Re: [Mixxx-devel] controller mapping guidelines

[Daniel S. <das...@mi...>]

Hi Tuukka,

this is not an easy question.
There will be a benefit for the manufacturer if we build a first class
mapping for one, but not for an other.

If we by a controller via the normal distribution chain, there are many
parties which earn money from that.
This feels like a bead deal compared to a free sample from an manufacturer,
which counts only the
hardware cost, on a depreciable marketing account.

Also the mapping work itself is an issue. Why should a contributor spend
time on a second mapping when
he has already a working setup at home? How should we decide which
controller will be the next when there
is no personal interest.

This deal sound better to me: A free controller for a mapping included in
the Mixxx setup and an entry on a recommendation list
sound fair to me.

Kind regards,

Daniel




2015-11-23 7:44 GMT+01:00 Tuukka Pasanen <pas...@gm...>:

> Hello,
> Should we have some fundraiser to buy some controllers for devs to make
> good mappings for the high-end contollers?
>
> Tuukka
>
> 23.11.2015, 00:22, Be kirjoitti:
> > (I'm consoldiating these related threads into one.)
> >
> > On 11/19/2015 08:18 PM, RJ Ryan wrote:
> >
> >   > The standards are much lower since there is little risk to doing so.
> If
> >   > we took in C++ patches without review we could cause crashes or
> actual
> >   > damage to the user's computer. The sandboxed environment we run
> >   > Javascript in is not capable of this. While it's true that Javascript
> >   > can trigger unexpected behavior in Mixxx there is not an effective
> way
> >   > to explore the set of possible inputs a script will provide to Mixxx
> >   > without the device in hand. In most cases, a preset is not going to
> take
> >   > down Mixxx.
> >
> > Mixxx crashing is beside the point to someone who has spent $600+ on a
> > controller if Mixxx doesn't even support the controller.
> >
> > Reviewers have been saying for years that the haphazard controller
> > support is one of Mixxx's biggest weak points and not much has been done
> > about it till now. They're not just talking about any support because
> > that is beside the point for someone who spent money on a controller and
> > could use other software that came with the controller and does support
> > it. They're talking about fully mapped controllers with documentation.
> > For example:
> >
> > http://djtechtools.com/2012/08/07/review-mixxx-1-10-dj-software/
> >
> > "Besides, Mixxx includes a good start on MIDI controller compatibility,
> > with Mixx Certified mappings for 13 controllers — including the Midi
> > Fighter Classic — and community supported mappings for 29 controllers.
> > They would love to have you fill in the blanks by creating a new
> > mapping. *The functionality and quality of documentation among the
> > community-supported mappings varies.* [emphasis mine]
> >
> > I tested Mixxx with the community-supported mapping for the M-Audio
> > Xponent. It was a nice way to resurrect a controller I had mostly placed
> > on the shelf. I was able to control Mixxx almost entirely from the
> > Xponent. With the exception of some browser functions and Sampler Deck
> > control, keyboard shortcuts could make up for what the controller
> > mapping didn’t provide."
> >
> >
> > http://djworx.com/mixxx-1-12-beta-still-free/
> >
> > "Setting up isn’t as plug ‘n’ play as commercial offerings, either, with
> > mapping options relying on DIY rather than a large existing library of
> > controllers."
> >
> > Including incomplete mappings like the Xponent mapping reviewed above
> > does not change this situation. It's still DIY to get one's hardware to
> > actually work.
> >
> >
> >
> http://www.digitaldjtips.com/2011/11/review-mixxx-10-1-0-free-dj-software/
> >
> > "Review Summary:
> >
> > Mixxx is now better than it’s ever been. If you’re an open source
> > enthusiast who knows a bit about Midi, XML and coding in general, you
> > can get involved and adapt Mixxx to suit whatever Midi gear you have,
> > but if you’re just a plug-and-play DJ, unless you have one of the
> > controllers it is already mapped for, *it isn’t going to be of much use
> > to you.* [emphasis mine]
> >
> > ...
> >
> > Midi and mappings
> > But the crucial area where Mixxx still lacks for me is in out-of-the-box
> > Midi control. Mixxx 1.10.0 comes with support of variable quality for a
> > small number of controllers, *but it’s not the ones that are selling
> > well today.* [This was four years ago and the situation hasn't changed
> > much.]
> >
> > Now, it is perfectly possible to produce your own mappings. Indeed
> > there’s a Midi Learn option where you are talked through the various
> > controls to get a rudimentary mapping going in a matter of minutes (I
> > got the Vestax VCI-400 we are currently reviewing partly controlling the
> > software in less than five minutes).
> >
> > But getting your mapping 100% right? That’s harder. Jogwheels are
