Linux in the field

P

Thread Starter

Patrick Allen

The newest Mandrake release has really polished Linux as a desktop and has made installation far easier and less painfull than loading a machine with Windows. But for some reason a lot of the open source control projects seem to have lost their steam in the last year or so. Links are broken, webpages go unmaintained. What happened?

Admittedly, I'm more of an electrician than a software developer. But I have been a Linux advocate and home user for a number of years now. But my employers have had no interest in using it commercially. Potential for troubleshooting has always been an argument against Linux. That, and the lack of at least one or two polished looking products out there; say...something along the lines of LabVIEW.

Is there something out there and I'm just missing it?
 
A

Alex Pavloff

Why Patrick, I'm glad you asked that question!

(Perfect segue into a little product-hawking of my own followed by some general observations about why Linux doesn't appear to be everywhere).

My company just released (this weeks) a Linux-based HMI. It doesn't run on a desktop -- rather, its using an embedded Linux to run our proprietary code that does all the things (and more) that one would expect to see from an HMI. The programming software is still Windows based. Here's the thing -- in all the companies that we talk too about applications, in all the contacts we've got, I can count the people that actually use Linux on the plant floor on one hand (and that's stretching). Linux (so far) is an IT thing. The small/medium business that my company deals with have little IT experience. They're good at making things move and things turn on.

They all use Windows on their desktops, and all the PLCs, motion controllers, and HMIs are programmed with Windows-based tools. Obviously, we'd be stupid to try and sell Linux HMI that runs on Mandrake/Red Hat/what have you to these customers. Now, the IT-savvy companies like OPC. Why? Because it works like its supposed to on Windows servers and desktops. The customers can mix and match OPC servers and applications from a variety of vendors.

In contrast, on the Linux side, there is no "critical mass" of automation software out there. Sure, you've got code here and code there, with some sample projects, but very little guidance on where to go next. Compare this to Windows, where googling for, say, "OPC tutorial" finds numerous examples on how you can quickly slap together a simple data viewer with VB.

As to why all various projects seemed to lose their steam, thats easy. The entire industry is sucking wind right now. Linux software devlopment won't help the bottom line _right now_, will it? Nope, so companies don't do it. Hence, developers that would like to work on Linux are forced to use Windows and other eeeeevvviilll things like that so that they can eat.

The difference between my company (with the just-released Linux product) is that its more of a classical embedded software system (albeit a very flexible one) sold as a piece of hardware rather than a fuzzy "customizable software-as-service" like many of the Linux advocates here and elsewhere like to push. From my viewpoint, Linux works great for embedded software projects, irregardless of any arguments about freedom or speech or beer. Hrm, its friday, I think I'll have a beer.

Alex Pavloff -- [email protected] Eason Technology --- www.eason.com --- Linux-based industrial HMI ---
-------- www.eason.com/5k --------
 
M

Michael Griffin

On April 25, 2003 12:26, Patrick Allen wrote: <clip>
> The newest Mandrake release has really polished Linux as a desktop and has
> made installation far easier and less painfull than loading a machine with
> Windows. But for some reason a lot of the open source control projects
> seem to have lost their steam in the last year or so. Links are broken,
> webpages go unmaintained. What happened?
<clip>
> Is there something out there and I'm just missing it?
<clip>

I'm not really the one to answer this, but since you haven't had any other replies yet, I'll make an attempt. "What happened" is that not a lot has happened yet. I think the biggest problem is that people don't know what needs they should be fulfilling.

--

************************
Michael Griffin
London, Ont. Canada
************************
 
C

Curt Wuollet

What's missing in this market are customers willing to demand Linux solutions and become part of the community. The intentional barriers set up make it very painful to get out of whatever same old thing you are into. For example, the company I work for has consolidated their enterprise IT on Linux on huge IBM mainframes, but I'm still stuck using RSLogix on Windows because there isn't RSLogix for Linux.

Regards

cww
 
Michael:

Actually, at least as far as my stuff is concerned, my projects have languished because of unemployment and lots of it too (essentially since last June).

My projects, CELL, ABEL, etc..., have all fallen by the wayside as trying to find employment has moved into the "job #1" position. This coupled with the strong urge to change careers (after 8+ years in this field) and also keeping my family happy and fed has consumed what little free time I have had for my Linux based projects. End result - the projects sit stagnant.

--
Ron Gage - Saginaw, Michigan
I am looking for work - resume at http://www.rongage.org/resume.doc
Electrical Engineering, Linux Programming, Networking
 
C

César García

I think, that many people don't say anything about that.

