As much as I like to see new features added - to a point, they need to be well implemented and without making a device clunky to use. Some gear has too many features to be a bit overwhelming, other gear has infuriating implementation or daft limitations.
But I don’t think commercial products should be open source, I like open source stuff but I don’t think it generally fits with commercial products, personally.
Also I much prefer ‘the old way” of product release, get a smallish bunch of knowledgable early adopters or beta testers, and get the firmware fully fleshed out and feature complete before it hits the general market.
I think there is a hell of a lot to be said for focussed and well designed firmware, with solid reliable performance, but it seems less common these days, feature creep and flaky performance are the quickest way to turn me off of an instrument.
Entitlement is a funny thing, it means different things to different people, for me it means if I had to work hard to afford a product I should reasonably expect it to work reliably, and to be serviceable for a long time after purchase. On that score I think Elektron fare pretty well, I can’t say the same for some other gear that I own.
But it’s not just one app among many, it’s the OS.
If you are deploying to containers in a cloud, and you brick something, you just destroy the container and spin up a new one. You can’t do that if a piece of hardware in a user’s studio stops responding.
Yes, it’s possible to make it work, but it would be a very different system to all the current Elektron devices, and not something you could just add as an after thought. It might be possible to design a future product to work that way (and maybe that’s why they wanted a new CEO with SaaS experience?) but I think it would be impossible to do right for the existing devices.
WRT the updates, not all Elektron users are into computer stuff, so the less updates the better for them. But Elektron did open the beta testing program for Overbridge. Perhaps they could do this with firmware too. Testing hardware is not easy so the more the people interested in the beta testing the better for all of us.
I very much welcome updates that add useful stuff. Everyone loves the last Digitakt update. What we don’t want is false promises, like when the Digitakt had “Overbridge” printed on the box. It just was not true and took years to implement.
See the current line of MPCs. All the features that have been added since 2017 were totally identified at launch. Heck, some of them were even available in previous generations (yes, even plugins). Yet more than 4 years after launch, still no streaming in spite of Akaï representatives repeteadly stating it was in the works. They just released an unfinished product… because “we’ll add that later”.
Agreed - existing products are probably out of this CD scope - but I just wanted hear everyone’s takes on how you would like this to be for the best experience/solution for you?
So, I hope the hw device companies are really considering CD as an option for future products.
Basically, what I want to overcome (CD or something else), is the tendency to “ahh come on, we need to get just another fix or features in this build, before we release…” <- we know how this goes
For the OS or software to work as open source in a commercial product, you need a recurring revenue stream that isn’t based on that software. This works if the software is just a commodity and the real business is actually selling value-added extras, like annual subscriptions to unlock “premium” features that aren’t in the open source version, or IT support contracts or training programs, or the hardware is so unique/niche/desirable (and expensive) that the customers are already locked into buying the hardware, and the software can just be thrown in on top. Or maybe giving away the software is actually an investment to get user engagement, because that drives the actual profits (e.g. you don’t pay for facebook, because other companies are paying for access to your eyeballs, and you don’t pay for Android because Google wants access to users’ eyeballs, emails, photos, etc.).
I don’t think any of those business models would work for Elektron. The hardware is good, and but not distinctive enough (and doesn’t sell in large enough quantities) to sustain the whole business if the firmware could be loaded into “hackintosh” style DIY hardware, or a Blektron hardware clone.
They probably could do it already, but constant updates would mean people would likely get annoyed with updates. Then you get people on different minor versions.
Also, regardless of the automated testing, there’s more chance of bugs getting through then having to release fixes for those.
People want updates more quickly, but with instruments you don’t want to do risk things breaking too often.
I guess making bigger releases makes thorough testing and subsequent fixes easier to manage and more stable releases possible.
I think you can always overcome this, by agreeing that testing always needs to be done properly.
Quickly should NEVER mean lesser quality.
Small release can be tested perfectly as well as big releases, without using 2 weeks testing for 1 small improvement - it really all comes down (90%) to how you have structured your code and tests, and how many hands you have to do a manually functions test.
I haven’t really needed anything that’s been added to phones since iPhone 8, I was perfectly content with it. I only upgraded because having a new phone barely cost more than changing the battery and broken screen. I don’t know what they can really add to phones at this point tbh
Right on, I guess I sometimes have expectations about what might comes as future features, maybe we all do, because we get used to the idea of “the next update” but I try to be realistic and not be a knob about it.
Seen some pretty daft feature requests and expectations posted before though, worst kind are when people want to change hardware things through a firmware update, or that they don’t understand that a hardware limitation makes the feature they demand impossible, despite the manufacturer making it clear in the specs/manual/website.
Well what they do is phase something out over a few generations, make sure the suckers customers get used to the new shiny, and forget about the old, before re-introducing it as a new feature on the new model. Profit.
This may be true… but I don’t want my hardware connected to the network and I want the people who make my hardware to run their business on the understanding that my hardware doesn’t connect to the network more than once a year.
Any other pattern leads to a reduction in my ownership, and likely interruption in my usability.
Can you imagine taking your sequencer on stage and it popping up a “your software is out of date” just as you’re hitting the drop?
Insane.
(I know a finite state machine driven UI could avoid that problem, but that’s just narrowing the focus down onto specifics. The whole business-product-ownership-usage-maintanence system for instrument-like hardware needs to operate on the premise there is no network for most of the time)
I actively choose products from companies that upgrade their software and support their devices for years on end (for example Elektron and Novation ), instead of companies that abandon their stuff because the Next-new-shiny-Thing™ is around the corner.
Considering that us customers are often treated like free Beta-testers, I also dont think it is entitled to demand some free software updates after launch. Bonus points if the company actively listens to user-feedback and doesnt operate from an ivory-tower with the “we know better what you want than yourself”-attitude.
I think the “back in the days, we got bugfree, feature-complete hardware on release and we were happy forever”-argument doesnt make a lot of sense in a world where every product around us like TVs, smartphones, coffee machines, vacuum robots and whatnot gets free firmware updates all the time. Nevertheless I agree that products these days often release too early and get fixed later exactly because of this…
Financial sustainability for the manufacturer is definitely the other side of the coin because dev time is expensive but I happily pay a little more on the retail price if it meant that a product gets a limited support window for software updates.
haha I can see the sequencer updating… of course we don’t want that And just to clarify, I do not want my devices to be on the internet. I just think you could split the firmware into different types of releases.
My experience with “firmware” is, that is is a package including a sh*tload of code not actually relevant to the real lowlevel firmware - meaning, a lot of the code is really not updated, like the application layer.
“Usually” it comes with a middle layer and on top the application layer. Those could be structured to be updated separately. And the user would be free to choose when ever they wanted to update their device - not different from today.
I like the clarity you’re bringing to this, but I still disagree when it comes to my instruments. It’s annoying enough when my text editor tells me there’s an update instead of letting me write the note I want to write. Docker hits me with a modal popup twice a week at the moment. It’s rubbish. Don’t make me deal with this shit on my instruments. Creative motivation is finicky at best; constantly maintaining my instruments’ software/firmware will kill it dead. I’d take up the piano.
), instead of companies that abandon their stuff because the Next-new-shiny-Thing™ is around the corner.