Is MOD building products or a platform?

aspiers · July 15, 2020, 10:05pm

This might sound stupid or weird, but it is a rhetorical question intended to provoke thought and discussion: what is the main goal of MOD Labs? Is it to build a range of products, or to build an audio platform which can evolve into a rich ecosystem?

You can probably tell by the way I worded it that I strongly believe the second option is better. This is because when you have a rich ecosystem, the demand for the core product (the MOD hardware device range) will naturally grow, and so will MOD’s revenue.

Building Open Source technology is great, but IMHO it’s a huge missed opportunity if you don’t invest into seriously scaling the developer community. The main benefit of FLOSS is that anyone can join in and create great new things. If they don’t, that benefit is lost. The current community is fantastic, but it is tiny and I don’t see many signs of it picking up any serious momentum.

Getting individual 3rd party companies to develop or port plugins to MOD, like was done with the Dwarf Kickstarter, is really really cool, but this alone is not a scalable approach - it would take years to grow this way. And obviously just having MOD Labs do all the work is not scalable either.

So I would like to humbly suggest that MOD thinks more strategically in terms of growing the platform, not just the product range. E.g. dream big - what would be required to attract 100 or even 1000 new plugin developers to the platform? I think it might not take much work actually, because already the software and documentation is pretty good. The main thing missing is probably the economic incentive - plugin developers can make much more money coding for other audio platforms. So how can this be fixed?

One way of jumpstarting more plugin development would be to facilitate Reverse crowdfunding section for new plugins and make it really easy for our great little community to crowd-fund new plugins / improvements to existing plugins, and drive plugin development in the direction we want. Everyone would win from this:

Existing MOD users get the plugins they want. The only constraint is how much money they’re willing to contribute.
Plugin developers (especially existing freelance FLOSS audio developers) get new opportunities to earn money in a low-risk way, since the money is pledged up front.
MOD benefits since their hardware devices become more attractive with every new plugin developed.

I can’t imagine it would be too hard to facilitate this - it would probably just need one person who is good at marketing and building communities to spend some time on it, reusing existing platforms (e.g. gitcoin or BountySource).

If you are reading this and thinking “why is this crazy guy talking about platforms?”, or you aren’t convinced of their importance, please read this amazing rant by one of the best developers I’ve ever come across. Granted, that is a totally different context and different type / size of platform, but I’ve worked with many software projects over about 25 years, and I’ve seen many of them live or die based on the strength of their community and ecosystem. The same is true of FLOSS projects, and commercial audio products. Would Pro Tools be as successful as it is today if it only had a few other companies building plugins for it?

Hope this is helpful, or at least thought-provoking. I would love to see MOD become the next Boss or Behringer in 10 years from now, but based on FLOSS

Klaustrophil · July 16, 2020, 12:53am

Great topic! I was thinking about this a lot in the past. I’m just adding my thoughts now. Warning, huge post ahead

For me, the ultimate selling point for the MOD products is the bridge it is building between the PC and the hardware world. I can use my hardware instruments (which I prefer over any DAW) but I got the power of software. With that, I don’t mean only the DSP for the effects but also the framework. It is open for extension and modification. I’m still blown away by the fact that I can use my digital mixer as an audio interface to get a lot of I/O. This is basically all I need to empower my 10 voice (18 channel) hardware setup.

But I think I’m part of a small minority. The tech savvy users who are into hardware and like to have increased influence in the products they’re using. I can’t imagine that there are a lot of us because it would be a way easier to just use a PC which has more power and more possibilities.

Of course, my wish would be to always keep the platform as open as it is right now. And I think that the MOD crew has the same mindset. But when it comes to plugin development, we should be careful about what we’re demanding.

First of all, it doesn’t make any sense to develop a plugin solely for the MOD platform unless you get paid a lot of money for that.

The current effect plugin market it very saturated. You can spent thousands of Euros in plugins which are sounding breathtakingly good. But even 100 Euros are enough for really good stuff. Sometimes it’s even for free. But one thing is clear: it’s all proprietary and it works. I can’t think of a lot of outstanding FLOSS effects. And I can understand why developers don’t want to give away the “secret ingredients” of their plugins to everyone. The market is so competitive that you don’t want to risk that someone else is just reusing your code to make money.

So my conclusion at this point is: we already can call it an achievement if a good number of developers of proprietary plugins are accepting that the MOD platform is open and that there will be no fancy DRM in place. Demanding anything in regards of open source code is very unlikely to work.

But let’s leave the FLOSS and DRM topic aside and ask: why should any developer port their plugin to the MOD platform? From the perspective of a (casual, average) developer who just heard about the MOD platform there are some showstoppers:

I’m not familiar with the processor architecture which is also slower than most PCs
The system is based on Linux which I’m not familiar with
I don’t know anything about this “lv2”
I have no clue how to start playing around with the platform without buying a MOD device
I have to completely re-write my GUI to fit this rather limited representation

Overall this looks like a lot of work just to get started. How much is the expected ROI again?