The backend = Linux
The frontend = Windows

Regards,

César García
 
On April 25, 2003, Patrick Allen wrote:
> But for some reason a lot of the open source control projects seem to
> have lost their steam in the last year or so. Links are broken,
> webpages go unmaintained. What happened?

In a couple of weeks, there's a Libre Software in Automation Workshop in Leuven, Belgium. It is expected that it will improve cooperation between the various projects, so that instead of disjoint pieces of code there will be a more-or-less coherent system. It should also raise the profile of work which now might be taking place behind somewhat closed doors. (We are aware of a number of projects running in parallel and/or orthogonal with ours.)

As far as MAT is concerned (I can't speak for anyone else), we're still going! We're aware that there was some slow-down, but we expect that the Leuven workshop will give this work new impetus.

We will keep you informed.

> Potential for troubleshooting has always been an argument against
> Linux.
Isn't it the other way around? :)

> That, and the lack of at least one or two polished looking products
> out there; say...something along the lines of LabVIEW.
Agreed.

Jiri
--
Jiri Baum <[email protected]> http://www.csse.monash.edu.au/~jirib
MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools
 
C

Curt Wuollet

Hi Alex

Congrats on the new product!

On April 26, 2003, Alex Pavloff wrote:
> My company just released (this weeks) a Linux-based HMI. It doesn't run on a desktop -- rather, its using an embedded Linux to run our proprietary code that does all the things (and more) that one would expect to see from an HMI. The programming software is still Windows based. Here's the thing -- in all the companies that we talk too about applications, in all the contacts we've got, I can count the people that actually use Linux on the plant floor on one hand (and that's stretching). Linux (so far) is an IT thing. <

It's that chicken and egg thing again. I know quite a few people who would happily run Linux on the factory floor if the products were available. And the products aren't available because so few people run Linux on the factory floor. Linux is becoming very big in IT because a few of the people with the apps finally broke down and did some ports. And between Oracle and IBM and what you can get from the OSS community, you can do IT with Linux fairly comfortably.

The critical path was convincing application vendors, not customers. That is especially true in automation since the hardware is not generic and the OSS community can't help as much. No matter how much I want to I can't really make products for an AB shop or a GE shop. But, it's crazy to cite a lack of customers when there's nothing for them to buy. It's like saying I'm not going to bring out this new car line because nobody's bought any of them yet. Except for people like myself who can roll their own, how could there possibly be Linux customers with no Linux products?

The small/medium business that my company deals with have little IT experience. They're good at making things move and things turn on.

> They all use Windows on their desktops, and all the PLCs, motion controllers, and HMIs are programmed with Windows-based tools. Obviously, we'd be stupid to try and sell Linux HMI that runs on Mandrake/Red Hat/what have you to these customers. Now, the IT-savvy companies like OPC. Why? Because it works like its supposed to on Windows servers and desktops. The customers can mix and match OPC servers and applications from a variety of vendors.
>
> In contrast, on the Linux side, there is no "critical mass" of automation software out there. Sure, you've got code here and code there, with some sample projects, but very little guidance on where to go next. Compare this to Windows, where googling for, say, "OPC tutorial" finds numerous examples on how you can quickly slap together a simple data viewer with VB.
>
> As to why all various projects seemed to lose their steam, thats easy. The entire industry is sucking wind right now. Linux software devlopment won't help the bottom line _right now_, will it? Nope, so companies don't do it. Hence, developers that would like to work on Linux are forced to use Windows and other eeeeevvviilll things like that so that they can eat.
>
> The difference between my company (with the just-released Linux product) is that its more of a classical embedded software system (albeit a very flexible one) sold as a piece of hardware rather than a fuzzy "customizable software-as-service" like many of the Linux advocates here and elsewhere like to push. From my viewpoint, Linux works great for embedded software projects, irregardless of any arguments about freedom or speech or beer. Hrm, its friday, I think I'll have a beer. <

Again congrats on the Linux product, even if it's a closed one.

Regards

cww
 
Maybe the time has come to make two lists that should reinforce the role of Linux in the control industry.

List 1 - The strengths of Linux as control system.

List 2 - The weaknesses of Linux as a control system.

If Linux proponents and Linux detractors wish to participate, by posting their thoughts and perceptions, I'll keep a running record and post the results when the two lists appear to be complete.

Bob Pawley
 
P

Peter Whalley

Hi All,

To which I would add:

embedded controller with TCP/IP comms = Linux

This is happening big time but in most cases the operating system is all but invisible.

