A conflict between Open Source and Business

Open source and business (making money) will always be a more complicated equation to resolve. Especially the side where the (paid for) effort you do to produce an open source product can be reused by others for free. You basically give it away. But you also get it: all the work needed to produce a full software stack (linux, jack, apache, etc) is the result of open source work. It works as intended. You enrich an ecosystem, and others will benefit from your expertise, as much as you benefit from others’ expertise.

Not always as in “they take your software and sell their product with it”. They could (should?) participate in improving mod-host/mod-ui… or not (unless the license forces them to release whatever improvement they make, but they might not make such improvements in the first place, I am not sure). But also, plugin developers could take advantage of the MOD to have a platform to use to distribute their work (and they do! see some of the comments above), so using Free Software makes things complicated, but also is a bless. I wish I could get “community upgrades” (like plugins) to improve my Zoom R24 or my Roland Fantom X8. It’s never going to happen. But MOD Dwarf (btw I just received one I ordered, knowing that the Tier 3 may still take a while), I am sure I will be able to get new plugins for years, even make my own if I have a specific need (or ask somebody to)!

I think the business model for MOD is still the hardware. The software enables the hardware and gives it value, and the hardware gives value to the software like no other (MODEP - which I have tried - and others are not in the same league). Another thing that gives special value to the hardware are the online services, and more specifically their integration with the hardware, the access to the plugin store, to the pedalboard sharing (from my perspective, the hardware should be what gives the key to these services, so it does not make sense to open them to other platforms - these are services, not software). In that respect, what needs to be communicated is that the hardware makes the software shine and that the software makes the hardware shine. You can get clones elsewhere, but it’s never going to be the same.

What I would find difficult is how to keep the balance between on one hand the one-time fee that is the hardware, and on the other hand the continuous improvements (software upgrades, plugin development) and service access (plugin store, pedalboard sharing, etc) that keep costing the company. One is of course the fact that the existence of these services gives more value to the hardware, hopefully convincing people to buy a unit. But at some point, the existing users could be willing to give more, in order to get more.

I don’t have an answer, I think putting services behind a subscription model (like Roland and others are doing) is a bad idea. I would not want to believe that I “rent” my unit, that it becomes worthless (or at least less valuable) without a monthly fee. At the same time, if I think I can get value by some subscription, that wouldn’t be lost if I stop subscribing, I might be willing to participate. E.g. if it’s a bit like a “patreon” where once a certain amount is collected, it could trigger some paid work (e.g. some specific plugin, some additional platform feature that would not otherwise be able to see the day). Everybody would benefit, even those not participating, but by participating, one would help this development to come earlier. This kind of things (not saying ALL development should be financed this way, don’t get the wrong idea! :wink: ).

In the end, around open source software, there are generally two ways of making money: on either side of the software: the hardware and the extra services. the software itself should enrich these two, and in turn be enriched by these two. And then, you need good communication so that it is clear to everybody :slight_smile:

9 Likes

the same for me. You pay for the Hardware and the service (plugins, pedalboard cloud). The same thing google does.Using open source as tools and base but keep the Hardware (pixel phones, tensor) and some software (gmail) closed.

isn’t this the the reason why the whole ecosystem exists? You build your system around existing and free (open source) software (80% of your plugins are not develop by the mod team) and save on the initial costs that even a portion of the guitarix plugins would have cost in development. Not to mention the linux core.
Don’t take this the wrong way many businesse operate that way and it makes sense on a basic level. But to state you did all the work without getting paid in some form is a bit shortsighted.

4 Likes

Basically, since open source is everywhere under the hood, maybe it is counterproductive to advertise that “our platform is 99% open source” and make a leverage point out of it. It just makes noise for non insiders.

4 Likes

For me it was also open source that brought me here, and for users and the MOD ecosystem it has many advantages as many had pointed out… but I also appreciate all the closed stuff that MOD offers.

