Questions & thoughts about VVVV development

[lasal]

Creating a language can be a very time consuming and alienating process for the creators, for this reason the users can give a reliable perspective about how to use this language, creating vocabulary and polishing the grammar aspects that are obsolete or not intuitive to use.

The best way to create something, is implementing the theory in a real situation on each step of development iterativelly. Of course doing that is a quite time and effort consuming task.

I would like to invite all the community to help the developers to understand the current state of needs, not only foccusing on features but also in other escential topics.

Assuming that the current VVVV development is focussed on
Designers.
Artists.
Engineers in arts.
Developers with design & art background.
Developers without design background but building tools for designers & artists.

I would focus on that part of the spectrum.

Topics to discuss that are around my mind since a while:

When you use VVVV in your work…

1. VL Grammar:
   1.1 Do you feel that VL language should be more than a visual translation of code?
   1.2 Do you feel that you spend more time solving problems than creating outcome?
   1.3 Do you think that you work faster than people using code?
   1.3 Do you miss strategies to translate code based languages (Java, Python…) into visual language?

2. Features:
   2.1 Do you find yourself buiding the same things on every project?
   2.2 Do you miss something that under your perspective is escential in your work with vvvv?
   2.3 What do you miss that was already in VVVV Beta?
   2.4 What do you think VVVV has lost along the way from Beta to Gamma?
   2.5 Do you envy caracteristics in other similar tools (or pieces of software)?
   2.6 Do you miss industry standars?
   2.7 Do you feel embarased when your client ask for something that already exist in the competence but not in VVVV?
   2.8 Does the VVVV UI make you feel confortable tweeking parameters (like setting up colors, or animating 3D cameras)?
   2.9 Do you miss a more designer oriented way to implement concepts?

3. Documentation:
   3.1 Do you think that the documentation is oriented to the current target VVVV user?

4. Education:
   4.1 Do you miss more workshops/lectures/documents by the VVVV developers or specialists in your field?
   4.2 Do you miss strategies to learn new concepts?
   4.3 As teacher, do you think that you are teaching meanful material?
   4.4 As teacher, do you think that you are teaching concepts that should be explained by the VVVV developers?

5. Community:
   5.1 Do you think that the most of the users don’t acknowledge the ammount of work / effort / knowledge behind your contributions?
   5.2 How supportive is the community in the chat?
   5.3 How supportive are the developers in the chat?
   5.4 How do you feel about speculating with knowledge in an open source community context?

• Your turn:
  How do you can help to solve any of the previous topics?

• Please answer and improve this collection of thoughts. Don’t be shy but be constructive and sensitive.

I hope i don’t make anyone feel bad, i’m just doing this to create a healthy discussion about something i love and i’m proud of.

[lasal]

I just want to be more specific about certain missing features.

1. Inspector.

2. Data Binding (via inspector or programmatically) with:
   2.1 A powerful, flexible and intuitive Animation engine.
   2.2 A powerful, flexible and intuitive Data Input engine (OSC, MIDI…).
   2.3 A powerful, flexible and intuitive Linear Animation engine (Timeliner).
   2.4 Clocks (Time & BPM based) & Industry syncronization standars (MTC, LTC, Ableton Link).

3. A mechanism to translate code into VL language.
   In order to implement new concepts found online, in the best case, first i have to search for compatible C# material & translate to VL. The most of the cases, the online documentation is written in other languages that first, i have to understand, and second, i have to rewrite in VL. That creates a clear desadvantage in comparison with, for example, Python programmers.

4. A powerful, flexible and intuitive Editor Framework
   4.1 3D Mapping / Warping / SoftEdge editor.
   4.2 3D Mesh editor.
   4.3 3D, 2D Spline, Bezier curve editor.

[thomashelzle]

Nevermind…

[lasal]

I think this is just a place to collect ideas and discuss.
Let’s see how it takes shape and how it continues.

[lasal]

I would love to have a VST host in Gamma

[thomashelzle]

Nevermind…

[tobyk]

@lasal I like your initiative here to start a discussion.
I think your question format is a good start, although in fairness to the Devs it’s perhaps a little negative.

It feels to me like this thread is becoming somewhere for users to post longform detailed critcisms and wishes, which is valuable. I will also contribute my thoughts soon.

At the same time I think it would still be good (as I was proposing in the chat) to conduct a user survey to collect a little bit of quantified data on how people are using vvvv. I think with a more neutral question format that has room also for positive feedback. I will make another thread about this.
.

[lasal]

@thomashelzle

@lasal

5.4 How do you feel about speculating with knoledge in an open source community context?

Any speculative way to create value in a competitive contex, like holding back meanful information that makes you more competitive than others.

It’s one of the columns of capitalism, if not the principal, but not so common in other economic systems where the knowledge has no “owner” and the authorship gets vanished in favor of a collective growth.

I know we all do it, but the question is if it makes sense in an “open source” community.
I’m not saying that VVVV or the community around it is open source, but rather wanted to make that part visible and create a bit of self reflection.

