migration to software?

As you could read I was struggling with the usability, not happy that we have no possibility to give VARs, RAMs, UROMs and WEBSETs real names to be stored in a table on the WC and other deficiencies when testing. So I wrote a simple PLC-interpreter in python 2.7 for Windows, to have some debugging/tracing capability. But this lead me to a ground shaking thought. What if the WC as we know disappears and CAI only supplies software to be installed on micro controller systems? This would immediately wipe all the limitations allowing an integrated development environment, rich documentation and debugging capabilities and the like. A virtual WC. Especially I would stick to the instant deployment capability using more boards to be connected. The environment may be tested in advance in a simulation tool.
I do not have ready the real email and WEBSET functionality in python so far, but will check that out when I get a suitable Linux board. There the first steps of a migration to software could happen.... Seeking allies to get further into that I'm open for your thoughts. In the best case CAI will jump on the bandwagon, but also other business models may be viable.
If we get that far the software may also see some XML interpreting WEBGET (xWEBGET), Network diagnosis (WEBCHECK), remote software reload (swUPDATE) and the like.... The hard work implementing software for new sensors could also be made through user defined libraries...
Having some free ressources CAI could develop piggy back solutions to be used with standard microcontrollers (industrial raspberry PI,OLinuXino and PCDuino naming a few). My guess is that an CAI PLC-code interpreting software will be able to run on much smaller boards in future compared to the ones I have listed.

Efried,
 
We appreciate your great efforts.  We do not plan to develop on other hardware platform at foreseeable future.  A lot of companies want to enter the market and sell a large number of boards.  Then they found this market is not really fit their business plan.   Go check out those boards you mentioned, you probably will find so many new compatible boards out there, and none of them really made money.  It will not stop any new company jump in, there are a lot of people on this world think they have more money or they are smarter than others.
 
As to us, we believe both WC8 and WC32 will fit most our customers needs.  We will continue enhance our firmware, continue to provide support and services to our customers.  To see our customers successful, that is our fun. 

CAI_Support said:

Efried,
 
We appreciate your great efforts.  We do not plan to develop on other hardware platform at foreseeable future.  A lot of companies want to enter the market and sell a large number of boards.  Then they found this market is not really fit their business plan.   Go check out those boards you mentioned, you probably will find so many new compatible boards out there, and none of them really made money.  It will not stop any new company jump in, there are a lot of people on this world think they have more money or they are smarter than others.
 
As to us, we believe both WC8 and WC32 will fit most our customers needs.  We will continue enhance our firmware, continue to provide support and services to our customers.  To see our customers successful, that is our fun. 

Click to expand...

 
It's not to lure you into that business porting the WC-user interface to exchangeable hardware, but just an idea how it could happen. And in the ideal case this has nothing to do with the luck of new start-ups but the ability to take standard boards of several vendors, plug in some hardware interface modules having also signal conditioning if necessary, fire up an standard µC OS and load the PLC-software having the same functionality we now have with the WC.  CAI not taking that idea as chance it might well result in a threat if ST, TI, NXP or others will offer an interpreting PLC environment on their boards. My understanding is such that it is better to start thinking now about programming paradigms for distributed systems of the IoT and achieve some unique selling proposition with an integrated development environment for that purpose.
Let's come back in two years time to that hypothesis and see what happened. I think a software configurable system might also allow new business models.

From my point of view, the existing WC platform does what it does quite well, and fills a niche that others simply don't.
 
I frequently have "instant need" for a system. Some kind of software-defined PLC running on arbitary hardware needing some sort of development/programming/documentation environment as you propose doesn't suit me at all.
 
Real world example: I was recently doing some work on an industrial facility. All computer-controlled gizmos, only "something" had gone belly up. Neither the guys on site, the national support agent, or I could see what was wrong. Problem was that the process had stopped dead in its tracks and substantial product loss was going to happen if we didn't get things back and running within the hour.
 
Fortunately, I had a WC8 in the car. I was able to borrow one of their computers and use NO MORE THAN A BROWSER to write a small program to make this thing produce enough output sequencing to allow the site to operate - perhaps not at optimum efficiency, but at least enough to save the product and continue production while the problem was solved.
 
Due to the nature of the original failure, the WC8 board ended up running the production for almost a month! The ability to simply jump in with a browser and change a couple of timing constants meant this "irritatingly limited device" with its 100% self-contained everything, was perfect for the job.