For investors, I suppose it would be better to consider them as potential customers and no competitors… I suppose going to user-based cloud instead of device-base goes in this direction. For example If all mod-ui with rpi or pc users could also buy closed or even MOD plugins should be very interesting for investors. And now the price difference with a mod dwarf is not so big to do the step forward eventually.

For big customers I don’t see the conflict. For example I doubt Thomann could sell them.

3 Likes

I’ve been involved in open source for longer than should be allowed. I spoke at the first Linux Kongress. If the main selling point of your product is the software, then you only win if you are selling services such as maintenance contracts. However, if the winning part of your product is the hardware design, then the open source should enhance and not detract from your product. Blokas doesn’t have a polished hardware package and Zynthian is not the ideal user interface for a guitarist. Offer a quality product and if possible, find something in what you do that can be patented.

16 Likes

Exactly what C&G did with the Organelle : pure data vanilla in a functionnal cute little metal blue box. With a community that makes more (useful and/or experimental) patches than you will ever need. Love that piece of Gear, and even though you can do the same thing with a Rpi for a fifth of the Organelle costs (and I tried), what makes it awesome is the immediate physical interface the hardware provides with the patches you or others made.
Genius!
I always saw my DuoX and Dwarf the same way I see the Organelle.
I have a Pisound with Modep, and I never use it to run Modep, but rather instances of Orac in Pure Data. I bought the DuoX the day after I gave a try to Modep. In a way, Blokas made me buy MOD.

15 Likes

I’m not sure there’s a clear question in this topic.

If ii is “How do we make money with open source software ?”, others with more knowledge than me about open source business models have already given answers : it seems like it’s either with hardware or service.

If it is “Should we keep our software open source ?”, I can share my experience and my opinion.

  1. How did I hear about MOD ? On some Linux musicians forum (Ardour, linuxmusicians or something).

  2. Why did I invest in this project ? Because the software is open source. There’s some political value about it and I can spend more on something I believe in than on just a product from a company that just wants to make money. Of course, the product had to look good. And it did : the hardware seemed of high quality and the possibilities of the device were attractive. Maybe I’ll have the chance to check, maybe I won’t.

So I may belong to a niche (musicians who are also Linux and open source software lovers) but it’ might be interesting to keep this niche on your side and market more towards this direction. What other guitar pedal of this quality runs on open source software ? I don’t know any.

What are the best selling points from all musicians (open source supporters and others) perspective beside the high quality hardware ? In my opinion, it’s something you can’t find on other pedals : hundreds of plugins and an OS that is constantly improved. When I invested, I thought The Dwarf would receive updates for years and would have a constantly growing list of plugins. Amazing ! And what makes this possible ? The open source software. Most companies just give you a couple of firmware updates.

I understand your disappointment when you see others take advantage of your investment in the software without any compensation. It’s a bit like what Tier 3 backers can feel when they see people playing with their Dwarf. But, as others have pointed out, remember that one of the selling points you used for marketing is the huge list of plugins, most of which being open source. So you also benefit from the open source community. Maybe it’s worth considering being part of it.

10 Likes

I am also in this niche

I think my selling points are pretty much the same as yours. Additionally, I thought that since there’s significant open source software involved, I should still be able to use and even develop “content” for the device independently of the health of the company.

11 Likes

There isn’t. It is an exposition of a business issue related to the fact that our software is OS.

I am happy to hear that multiple users made their decision of moving into our platform influenced by MODEP. I would love to understand if that happens at scale level or if this is just a handful of isolated cases. Also it would be great to discover how many people have not joined MOD and decided to go the low-cost path.

On a personal POV, I share the opinion of the majority here. I think that there are multiple reasons to being OS, from philosophical to practical ones, and, as I wrote in the initial post, I am neither attacking OS nor willing to go proprietary.

But on the business level, there are issues related to being OS that, in more than one occasion, worked against our growth as a business.

On the long run, I share the opinion of many here, that the hardware and services are the means to make the revenue and the software should be distributed.