[lasal]

@tobyk

I think your question format is a good start, although in fairness to the Devs it’s perhaps a little negative.

Sorry if it has this tone, it’s not my intention, i try hard not to be bias. And of course i deserve the criticism.

[lasal]

@thomashelzle

A massive undertaking like Kairos can’t concentrate on the “philosophical journey” as Natan put it in the chat but since it is the only curve editor, users would like to also use it for simple keyframing and then collide with the very hard to use GUI.
And IMO that is one source of tension, that on one hand, something like Kairos is a massive piece of work, and on the other, working with it for simple keyframing is no fun to put it mildly.
So Natan is frustrated since his love child is undervalued and the user is frustrated since he’d rather have a right mouse click menu and a fluid to use GUI than some esoteric blending, as fascinating as that may be.

This is a problem that is being around since a while with contributors, that in my opinion (and many others think in the same way) could be solved creating a Foundation around VVVV, where the users can decide where to put the money on.

Look at Blender example, it was a niche tool and now is the industry standard.

[yar]

@tobyk

It feels to me like this thread is becoming somewhere for users to post longform detailed critcisms and wishes, which is valuable.

I propose that one day a month we all get together and talk about how VVVV has changed our lives. We will declare our love for the developer, admire his talent and describe our favourite solutions.

In general, I guilty of criticising some of the solutions used sometimes. But I would like to say that I love VVVV and everyone involved in it with all my heart. I believe and hope that the developers understand that my goal and purpose is to help develop, not to criticise. I try very hard to strike the right tone. I’m sorry if I don’t.

Perhaps we should start our comments with this kind of disclaimer.

[tobyk]

@yar if I responded to that I would be criticising your criticism of my criticism about Lasals criticism. I think everyones heart is in the right place and all this discussion is valuable.

[yar]

@tobyk

if I responded to that I would be criticising your criticism of my criticism about Lasals criticism. I think everyones heart is in the right place and all this discussion is valuable.

Yep! I understand
I wrote rather in agreement with you.

[lasal]

Regarding default format in IO Boxes:
Floats and vectors should have higher resolution, just 2 digits is not enough for the 90% of the cases.

Maybe the motivation was to have a much lighter UI, but right now the UI looks lighter enough to display more digits.

I would say that 4 or 5 digits by default would reduce the users level of frustration.

[tobyk]

Ok here’s my thoughts on vvvv development at the start of 2024.
It’s an essay rather than answering the questions.

---

Positive

I want to start by acknowledging that I think vvvv gamma and vl is a huge achievement.
IMO it’s the worlds first working visual programming language that gets beyond ‘wires’ to actually being able to describe programming paradigms with the same complexity as professional textual coding languages. I cannot stress enough how big of a deal that is.

If you look at media art tools of similar vintage as vvvv (Eg TouchDesigner, Resolume, Reaktor, Ableton Live, Arkaos), to their credit those tools are all still around but I don’t think any of them achieved such a huge evolution as vvvv over the last 10 years.

To me vvvv is the clear leader in real-time visual programming today.
In that same time some frameworks have seen the value of visual coding and implemented their own node based solutions (Resolume, Unity, Unreal) and new tools have appeared (Bitwig, Cables.GL) but these visual environments are at most using vvvv beta type concepts.
VVVV gamma is literally a decade ahead.

I also want to acknowledge that vvvv group is a small business, not a huge corporation or an overhyped start-up. And a small business that has prospered for over 20 years and achieved more with much less resources than corporations and start-ups, also with an incredibly generous licensing model and many hours spent on near real-time community engagement and support.

---

The cost

vvvv has had absolutely breakneck feature development over the last 10 years. Check it out:
https://thegraybook.vvvv.org/roadmap/past.html

This feature development has made vvvv capable of general software development, which as I say I think is amazing. But there has been a cost.

IMO The cost of such a tight focus on shipping features and expanding general programming functionality is that documentation, UX and media arts specific features are not where the community expects them to be for a media arts tool.

I think a good example is the Reactive functionality.
Reactive provides an incredibly flexible and robust way to deal with all IO.
And crazy IO is close to the heart of interactive media arts, MIDI, OSC, LED protocols, Key/Mouse, MQTT etc.
Reactive enables message based programming where all of this IO can be reacted on with different timings to the mainloop rendering.
It’s a complicated paradigm but was necessary to implement to really unlock the full potential of these protocols without restriction.
I would say Reactive is an example of a general programming concept that brings great benefit to the media arts specific Iusers.
YET IMO Reactive is relatively poorly documented and poorly understood by the media arts users that it is there to benefit. The learning curve actually got steeper to use protocols you thought you understood from vvvv beta.

What I’m trying to illustrate is that the huge investment in general programming functionality over the last 10 years with vvvv gamma is not always able to be accessed by all media arts users.