This might sounded a bit provoking and exaggerated now. But besides the competitive and saturated market I know that developers are lazy. As long as the developer isn’t totally stoked by the MOD idea itself or is being payed in advance, I don’t see big chances that the dev will overcome this obstacles.

Of course this is where the crowd funding could come into play. But it’s hard to say how much money is needed as a sufficient incentive. I think the only way to find it out is trying. Let’s find a developer of a great plugin who’s willing to port it, provided that the following crowd funding succeeds.

When we want to convince devs to port their plugins without a crowd funding, I believe we should at least try to lower the entrance barrier significantly. Some kind of development platform would be helpful.

The first thing which comes to my mind is MODEP by Blokas. However, it differs to the MOD platform a lot when it comes to the detailed software configuration. It’s outdated, some functions are deactivated and the OS is not very optimized. But maybe getting this right is a quick win? A Raspberry PI 4 is reasonably close to the actual MOD hardware and it’s just COTS which means zero hardware work for MOD, assuming the dev owns a class compliant USB audio interface.

Another solution could be some kind of development VM/Container. But there are some additional obstacles with getting audio and GUI out of a container while fluent audio is an issue within a VM. It might be also difficult to simulate the MOD hardware limitations.

In the end, I’m not sure how I feel about trying to convince random developers to build plugins for this platform. Having strategic partnerships with some elite devs could simplify many things. And maybe less is more? Who needs 2k different effects anyway? On the other hand, a working platform could be a more sustainable business model. And we also see that a high diversity is somehow working out in the PC world. There could be also some nice side effects like an increasing number of commercial Linux plugins.

aspiers · July 16, 2020, 1:04pm

This already exists: https://wiki.moddevices.com/wiki/MOD_SDK

So there is no need to build a new one here. That’s why I stated my belief that not much work would be required to make this happen. The quick win is already there waiting to be seized.
I’m sure there is room for improvement with the SDK / docs / Docker images etc., but that can happen incrementally. IMHO the main barrier is economic / organisational, not technical.

I don’t think anyone was suggesting that. My suggestion was to coordinate funding of bounties for specific plugin work, and the community can pick which dev(s) it trusts the most to deliver good results out of those who have indicated interest. In practice, I think it is likely there would be a naturally high correlation between devs who are interested in the work and devs who are good at it

I agree there is a risk that the money raised might not be enough to incentivize increased levels of development. But we will never know unless we try.

Klaustrophil · July 16, 2020, 1:56pm

We are both insiders to this topic. We know how to work with Linux and what tools to use there. We are used to the pain working with all this quirky FLOSS software. But I don’t think this is very helpful for most developers coming from the Windows/Mac VST world. The SDK just a GUI designer and a cross compilation environment. There’s still no test environment without owning a MOD device. I still have to read into this whole lv2 topic and figure out how it works. There’s no interface provided for debugging my code on MOD hardware. Sure, a good dev can figure this out. But only with the right incentive.

You suggested to think about attracting hundreds of developers So for me, this is one of the next steps. If a dev is interested in creating Linux audio plugins, the chances are very high that this already happened. In my perception the MOD platform in it’s current state is already a pretty comprehensive collection of available (free) Linux audio plugins. I don’t think that there’s that much to gain in this area. But there are some low hanging fruits which could be eligible for a first funding:

Fix/extension of existing plugins (Some major changes might be nice but I don’t see a huge the impact to grow the customer base)
The port of existing free DSP code to lv2 like from the VCV/Eurorack world
The port of existing commercial Linux plugins to ARM/lv2

After that, there’s only the commercial Windows/Mac market left.

Yeah I’m still in to tackle this topic

plutek · July 16, 2020, 2:09pm

i’ve got a bunch of pedalboard dev to do right now, so i won’t get into a detailed discussion…

let me just say that the principle reason i’ve adopted MOD’s line of gear so wholeheartedly is precisely that it is an expression of a larger ~idea~ about how integrated hardware/software solutions can serve musical needs, and that it is centered around an active and communicative community of users and developers.

so, yes, finding ways to more explicitly harness an integrated world of software, including plugins, with a view towards unifying the hardware product line, seems like a natural expression of how this community (the MOD folks, the users, and 3rd-party devs) thinks.

to extend the “platform” idea: i’d say it would be a huge leap forward if there was a unified MOD control UI, which allowed users to organize an entire collection of MOD gear. …i.e. implement an inter-device communication system, and then make it possible to build “meta-pedalboards” which leverage the combined, distributed capabilities of all the interconnected MOD units.

khz · July 16, 2020, 3:07pm

Don’t know if this is off-topic or even helpful, but there is an
LV2 wiki >> https://github.com/drobilla/lv2/wiki.

aspiers · July 16, 2020, 6:16pm

I was under the impression there was - with the combination of the SDK, MOD Plugin Builder, the various LV2 dev tools etc. Maybe it’s harder than I thought, in which case of course some work would be needed to make development easier.