What if the WC as we know disappears and CAI only supplies software to be installed on micro controller systems? This would immediately wipe all the limitations allowing an integrated development environment, rich documentation and debugging capabilities and the like. A virtual WC. Especially I would stick to the instant deployment capability using more boards to be connected. The environment may be tested in advance in a simulation tool.

PLC code is too hardware dependant and low-level to be ported from system to system if I read your concept correctly.
 
The portability of a medium level language would be awesome but look at all the languages out there now since manufacturers can never agree on anything and give up their "day in the spotlight".  In the few cases out there great things have come about for users and industry but many fear losing control and losing profits. Apple and Blackberry may be good examples.
 
OTOH Radio Shack tried locking down their computers, with a closed architecture, years back, and and went from #1 to the bottom in one year.
 
It would be nice to see and could even improve the speed of development and testing for the open minded but not likely to happen at this end of the playing field.

In the industrial side, a few PLC manufactures developed "ladder" diagram/logic on windows, then when save for different PLC, they translate the diagram into that particular PLC logic.
 
We found that ladder diagram is becoming a challenge of its own over the time.  A lot of companies end up offer extensive classes and trainings for how to use those ladder diagram.
 
That is why we selected this BASIC like PLC language, less those user randomly define the variables, added some hardware related commands and data type.

CAI_Support said:

In the industrial side, a few PLC manufactures developed "ladder" diagram/logic on windows, then when save for different PLC, they translate the diagram into that particular PLC logic.
 
We found that ladder diagram is becoming a challenge of its own over the time.  A lot of companies end up offer extensive classes and trainings for how to use those ladder diagram.
 
That is why we selected this BASIC like PLC language, less those user randomly define the variables, added some hardware related commands and data type.

Click to expand...

 
My first objective was to test my code via the PLC simulator. I think this would be helpful for all using the existing and future CAI products and needing some unit testing capability. I'm attaching a handy chart having the execution frequency for the program lines, generated by my quick and dirty python emulator (not real time capable) when varying temperatures randomly over one day.
The success of  a software solution on the other and depends on the size of the market for a board standard or standard boards. We will see, I'm not totally with you, because looking at the block diagrams I guess a lot of boards simply connect the processor pins to headers which would give the same condition for different boards having the same processor specification. But of course this is only a guess, supported by the attitude of the board vendors referring to CPU specification often (http://dl.linux-sunxi.org/A10/A10%20User%20Manual%20-%20v1.20%20%282012-04-09%2c%20DECRYPTED%29.pdf).
As for the improved unit testing capability I think my approach emulating the WC PLC logic in software is helpful.

LarrylLix said:

What if the WC as we know disappears and CAI only supplies software to be installed on micro controller systems? This would immediately wipe all the limitations allowing an integrated development environment, rich documentation and debugging capabilities and the like. A virtual WC. Especially I would stick to the instant deployment capability using more boards to be connected. The environment may be tested in advance in a simulation tool.

PLC code is too hardware dependant and low-level to be ported from system to system if I read your concept correctly.
 
The portability of a medium level language would be awesome but look at all the languages out there now since manufacturers can never agree on anything and give up their "day in the spotlight".  In the few cases out there great things have come about for users and industry but many fear losing control and losing profits. Apple and Blackberry may be good examples.
 
OTOH Radio Shack tried locking down their computers, with a closed architecture, years back, and and went from #1 to the bottom in one year.
 
It would be nice to see and could even improve the speed of development and testing for the open minded but not likely to happen at this end of the playing field.

Click to expand...

 
Hi,
now having some hands on experience with a (Chinese) PCDUINO, I may add:
1. It is quite common to offer graphical programming software like ArduBlocks, Scratch or others simplifying the engineers life, so he/she may concentrate on the process
2. file based systems like the PCDUINO where single data is stored in files may not viable for real time processes and I doubt that the SSD storage will survive long for high frequency polling. Also for industrial usage such a Linux system shall be deprived of the (may be energy hungry) graphical user interface and be capable of starting the control software after reboot in superuser mode because of real time requirements. This is not impossible so I will stick to that board for a while, because the integrated WiFi really worked flawlessly (but not the browser). So far the ADC did not fully convince me, having a strange behavior, I still have to try out PWM and GPIO
Concluding , the software only solution might not be viable for the cheaper Chinese boards now. But I will check out the new Raspberry PI too.
The WC is certainly now the better alternative for quick entry but has more limitations. CAI will certainly see a strong competitor with the PCDUINO Scratch software if the board is hardened a little for industrial use. 