In our long-term vision, the company would migrate from a hardware based revenue model (which is what we have today) to online services and marketplace in the future. In a further scenario, when we are big enough, the price of the hardware should even be “subsidised” by the other revenues so to diminish even more the barrier to entry of new customers.

But that is the thing. All these ideas only work at scale level.

In order to have the required revenue size coming mostly from software and services, the population needs to reach a critical mass.

We are still stuck on the growth to get there, so still no practical solution for us here.

Sorry guys. Woke up sadder than normally today :frowning:

9 Likes

For my side, MODEP served as a way to validate the MOD Dwarf: enough to understand how MOD UI works, but also enough to understand MODEP itself is not enough. It may be a stop-gap, but in the end, somebody that tries MODEP and wants to get involved has (had?) not really other places than moddevices.com to try to find answers, tricks, explanations, and then be exposed to the real thing. So, it is good as an enabler. Somehow, it helps that MODEP is not really making any big effort to keep at the forefront. It’s more “here! we enabled this for you, do what you want/can”.

As I mentioned above, I believe the extra services (plugin store, pedalboard sharing) are really adding value to the platform, and the Dwarf is the “key” to access them. Paid services (e.g. paid plugins) are also fine - to a certain extent: when I buy a device, I wouldn’t want to feel my device is sub-par unless I invest much more. Also, I wouldn’t like to feel that I have “rented” my device (which is often the case when the hardware is subsidised), that if I stop paying, then my device is somehow “disabled” (for some functionality). In summary, I wouldn’t want to feel a hostage of a subscription model. Not all subscription models are bad though, subscriptions that are giving access to additional services are good. A subscription to pay to have access to e.g. a certain “sound” and that you lose if you don’t pay are bad (e.g as would be renting plugins or features of the device itself). By this, I just want to say that MOD Devices has hit a sweet point with the Dwarf (many plugins available from the start, yet options for some advanced paid-for plugins, and no renting model), and it just needs to find a way to keep it viable on the long term.

Actually, I think that, in these times where everything costs more, a well-polished hardware which caters to all your needs is actually more valuable than ever. And its value should not appear less than it is. Given the cost of individual pedals, if I invest in a Dwarf expecting to replace most pedals (if not all, in certain setups), then I would not expect the hardware to cost peanuts. It’s just that the experience of using a Dwarf should be top-notch from the moment I take it out of the box :slight_smile: (which comes back to the fact that “cheap” alternatives won’t be on par).

8 Likes

My path was built a MODEP, built a zynthian (still use it as a zynthian), invested in the Dwarf.
Can’t you turn this open source “weakness” around? From I have 2000 hardware/software clients limited by some cheaper solutions to I have 2000 hardware clients and 5000 software clients all “selling” how good the platform is, the entry to be a software client is (was) a 35$ raspberry and we have the nicer hardware by far.
You can setup a survey for this, I really don’t know the size.
Best,

5 Likes

THIS!!!

5 Likes

Let’s be clear : that was totally reasonable in a context where NOBODY could foresee the collapse of the supply chain for electronics from China. If one had told an investor in early 2019 that they wouldn’t want to develop a product based on electronic components made in Asia, they would have been shout at and treated like fools.

6 Likes

There are already plenty of proprietary solutions on the market. Adding one more to the list wouldn’t be a win in my opinion. They are backed by much more wealthy investors and armed with tons of marketing experts. Think of the Spark by positive Grid. So if it was just to have that, I don’t think a lot of us would have been attracted by MOD’s product. What makes it unique is the Open Source model.

I discovered MOD while thinking about creating my own DIY pedal running open source simulators like Guitarix or Rakarrack. I didn’t want to spend loads of money buying pedals that can only do one thing, plus an Amp that can only do one thing, or even a Line 6 simulator which is locked to do only a couple of things. At the time I believe they were starting to sell small pedals running software based effects, but you could only run one effect at a time on one pedal and it was all closed and black-box like. I didn’t want to be trapped in that. I always hated Apple products because of that, when you are totally dependant of arbitrary decisions from the manufacturer.

