Iain Buchanan <iaindb <at> netspace.net.au> writes:
> Me too. > > > Yet Linux lacks a robust open source SCADA plan. > > Because there isn't the money for some company to come in and push Linux > as being the Way To Go, sell lots of licenses, and make profits. Linux > doesn't work that way - too many distributions, too much open source - > it scares people (people who want to invest). Well that's why a SCADA plan on Linux is what's needed. It not that it's linux, its that it is a robust SCADA on Linux. Then it can run on any (linux)distro. Gentoo just happens to be one of the better choices. NO you do not have to sell the SCADA software. The money, as IBM puts it is in the installation, consulting, customization, and support of this proposed Linux based SCADA software. Young/new developers would gain keen insight into how the software works, and could become consultants to large organizations or use it for a small business. The bottom line is the number of DSPs, micro controllers, CPU, FPGA and many other sorts of processors are getting 'connected'. Sooner or later the collective Linux community is going to have to address massive machine control. Technical persons will be managing tens of thousands of processors, via computer, and consumers and other users will not be able to surf thru thousands of vendor supplied software applications. For exammple, just go and try to find your copy of a manual to something electrical you purchased more than 3 years ago. Chances are (statistically) the manuals are gone. With operating systems just look at the amount of hardware that is controllable via the kernel, compared to 5 years ago. We are building many, many more electrical devices and it sure would be much simpler to develop software as a management systems approach for thousands of different devices, as opposed to getting the latest vendor supplied application happy with yet another device. > But, if something you want in Linux doesn't exist yet, write it! That's the whole point. Why should lots of individuals and small companies spin their own SCADA software package for linux? Why not get a few top developers to develop a SCADA package for linux, then many, many small companies can make money migrating companies with Microsoft base SCADA to Linux Based SCADA, then there is money to develop the SCADA package further. Once the LINUX SCADA system is organized, many other folks could contribute. > > I have offered money to any young, talented person wanting to make > > a name for themselves by championing the cause to develop an open source > > SCADA system for (Gentoo) linux. No takers. none. Why? This could become > > an excellent opportunity to teach software development, and migrate > > the industrial world to Gentoo. > > We got sick of the pricey commercial SCADA packages that just didn't > work well enough, so we wrote our own. We started with the > visualisation software, wrote our own data logging software, moved onto > a communications library (which we have running on a number of PLC's, > real time PC's, windows PC's, industrial controllers, etc) and we're > improving it all the time. I'm sure many other companies have done the > same thing, but you just don't hear about it. We've just nearly > completed a rewrite of the visualisation software, that allows people > (for now, just us, in the future, customers) to easily create their own > 'screens' to view and control devices. Yes, but with hundreds of different PLCs and other devices, wouldn't it make more sense to develop drivers once and a unified SCADA package for linux that is widely used and supported? Imagine if every company wrote their own MTA or web browser.... what a waste of talent. > In fact, not long ago we tested our version of our visualisation and > communication software against a commercial SCADA package using Modbus > TCP over a satellite link. Let me just say that Modbus was atrociously > unresponsive in comparison. We've built our software to handle bad > connections, especially poor 9600bps modems to remote areas (but I > digress)... > We've chosen Gentoo to go on our HMI PC's, as well as our industrial > controllers (Look back in the archives for some discussions on getting > Gentoo to fit in under 64Mb). Which/what controller hardware runs Gentoo? > > My answer to Savior Linux is "put the check into the mail, and I'll be > > right there". > I think it works the other way around. You might have to do something > first, and then see if the check turns up. No, I think you missed the point. The original author is offering vapor-financing. I'm willing to finance a serious SCADA development effort, particularly in the early stages with an open source public license. > > video controls and display of video needs to be added to the SCADA software. > Not there yet. Don't know that we're heading in that direction > either... Well let's consider an example, forget intrusion (physical) detection as it's obvious. Consider a very large pump station pulling water out of a lake or river. It has backwash screens, which need to be 'pulsed' periodically with large volumes of pressurized water and/or air. It's a good idea to have a video camera on the pump intake screens to prevent boaters from getting hurt, if they stray into a restricted area. Another example: a young technician is troubleshooting an equipment problem, and you are trying to help remotely. It's sure nice if you can see the equipment and problem.... > So, do I want your check? Probably not. Will we ever release our > software to the public domain? Maybe. Will people use it? Maybe not. > The fact is, we started as a small (2 person) company, started writing > our own software and hardware design, we were about 8 people when I > joined, continued with bigger and better software and hardware with less > components made by other people and more by us, and now we're over 30 > people, and still moving. I'm sure we're not the only success story in > the world. That's great. And everyone that took that approach are now employees of GE, Invensys or Microsoft, in the windoz world. NOt to mention that if a company only had to build hardware and sell it, knowing that there was a well define development approach to getting their hardware supported by a very large SCADA package, it would tremendously lower the cost for small businesses. Just look at how OpenOffice is used by linux folks to deal with a common document format problem. Could you imagine if everyone wanted to develop their own solution? Futhermore, as the amount of hardware in the world increases, so does the amount of vendor supplied software. Wouldn't it be nice if there was a large, robust Linux software package that was available, so when a small company wanted to develop a new product, it could easily be supported from a well know software package? After a while the larger companies would join in, as customers ask if their hardware will be support in XXX-SCADA. A paradigm shift, my friend, a real paradigm shift is what I'm talking about. Thanks for the comments, James -- gentoo-user@gentoo.org mailing list
