OSS revenue models (was: New forum topic - Open Control)

J

Thread Starter

Jiri Baum

Greg Schiller:
> The linux PLC project and others applications involving linux must have
> some other form of revenue otherwise how can they provide techsupport for
> the complexities that arise post sale and pre production.
...
> I am looking at all this development and energy being put into making a
> version of linux that will provide an alternative to the competition in
> the PLC market. I am looking how I can contibute or benifit by helping
> develop this technology or by using it. I'm not sure how it would feed me
> down the road.

Another revenue model is the Apache way: a group of webmasters decided that if they each write a piece of a webserver, then they'll all have the software they need to do their jobs.

Similarly, a group of integrators can each write a piece of a SoftPLC and then they can all use it to provide automation solutions.

In reality, things aren't quite so organized, of course. However, if MAT is almost but not quite what an integrator needs, then he/she can extend it so that it *is* what is needed. The integrator benefits by getting the majority of the code for free, the customer benefits because the machine works, and the MAT project gets an improvement to its codebase[1].

Of course, if MAT is exactly what an integrator needs for a particular job, just use it off the shelf.


And then there's the traditional Linux way: procrastinating PhD students.

Jiri
[1] Or not - there's no obligation to contribute.
--
Jiri Baum <[email protected]> http://www.csse.monash.edu.au/~jirib
MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools
 
A

Alex Pavloff

> Another revenue model is the Apache way: a group of webmasters decided
that
> if they each write a piece of a webserver, then they'll all have the
> software they need to do their jobs.

Yes, and at the time that the Apache project started (1995?), how many high quality web servers were in existence?

None.

How many high quality and cheap PLCs are in existence right now?

A whole pile of them.

Why does this particular wheel need to be reinvented? I think what some of us are saying is that "Open Source", by itself, isn't a good enough reason for us to jump on the bandwagon. It's got to provide some clear benefit to me, right now. I look at the LinuxPLC (or MAT, or whatever) project, and I see nothing usable right now. In addition, I see nothing right now that
makes me use it as a base for anything else. You haven't achieved critical mass yet. If you want people to use open source software, then go DO
something with it, make something usable, show us, do a demo, something. Posting to control.com and the automation list don't count!

I'm not being critical for the sake of being cruel. You're just going to have to give me (and other people out there) a better reason for being open sourcing everything than vague promises about how things will be better!

Alex Pavloff
Software Engineer
Eason Technology
 
G

Greg Schiller

Here is how I would pursue trying to undermine the PLC market ;).

Look at the top 10 plc's out there and their pricing. Then decide what you can live off of financially. Come up with a pricing scheme that is
competitive.

Go get someone to design a strong Rack and IO scheme that wants a piece of the hardware sales. Also design in a remote IO structure. Make Ethernet your first method of interface with serial second. Support a flash based memory
system so I can rip the program out of one Rack and put it into the other with out having to have a laptop or more importantly the person who know how to use the laptop.

Then be different. Think of multivender motion control first. Make it a part of your system approach. Provide open connection to your tag/data table. Put it all in a box with what ever OS you think is necessary to make this all happen and maintain hard sub millisecond or millisecond real-time execution. Allow thin clients to be the HMI. Give me your programming/setup software for life, not upgrades, no keys, no hacked stuff. (roll the cost into your hardware sales). Then sit and wait for me to call you when it does not work or for advice on how to approach my application.

Then sell it to the public as "one box... start here!".

Only then will people choose this technology on a mass scale. You can see that the OS is only a brick in this structure.
 
C

Curt Wuollet

Hi Alex

Where can I get all this good cheap automation stuff that doesn't need to be reinvented. And you've been holding out on the great software too. If I could find some of that I could save a lot of effort. And I painstakingly described the automation project I did with Linux. I must be missing something. I don't know what to tell you if you can't see anything wrong that Openness, Standardization, and cooperation could fix. I sure see a lot of things here on the list that could benefit. But, maybe I just see things differently. The revenue models of the existing vendors are responsible for a great many of the complaints, problems and infelicities I see here. Try for a day to view them in that light and then tell me it's not worth trying some things differently.

Regards

cww
 
> > b) There is, in fact, a benefit to adding a feature that other vendors
> > then adopt. As the author and copyright holder, your name is in the
> > code they distribute; you and your expertise are being advertised to
...

Ralph Mackiewicz:
> I'm sure the investors will be lining up. Instead of a return on the
> investment of software development enabled by having a unique feature
> available, OSS offers self-gratification and ego-building.

The accounting term is "goodwill"... but it can be hard to put an accurate figure on it.

> There must be better reasons to spend your time developing OSS than this!

One concrete, income/expenditure reason would be if the benefit of the already existing code is greater to you than the cost of sharing your improvements.

Admittedly, that's hard to put figures on too. But sometimes it's obvious.


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

Ralph Mackiewicz

> > > b) There is, in fact, a benefit to adding a feature that other
> > > vendors then adopt. As the author and copyright holder, your name
> > > is in the code they distribute; you and your expertise are being
> > > advertised to ...
>
> Ralph Mackiewicz:
> > I'm sure the investors will be lining up. Instead of a return on the
> > investment of software development enabled by having a unique
> > feature available, OSS offers self-gratification and ego-building.
>
> The accounting term is "goodwill"... but it can be hard to put an
> accurate figure on it.