I'm involved in a project with some 300 embedded servers (Axis cameras) running Linux and talking to 10 or so servers running Win200 and similar number of HMI also running Win2000.

Regards
Peter Whalley
Magenta Communications Pty Ltd
Melbourne, VIC, Australia
e-mail: peter*no-spam*@magentacomm.com.au
delete *no-spam* before sending
 
M

Michael Griffin

On April 26, 2003 07:12, Alex Pavloff wrote:
<clip>
> My company just released (this weeks) a Linux-based HMI. It doesn't run on
> a desktop -- rather, its using an embedded Linux to run our proprietary
> code that does all the things (and more) that one would expect to see from
> an HMI. The programming software is still Windows based.
<clip>
> They all use Windows on their desktops, and all the PLCs, motion
> controllers, and HMIs are programmed with Windows-based tools. Obviously,
> we'd be stupid to try and sell Linux HMI that runs on Mandrake/Red Hat/what
> have you to these customers.
<clip>

The MMI panel is "automation". The programming software is "office use", just like the CAD or project management software that gets used as part of the same project as well. I think this is a very important distinction to make. "Automation software" runs on systems which control or interact closely with machines. "Office software" runs on computers which get used for general
purpose computing tasks.

There are other people debating whether Linux is ready to take over the office. That sort of discussion belongs on another mailing list however. If you wanted to hedge your bets, you might think about making sure your software is desgned so that a port to a Linux desktop won't be a catastrophic experience if you feel the need for it several years from now. Most of what would be required for that would probably be prudent software design in any case.

************************
Michael Griffin
London, Ont. Canada
************************
 
M

Michael Griffin

On April 27, 2003 06:08, Ron Gage wrote:
<clip>
> My projects, CELL, ABEL, etc...,
<clip>

These are drivers for communicating with AB PLCs. This is exactly the type of basic software "infrastructure" which needs to be in place before Linux is used widely "in the field".

I believe that industrial communications drivers is one of the key technologies Linux requires for use in automation. What is also needed is some sort of common API for these drivers, equivalent to what OPC does for Windows. At the moment though, the common API is probably less important than the drivers being available.


************************
Michael Griffin
London, Ont. Canada
************************
 
C
Hi Ron

I'm sure this has affected many OSS folks. A great many do something else during the day and so, don't benefit from the fact that Linux is
growing and expanding rapidly as even PHBs are required to look at the value/cost/benefit equation. The problem is, that Linux in the roles it is dominating, doesn't create many jobs. Replacing dozens of Windows servers with a few Linux boxen may actually create unemployment as one person can manage quite a few servers that stay running. So far, the "lack of Linux talent" argument that is ballyhooed whenever people want
to keep Linux out, has been shown to be a farce when a Linux job posting produces a very vigorous response from qualified people, both unemployed
and those who are employed but doing something else. I was told by one employer that I was about applicant #240 for a Linux SA job. As this thing continues to snowball, I expect to see great opportunity, if we don't starve to death first.

Linux in automation will probably lag far behind, as the monopoly is well entrenched and the vendors enforce it and openness prevention
vigorously. Birds of a feather and all that.

The unfortunate part is that folks like you and I are interested in automation rather than ISPs or corporate IT and are victims of a vast and successful "dumbing down" movement. This is focused on ensuring that most computer related jobs can be filled with lower paid, "windows
operators" rather than skilled labor or specialists. This is good for business but very bad for progress and advancing the state of the art. The number of people who actually create things has plummeted as the number of folks who simply run other people's creations has skyrocketed. This is not meant as criticism, but as a reflection of what is really going on. Is the revolution over, so now they are shooting
revolutionaries? Is it now just a business thing rather than a science or technology thing? Is automation a place for machanics rather than
engineers? Should I be looking for a haven with the craftsmen of yesteryear? Or perhaps a tar pit with a friendly compliment of other
dinosaurs? Are we there yet?

I just went through the unemployment experience and for the first time, I was left with the impression that being a knowledgable and experienced "computer guy" with a hardware background was a liability rather than an asset. And to add insult to injury, people would actually ask me if in addition to designing, building, and coding test systems, if I could
possibly "run Windows" and would even look skeptical when I assured them that I could operate a mouse and reboot at need. Isn't there
something very wrong when this is held as the height of achievement? This was held in much higher esteem than even whether I could interpret
ladder diagrams, which was also grudgingly accepted with some skepticism.