PCnuino might take Raspberry Pi market places, but they are both not same as WebControl.  You don't go to Lowes and point to all the trees or flowers and claim they are all the same.  They are different. Until you really know what they are, it is better not mix them all together.
 
Some new vendors came to Cocoontech site to post marketing information that has nothing really related to the product in that category.

CAI_Support said:

PCnuino might take Raspberry Pi market places, but they are both not same as WebControl.  You don't go to Lowes and point to all the trees or flowers and claim they are all the same.  They are different. Until you really know what they are, it is better not mix them all together.
 
Some new vendors came to Cocoontech site to post marketing information that has nothing really related to the product in that category.

Click to expand...

 
Hello CAI support team. My posting was looking in the future not fuelling marketing wars. I wanted  discussing about automation paradigms and not details of current boards. One theory is that automation goes PC in terms of standards and the big business will be software. Otherwise automation may not become mainstream for homes because of lacking interoperability. Secondly I wanted to convey that tehre is need for unit testing. And my software emulator of the PLC is the first step improving robustness of quick and dirty PLC-ing.

It seems that CAI staff really cares about users. We now have a commenting possibility and forced rebooting amongst other new functions. Still lacking for me is PWM with WC8, encryption, integrated wireless and roll out support for multiple board arrangements.
But anyhow programming a WC master controller might be to complicated in terms of testing. One possibility solving that problem could be some collaboration between boards capable of running high level languages on the one and WC8 boards on the other side. So far RaspberryPi and PCDuino were a bit disappointing in terms of start-up time requiring long install procedures before the board may be used.  My latest try will focus on this board running python: https://micropython.org/
May be in the end it may also run my WC PLC code, since I started writing an PLC interpreter. Let's see, I'm willing to hand over this idea to specialists.

WC8 processor can not do encryption, since its CPU is not designed to handle that.  If you want encryption, then WC32 will do for you. WC32 also have PWM function.
 
LarsK developed a configuration copying software, that allow copy from one board the configuration and paste into many boards.  We do not plan to provide any software to compete with that. If you have any suggestion for LasK software, please contact him.
 
All those xDuino, yspberry boards will come and go, lower cost manufacture will come out with different board to replace the older one. WC boards with its own firmware and board will be there with our support.  We welcome 3rd party software development, but we will not take same path as those board only manufactures. We care about our customers by providing solutions and improvement for our customers.  We also encourage 3rd party software developers taking same path.

CAI_Support said:

WC8 processor can not do encryption, since its CPU is not designed to handle that.  If you want encryption, then WC32 will do for you. WC32 also have PWM function.
 
LarsK developed a configuration copying software, that allow copy from one board the configuration and paste into many boards.  We do not plan to provide any software to compete with that. If you have any suggestion for LasK software, please contact him.
 
All those xDuino, yspberry boards will come and go, lower cost manufacture will come out with different board to replace the older one. WC boards with its own firmware and board will be there with our support.  We welcome 3rd party software development, but we will not take same path as those board only manufactures. We care about our customers by providing solutions and improvement for our customers.  We also encourage 3rd party software developers taking same path.

Click to expand...

Yes I appreciate your work, but it may well be that there will be boards offering a high level language from the start and there is already one working with python. IMHO there is a short period of time where the WC8 will have some USP on the market due to its low cost and short time to set up working solutions, but having decreasing board costs the WC32 will replace the WC8 having some fierce competition too from boards offering graphic development.

Efried,  We have no intention to follow your instruction to discontinue WC8 or lower WC32 price,  you may go press other manufacture with your idea.  There are similar products out there made by Hitachi, Honeywell, and Allen-Bradley. They may be interested in your idea.

Hi, there. Sorry to bring up this old thread. Ive got five wc8's doing things around my house and a few Rpi's. I really like the webcontrol and the webGUI thats built into them, But im wondering if anyone has "ported" the webGUI to html or something suitable for my Rpi- as I'd like the same interface on my raspberry pi?

Thanks
mc