We are in violent agreement. That’s why I was talking about creating the right incentives

Yes, but “attracting developers” (my words) is different from “trying to convince random developers” (your words). I think the carrot is more effective than the stick

But again, I think we are basically in violent agreement.

Jan · July 16, 2020, 7:09pm

I think this is what you are looking for:

@falkTX has just recently updated it and can probably give more specifics

gianfranco · July 16, 2020, 7:22pm

Loved the topic guys.

@aspiers: you rock

@Klaustrophil: you’re spot on about the showstoppers for developers.

The platform is surely the vision and yes, scaling up the developer community is a must. I’ve been toying with ideas around this on a daily basis.

What I can tell at the moment is that MOD Labs is coming as an MVP in our next software release.

I will follow-up on this topic after the release

aspiers · July 16, 2020, 8:25pm

Thanks @gianfranco, sounds very exciting!!

Steve_Lee · July 23, 2020, 7:08am

Great thread. One thing to keep in mind is that with a few notable exceptions FOSS projects are for developers rather than end uusers.or af least those very technical users who are happy to be both.

Another way of looking at it is end users usually just want fully supported products. Some also want them to be part of of a vibrant ecosystem.

Yet another way to view it is that for companies customers and community are not always the same group. They need different interactions.

Personally as a long time OSS dev I want a mature open ecosystem. Yet as a user I want my product to be reliable, consistent and fully supported. Perhaps the messaging was a little confused or just my assumptions as I was surprised when @falktx pointed out I could not install the full image I had built onto my modx. But I do understand the commercial sensitivities at play Here.

But the Modx is a great product AND a really nice platform

Thanks guys!!!

aspiers · July 23, 2020, 12:01pm

There are many notable exceptions, not just a few.

This is true. It’s also true that end users often don’t know what they want until they experience it, e.g. they will often underestimate the value of a vibrant ecosystem around the product they buy, until they start exploring that ecosystem.

However what a lot of people miss is that it is possible to pay for support of FOSS products (and earn from it), mistakenly assuming that if you want a supported product, you will have to use proprietary technology. Of course this is totally wrong, as proven not only by MOD but many other companies (including one I worked for for ~15 years, which currently has an annual revenue of around $400million from FOSS products and services).

Agreed, although there usually isn’t a black and white separation between customers and community, but rather a large overlap with a full spectrum of behaviour: customers who are very active in the community, customers which occasionally interact, and those who don’t at all, and everything in between.

Steve_Lee · July 23, 2020, 12:28pm

There are many notable exceptions, not just a few.

oops thanks for pointing out the unintented weasle word!. I meant ‘some’

However what a lot of people miss is that it is possible to pay for support of FOSS products (and earn from it), mistakenly assuming that if you want a supported product, you will have to use proprietary technology

quite - so called “System Integrators” for example. also unfortunately some seem to think FOSS means ‘free stuff’ rather than a whole range of possible business models plus self service help that’s almost impossible to get with proprietary.

Anyway I don’t want to side-track this thread. I’m all in for supporing a vibrant ecosystem around the awesome MOD platform, as well as best of breed products. FTW!

redcloud · July 23, 2020, 1:15pm

As end user I constantly look for continuous improvements over time against “planned obsolescence”. I hate companies that abandon their products after 3 years since release to market. If the hardware has some potential to run updated software, why not work on it even charging end users for update costs? FOSS should make this stuff easier.

Klaustrophil · July 23, 2020, 1:28pm

Yeah my wording was exaggerated. I just wanted to say: It could be a possible strategy to target as much as developers as possible. If 3% of the targeted devs decides to publish for MOD it could already kick start the platform and the investments would pay off quickly. But the risks might be hard to calculate.

Never heard of this idiom Just to make clear: I don’t take or mean anything personal here, nor I want to spoil this idea. For some reason I often tend to be this annoying, questioning, polemically challenging voice…

[troll] From the UX point of view you can count end user friendly FOSS products almost on one hand[/troll]
SCNR

I’m really looking forward how the MOD Labs will look like

aspiers · November 30, 2020, 10:31pm

This discussion just prompted me to think about this a bit more. Based on my professional experience of working in other FOSS communities, I suspect the only realistic way to kickstart growth of the MOD platform and developer community would be to hire someone to be dedicated at least part-time if not full-time to being a community manager and focusing on creating the right environment and incentives for more people to get involved in development. I’ve seen that model work for FOSS projects in the past, and I’m guessing with the right person it could facilitate tremendous progress perhaps in only 3-6 months.

But in case anyone is getting suspicious of the motives for that suggestion, no, I’m not looking for work