as in fpga or combinatorial logic is ok. but in software engineering, it can become 'illogical'

In the Star Trek universe they control and program computers by tapping a few different abstract symbols on a keypad.

probably predefined standard construct, and probably higher level standardized functions, such as search, sort, compress function etc. but then there's IDE meant for kidz, there's for 'professional' users, and then there's for professional (core) programmer. think of Windows/Mac/Linux, we are unknowingly becoming 'professional users' using predefined construct (buttons, menu, file system etc) we are at the mercy of the 'OS/kernel core programmer' simplifying things for us. even though we program C++ or basic in it, but still... we are the OS users, not OS programmers.

National Instruments LabView is not a toy. It took them a while but now are a standard for graphical programing of instrumentation and control systems.

they are meant/excel to control instrumentation yes. but when it comes to controlling with some level of intelligence, then... that blocky blocky things are just toyz. been there i know. so i guess it depends on what you want to do. just to control ON/OFF an device, go for Lab/NI-View 'blocky thing', kiddy stuff? FlowCode (no offense, only from my experience, no hard proof). core stuff? go verbose, go textual and ready to invent 'new creative illogical semantics'. (but not for ascii drawing P)

[Bored@Work]

I don't think it's a question of people "becoming" more intelligent. It's a matter of teaching currently unknown concepts at an early age so we can develop abilities that we don't know we have.

The whole skill set of computer programming was unknown until computers were invented and the need for it arose.

I think most programmers would agree that if computer programming was taught as a major subject (like English or Maths) from a primary school level and up that... by the time they left high school the majority of people would be able to do semi-complex computer programming without any trouble at all.

I have to disagree here. Take Math as an example. Taught as major subject, but still most people suck at simple math after they finished school. Devising a computer program, as opposite to just the act of typing (or drawing icons), also requires talent.

One of the reasons there is so many bad software out there is lack of talent and people believing everyone can become a programmer by just studying. And many, not the least hardware specialists, even think the studying part is optional when becoming a programmer. Graphical programming environments cater exactly for these type of people. Similar to the Arduino Sketch junk this stuff targets the lazy ones with a lack of talent.

You have this in all areas of life. There are always those promising you to becomes something even if you lack the talent "you can become an XYZ, too.". Whole industries rips off talentless wannabe book authors, wannabe painters, wannabe DIYlers (aka "makers"), wannabe bands and musicians. Nothing new here.

In the Star Trek universe they control and program computers by tapping a few different abstract symbols on a keypad.

Star Trek controls explode whenever there is an unusual condition. I don't want to have that junk in front of me. Give it to the Arduino crowd please. Let them burn their fingers.

National Instruments LabView is not a toy. It took them a while but now are a standard for graphical programing of instrumentation and control systems.

It is a monster that works in a very limited area. And it rapidly breaks down when you step outside of the threaded pathes. E.g. in terms of speed.

[Psi]

Take Math as an example. Taught as major subject, but still most people suck at simple math after they finished school. Devising a computer program, as opposite to just the act of typing (or drawing icons), also requires talent.

True, but math isn't something that really goes anywhere. You can see you need some math but beyond add/sub/mul/div students find it hard to care much about it.  It's needed for problems but it really doesn't do anything (except maybe PHD level math on cutting edge science).

Programming allows people to produce things which make their lives easier.
So many times i see people in data-entry like jobs doing repetitive keyboard takes when they could write a simple program to do it with only a small amount of programming knowledge.  My point is it's immediately obvious why programming is useful.

I also heard a month or so back, not sure if it's true, but apparently computer geeks at school are starting to become popular due to being able to write phone apps.
It may, slowly, be changing to "cool" to be able to program or at least not as un-cool.

And motivational and enjoyment really are the keys to learning new things.

[Bored@Work]

True, but math isn't something that really goes anywhere.

Sure? People are ripped off daily because they fail at simple math. Take advertising like "now 10% bigger". People assume they get 10% more in weight, while it could mean the manufacturer managed to increase the volume by blowing in more air. Or people who end up in deep debt, because they can't figure out what they signed.


I recently met a nurse. She gave the usual spiel of "oh, I suck at math" when I mentioned that I now had 60% of my five required doctor appointments done.  Blank stare ... Oh great, I like it when medicine is mixed the wrong way, because a nurse can't do simple math.

But we can take another subject, languages. Except if you are an American, who don't believe in a second or third language, languages are a useful thing. But still many people suck at it after years of foreign language education at school.

It's needed for problems but it really doesn't do anything

Oh, it keeps your wallet thick and helps you don't get killed by a blond nurse.

Programming allows people to produce things which make their lives easier.

And math helps to keep people alive.

So many times i see people in data-entry like jobs doing repetitive keyboard takes when they could write a simple program to do it with only a small amount of programming knowledge.

Most of them can not, because they lack the talent for programming. And there is a surprising thing. People, especially GUI users, want it that way. It gives them a sense of achievement when they click themselves silly.

[steve_w]

From another perspective if you are doing something safety related the output of the software needs to be deterministic.  With a lot of code generating software the code is generally unacceptable for safety applications because it is unverifyable.

I have seen labview solutions rejected several times due to the inability to understand what is going on inside the software.  So a code generator is good in a learning environment but probably won't be accepted in a serious application.  So i'd say bored at work has it pretty well spot on, do the maths and understand what you are doing, it will make life easy in the end.

regards


SW