(Personally I am planning a simplified reactive tutorial video, hoping to improve this particular situation)

---

A proposal

Now that vvvv has a strong backbone of general programming functionality, what next? Because although it’s niche at the moment the potential is huge. You can write almost any kind of app with vvvv if you wanted to.

In my opinion it makes sense to focus again on features that improve the experience, capability and commercial viability of media arts users.
Consolidate all the incredible general programming capabilities of the last years by making them better documented and easier to use for media arts users.

Two reasons
-Media arts users are still the core users of vvvv, and as other posts in this thread discuss if you pick up vvvv gamma today some things are harder than they should be. Serving these users better and helping them thrive helps vvvv thrive in the short and medium term. Media arts is a good niche and vvvv is well positioned as the language that lets you really take control of what a machine does. I think serving these users better will grow these users.
-Long term I think the more that general programming can become like media arts programming (Eg fun, inspiring, creative, real-time) this is how vvvv gets closer to conquering the world.

I think we are already going in this direction, features like Global Channels and recent UX improvements are a great start.

---

Specifics

I am also advocating for a user survey, I would revise this list on seeing the survey results.
But as a first hunch I think there are four areas of action:

Beginners

Issue is still a huge learning curve coming to vvvv gamma. Now you have to think like both a Media Artist and a Software Developer.
Because I’m not a beginner it’s hard for me to appreciate their perspective.
I would suggest
-Interviews with students and teachers to find out what blocks them and improve it.
-Improve the F1 functionality further, both with better coverage of factory libs and maybe putting the ‘this node also used in’ patches directly in the tooltip.
-Improve visual communication options on a helppatch so they can be better tutorials. Could be improved formatting of links (Hidden links…), images, videos embedded on the patch etc.
-Help library developers improve the F1 coverage in their libraries, perhaps a tool to see which nodes don’t have F1 links. Incentives or prizes for best helppatches?
-I like @thomashelzle suggestion of more short intermediate videos that try and break open a complex concept for everyone to understand.

Pro Media Arts

I think the other posts in this thread already point to a number of improvements that can be made for these users. Its looking like quite a big backlog of small and big things.

Not all of this is responsibility of vvvv group, some of it is Fuse and Kairos and other library developers and thought needs to be put into how to ensure everyone feels good about their contributions when those are perceived as essential to vvvv as a media arts tool.

Long term in this area I would look towards innovations in the VL language and gamma editor that make complicated programming concepts easier to use. Like @sebescudie suggestion of visualisation in reactive links when a message passes through.

On the community side I would like to see more sharing of working media arts patches. The helppatches are a great start but imagine a kind of shadertoy vibe where people frequently share working ‘toys’ as inspiration. Also the amazing initatives we have had in the past to show how the generative gestaltung processing examples could be achieved in vvvv. Maybe some way to incentivise this kind of share-out of media arts practice? Cables.GL has a good setup of monthly featured patches, maybe something like that in help browser?

General programming

As I said I think general programming features should be a lower priority for a little while, but not completely zero. I think it’s always worth doing some work here because it’s the groundwork that enables the community to help itself and empower vvvv further.
-Upcoming work on attributes looks exciting, will enable more auto GUI and better auto serialisation.
-More tutorials that enable people to wrap .net libraries and become library developers themselves. Or improve the standard lib to eg get more out of stride.

Moonshots

Incredible innovation is still at the heart of vvvv, and still gets us ahead.
Still got to have a moonshot going on in the background, eg
-vvvv on macOS
-vvvv in the browser
-vvvv export for other platforms
-vvvv editor as a product that can be implemented in vvvv applications for end-user scripting, or to make ‘mods’ of vvvv applications.
-realtime collaborative programming (boygrouping)
-What is the role of a visual language in a future where lots of the ‘interior’ bits of code will be AI generated? It’s easy to tell ChatGPT to solve programming problems “write code for rectangle intersects rectangle” but it’s not easy to tell ChatGPT to solve design problems “Write me an app for playing music, it should have a sign up screen, then a terms and conditions screen with different t&cs for each language, and then a home screen like this, and then a play screen like that and then and then and then”.
I think describing a quality experience will always involve a visual language. (Like current development practice where you show a real developer a figma file)

---

That’s my thoughts on vvvv development.
Looking forward to using it for another 15 years.

[lasal]

@tobyk I couldn’t articulate it better, agree on 99%

[lasal]

@tobyk

To me vvvv is the clear leader in real-time visual programming today.
In that same time some frameworks have seen the value of visual coding and implemented their own node based solutions (Resolume, Unity, Unreal) and new tools have appeared (Bitwig, Cables.GL) but these visual environments are at most using vvvv beta type concepts.

Well, PureData and Reaktor were already there since the 90’s.

[tobyk]

I didn’t mention PureData/MaxMSP because I would say it truly is in an earlier generation. But also something vvvv beta improved upon vs their ‘everything is a bang’ paradigm.

Reaktor I would say is comparable with vvvv, just a few years before.

[thomashelzle]

Nevermind…

[thomashelzle]

Nevermind…