That is not "goodwill" at all. Having your name listed among (potentially) hundreds of contributors to a piece of OSS is going to
result in a neglible amount of goodwill from an accounting perspective. Goodwill may be listed as an asset but it does ZERO for cash flow or profit. Goodwill is an asset that reflects the results of previous marketing efforts (this is not a precise accounting definition). For instance, if Procter & Gamble ceased all marketing activities they would still generate significant sales. This increases the value of P&G over and above their physical assets. Over
time their goodwill would reach zero unless they started marketing again. You can't put goodwill in the bank and it can evaporate overnight. Investors are not that impressed with goodwill. They want cash flow, sales, profits, management expertise, etc. Goodwill is way down the list.

Owing a copyright can be of value. Although if you own it because of changes you made under a GPL I'm not sure that it is worth very much.

Getting your name in a contributor list will only boost your ego, not your bottom line.

> > There must be better reasons to spend your time developing OSS than
> > this!
>
> One concrete, income/expenditure reason would be if the benefit of the
> already existing code is greater to you than the cost of sharing your
> improvements.

Now you're on the right track.

> Admittedly, that's hard to put figures on too. But sometimes it's
> obvious.

Nothing is obvious to those that are unconvinced. EVERYTHING has to be explained.

Regards,
Ralph Mackiewicz
SISCO, Inc.
 
> > > > b) There is, in fact, a benefit to adding a feature that other
> > > > vendors then adopt. As the author and copyright holder, your name
> > > > is in the code they distribute; you and your expertise are being
> > > > advertised to

Jiri Baum:
> > The accounting term is "goodwill"... but it can be hard to put an
> > accurate figure on it.

Ralph Mackiewicz:
> That is not "goodwill" at all. Having your name listed among
> (potentially) hundreds of contributors to a piece of OSS is going to
> result in a neglible amount of goodwill from an accounting perspective.

Depends on how large the feature is - if it's a feature the client wants, and you're the *main* developer of that feature, it's perhaps a bit better.

You can also put it on *your* shingle - advertise that you were the original author of that piece of code.

> Goodwill may be listed as an asset but it does ZERO for cash flow or
> profit.
...
> You can't put goodwill in the bank and it can evaporate overnight.
...

There's a lot of other assets that can't be put in the bank or that evaporate if you don't watch them - take livestock, for instance. Even real
estate gets burdened by rights-of-way if you don't put fences up.

I believe you're mistaken as far as the impact of goodwill is concerned - it can do a lot for a business. Indeed, I suspect you'd do very little work if most of your clients didn't already know of you...

> > One concrete, income/expenditure reason would be if the benefit of the
> > already existing code is greater to you than the cost of sharing your
> > improvements.

> Now you're on the right track.

I guess it's the major way in which people can benefit - the other ways also apply, but they'll be much rarer.

> > Admittedly, that's hard to put figures on too. But sometimes it's
> > obvious.

> Nothing is obvious to those that are unconvinced. EVERYTHING has to be
> explained.

Sorry, I meant `obvious' in the sense of the difference being large (ie adding only a few lines to a large piece of code); you still have to explain it, but you don't have to be that accurate estimating $/LOC.


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

Ralph Mackiewicz

> Depends on how large the feature is - if it's a feature the client
> wants, and you're the *main* developer of that feature, it's perhaps a
> bit better.
>
> You can also put it on *your* shingle - advertise that you were the
> original author of that piece of code.

The value of goodwill is related to the amount of business that would continue to be generated by your name previously being in the author list if your name was suddenly removed from the author's list. I think that this value is very small. You won't be able to build a realistic business plan around it.

Putting it on your shingle does not generate goodwill unless you expose that shingle to a lot of people. That builds goodwill. Then, if those people no longer saw your shingle, goodwill is the value associated with the tendency of these people to give you business anyway.

> I believe you're mistaken as far as the impact of goodwill is
> concerned - it can do a lot for a business. Indeed, I suspect you'd do
> very little work if most of your clients didn't already know of you...

I did not mean to suggest that goodwill is meaningless to business. My assertion is that having your name listed in a piece of OSS along with many other contributors will not generate enough goodwill to be significant factor in a business justification. The value proposition to an investor has to be based on something other than having your name listed in the software.

Regards,
Ralph Mackiewicz
SISCO, Inc.
 
> > > That is not "goodwill" at all. Having your name listed among
> > > (potentially) hundreds of contributors to a piece of OSS is going to
> > > result in a neglible amount of goodwill from an accounting
> > > perspective.

> > Depends on how large the feature is - if it's a feature the client
> > wants, and you're the *main* developer of that feature, it's perhaps a
> > bit better.

> > You can also put it on *your* shingle - advertise that you were the
> > original author of that piece of code.

Ralph Mackiewicz:
> the author's list. I think that this value is very small. You won't be
> able to build a realistic business plan around it.

Well, it was one item in a list - it may well be very small. Like I wrote, it might help if it's a large feature that a client might be particularly interested in (or if you're the primary author for the whole program).

> Putting it on your shingle does not generate goodwill unless you expose
> that shingle to a lot of people.

Actually, the usual formulation here is `line on a CV'...

> I did not mean to suggest that goodwill is meaningless to business.

OK, no worries.

> My assertion is that having your name listed in a piece of OSS along with
> many other contributors will not generate enough goodwill to be
> significant factor in a business justification.

Quite likely - especially in the situation where it's one of a hundred equals, then it's practically useless.

On the other hand, I believe zope was originally released this way, as a
way of marketing a particular business. Last I checked, the company's still
around, so it seems it worked ok.


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