I was also worried about what would happen to my product if the company goes bust. With Hardware pedals, you don’t care, because you can still fix them if they fail. But with software based devices, you need updates, bug fixes, and ideally to be able to add your own effects, or effects that come from a different provider.

So I was dreaming about what it would take for me to hack together such a dream device, and by looking up some similar projects (so that I did not have to reinvent the wheel), I came across the MOD first kickstarter. And I immediately wanted to get a Duo. It was doing exactly what I wanted and probably way better than anything I could achieve myself. And I could even get involved if I wanted to.

I wouldn’t have jumped onboard for anything else.

The next best alternative to me is not MODEP but simply running the MOD interface on a strong enough laptop with a good soundcard and a bunch of MIDI controllers. Even in live settings. Because the laptop offers power and maximum possibilities, including editing the board live. Many acts have laptops on stage as well. But still in that case, the MOD interface is really bringing a lot to the table. Because of the model based on jack and the endless possibility of routing it offers. I actually think the MOD interface is a way better UI than all the Jack traditionnal UIs, like Qjackctl or Catia, and it makes a lot of sense running it on a PC.

What would be awesome is to be able to run ANY commercial plugin in the MOD system. I guess it wouldn’t be possible on an ARM based mod devices for performance reasons, but maybe embedding a CPU emulator able to run VSTs created on a different platform inside a MOD plugin would be a game changer (at least on the PC version). Is that just a dream @falkTX ?

6 Likes

not for any foreseeable future, I would say minimum 5 years.

emulating x64 on ARM is something not even microsoft is able to do properly yet.
apple only did because they “cheated” by adjusting their own hardware and chips to make that actually feasible.

and ARM cpus general performance is still not that great. this at least is where things can quickly improve. in my opinion the best path for the industry as a whole would be to leave x86/64 architecture behind, would love to see a generic all-purpose ARM chip that matches M1 performance…

8 Likes

I mean “how do they get paid for their work” in general. Not especially by MOD. What is the business model of Guitarix. Is that 100% unpaid voluntary work ?

2 Likes

Indeed : it might be good to have a page on the site talking about MODEP etc… explaining what they are (encouraging enthusiats DIYers to check them out) and how you are VERY different from them. That would clarify your relationship with them in the eyes of potential investors.

Actually, it could be a good idea to present these projets (including the MOD PC version, and the online MODs) like unique possibility of testing before you buy! Like great tasters or cheaper versions that people can try, and then move on to the MOD once they want the real deal (the rugged undestructible powerful PRO version).

3 Likes

To be realistic I think even on the bare software side original Mod is more advanced. This is because features on mod-host/mod-ui are mostly ad hoc for Mod product. On the plugin side, I think both my yocto porting of mod-plugin-builder and also the debian packages of MODEP are not up to date with mod-plugin-builder. In my case I simply don’t have the time to stay behind developement on this repo in particular. I remember I’ve checked out some time ago the plugins in MODEP debian packages and they were at least a year old comparing with Mod devel. For example at the time MODEP didn’t offer file browser feature. This is a way to say that derivative projects are not all that danger at all, in an open source project things like that exist, are part of the game and moreover the bigger the community, the better. More chance a musician will enter a studio and find a Mod device or derivative and ask: hey what’s that?

3 Likes

On a different performance aspect : wouldn’t it be worth exploring the opportunities offered by GPUs as well. How do GPU calculations integrate with the JACK engine ?

1 Like

I can ask since this pretty much overlap with neural network inference in audio plugins. Like I’ve already said my plugin is based on RTNeural lib from @chowdsp. In his paper, which I’ve studied, there is a mention on why not using GPUs or NAs (Neural Accellerators) when doing inference for real time audio. Well the problem, in this exact moment, is that access time to external hardware is too much for real time audio applications. The benefits in terms of computations are killed by the time it takes for the plugin to transfer the data to external hardware block plus initializing/configuring it. Let’s keep a look maybe in future it will be feasible.

3 Likes