What I heard was, "Experts?, We don't need no steenking experts!" Is this where the industry is where you are?

I now hold a maintenance job, which strangely, pays as well as any of the openings I have seen. What happened since the last time I changed jobs? And is this percieved leap in technology holding the adoption of Linux back?

Regards

cww
 
M

Michael Griffin

On April 27, 2003 08:46, Jiri Baum wrote:
<clip>
> On April 26, 2003, Michael Griffin wrote:
> > I think the biggest problem is that people don't know what needs they
> > should be fulfilling.
>
> Would you (or anybody else) be willing to contribute to our wish-list
> document?
<clip>

Perhaps you could tell us what you consider "using Linux" to be? Is this supposed to be about "using Linux", or is it supposed to be about "Open Source" which happens to use Linux? Siemens, Sixnet, and now apparently Eason all "use Linux" in some fashion. I could add several more companies to the list if I spent the time to look up their names. These companies are not in the "open source" business though. They just picked an operating system which they thought made good technical and economic sense. I sometimes get the impression that what some people have in mind when they say "Linux" is a
philosophy rather than an operating system.

To answer your question more closely, one of the biggest things that you lack now which can be readily solved is a good way to make information available. Someone who is new at this, doesn't know where to start. Telling someone to use Google, or to search around on Sourceforge, or to post a question on a mailing list isn't good enough.

You need a central web site which points to where to find things. This should include useful commercial items as well (e.g. people who provide Linux drivers with their special boards). The "automation" market includes a lot of overlap with the lab and scientific fields, so these sources should be kept in mind as well when they are useful (drivers for data acqusition boards,
signal processing libraries, etc.). "Automation" on Linux also requires conventional software tools as well (e.g. Python) as alternatives to things like "VB" which are used on Windows.

I would suggest you also need an "official" organisation to promote what people are doing. "Jiri Baum's Web Page" doesn't sound very impressive to people who don't know you. Something like "www.LinuxAutomation.org" however
sounds like something to take seriously when the trade magazines get your press releases. A lot of what counts in this business *is* just good
promotion. Arguing with people on this mailing list doesn't constitute promotion.

Finally, I would suggest that you give careful consideration to what your target market should be for the near future. I see three main areas which would have near term promise:

1) Machine OEMs who use PC based controls in standard machine designs. Some of these write their own softlogic or MMI software.

2) Automated production test equipment.

3) Interfaces to IT systems.

************************
Michael Griffin
London, Ont. Canada
************************
 
A

Andrey Romanenko

The site www.realtimelinuxfoundation.org provides
a lot of information about various industrial and
research applications where (real-time) Linux has been used. For the reasons the other people
have highlighted in this thread, these solutions
are mostly small scale, in-house, experimental, and research. But they are successful. The market is yet to be upset towards Linux, though.

Andrey
 
Hi all,
i don't know the actual situation but in 2000 i saw a shift to Linux as a platform for HMI applications.

At that time i had some projects - i'm a freelancer - for a mid sized engineering company in Germany. They mainly work for steel companies.

This company sells an own HMI System called PRODAVIS which is OS and PLC independent, available for Win - Linux - SCO.

The most projects were done with Linux as OS, followed by SCO Unix.

But maybe this was historical thing. They are suppliers for the steel industry since more than 40 years and supporting the Unix environment since the late 80's. So the shift was mainly from SCO to Linux and not from Win, what means their customers are already used to have Unix systems.

Btw. not only european steel plants were using it, i know at least one project that was done in the US (can't remember the company name) at that time.

Sorry if my English is not the best, i'm trying to improve it everyday.

Lothar Kontowski
 
Hey Curt,

There is no RSLogix for Linux because nobody is asking for it! There is no RSLogix for Apple computers for same reasons. 1 or 2 people ask now and then but not enough for any benefit for the porting of software.

As other people have mentioned here, Linux is great for embedded products because there are no runtime fees. Microsoft, however, is very competitive but will struggle to crack this market.

The only hope for Linux products on the desktop is when there is a .NET port available for Linux. This allows software development to target both platforms with little cost.
 
D

Donald W. Carr

Devices that communicate via ethernet can work very well under Linux, no special driver needed, just re-compile the communication routines for
your version of Unix or Linux. For example:

http://www.edasce.com/multifunctionedas.asp

I am not sure what operating system they use for the device itself, might be Linux?
 
Out of curiosity, has anyone done this work for Omron? Any links for finding it?

Thanks!


--Joe Jansen
 
Top