> > famously the hard bit about Midi mapping, and you need to get your
> > sleeves rolled up and start hacking in order to add this kind of
> > functionality to a custom mapping, using Midi sniffer apps and writing
> > XML. If you thought mapping Traktor was hard, wait until you get stuck
> > into this beast. There’s a friendly user community and an excellent
> > wiki, but plug and play it ain’t, unless you have one of the controllers
> > it natively supports."
> >
> >   >     None of us are lawyers, so we should not be making decisions
> about a
> >   >     legal grey area. I think the legally clearest and safest
> > situation would
> >   >     be to require mappings to be licensed under the GPLv2 or later
> > and have
> >   >     mapping authors sign the agreement.
> >   >
> >   >
> >   > I understand your position here. I've consulted lawyers about this.
> I'd
> >   > prefer to not discuss this any further on our public email list.
> >
> > The question of legality isn't my only concern about taking mappings
> > without explicit consent. I am also concerned about the quality of the
> > mapping. If mappers have to submit mappings themselves, I don't think
> > anyone is going to submit a mapping before they feel it is complete and
> > ready.
> >
> > On 11/22/2015 02:22 PM, Be wrote:
> >> On 11/20/2015 10:53 AM, Sean M. Pappalardo - D.J. Pegasus wrote:
> >>> On 11/19/2015 11:06 PM, Be wrote:
> >>>> Also, what devices do people want supported? IMO it is a big problem
> >>>>     that Mixxx lacks much support for popular brands that make quality
> >>>> hardware that is commercially available today. Where is the support
> >>>> for contemporary Native Instruments, Pioneer, Akai, and DJ Tech Tools
> >>>>     controllers?
> >>> Supporting newer controllers requires either manufacturer cooperation
> >>> (many of them have their hands tied when they do bundling deals with
> >>> other DJ software,) or funds to purchase the controllers ourselves,
> >>> which we definitely don't have because we have no income stream.
> >> Both of those would be great, but neither are necessary. For popular
> >> controllers, enough users come by Mixxx interested in a mapping. Soon
> >> enough one of them has the ability and time to make a mapping. These
> >> users need to be supported technically and encouraged, which I have been
> >> doing.
> >>
> >>>> ~3/4 mappings in Mixxx are for devices that have been discontinued.
> >>> That's due to a number of things 1) the market moves so quickly that
> >>> controllers are discontinued within a year in some cases and 2) a
> >>> majority of Mixxx users are in situations where second-hand controllers
> >>> make the most sense for them, so those are what people create presets
> for.
> >>> (This latter point isn't a bad thing because people new to Mixxx can
> try
> >>> it with whatever controller they already have or can get
> inexpensively.)
> >> The market moving quickly is not an excuse for not keeping up. We can.
> >> As I mentioned above, people want the latest controllers to work with
> Mixxx.
> >>
> >> I think a majority of Mixxx users are in situations where second-hand
> >> controllers make the most sense for a few interrelated reasons. It's
> >> kinda a chicken-and-egg problem. Before I overhauled the wiki this
> >> summer, it was difficult to tell what Mixxx actually supported, which
> >> made it confusing and difficult to pick a controller to use with Mixxx.
> >> So, it's not surprising that many people would get whatever controller
> >> they could find cheaply and hope it worked. Also, Mixxx has a reputation
> >> of being just a nice thing to try for beginner DJs without having to
> >> invest much, so these have generally been the people who came to Mixxx.
> >>
> >> I would like to see Mixxx outgrow this reputation. I would like to see
> >> new DJs coming to Mixxx and staying with Mixxx. I would like to see DJs
> >> switching to Mixxx from proprietary DJ programs and sticking with Mixxx.
> >> This requires fully supporting popular middle and high grade controllers
> >> like the Traktor Kontrol S4 and Pioneer DDJ-SR and DDJ-SX2. With 2.0,
> >> Mixxx is getting close to outgrowing that reputation in terms of
> >> features. But I think including mappings without any quality control
> >> will hold it back.
> >>
> >>>> Write the documentation, and people step up to do the work.
> >>> While good documentation is indeed essential, our experience with many
> >>> Mixxx users who want to add support for a controller is that they are
> >>> spooked even by the idea of editing an XML file*, let alone working
> with
> >>> JavaScript, despite the extensive documentation on the wiki which was
> >>> available when the scripting engine was released.
> >>>
> >>> *So often that one person made a lighthearted joke about it:
> >>> http://downloads.mixxx.org/mess/baddudes.gif
> >>>
> >>> In short, the average Mixxx user is not a developer by any stretch and
> >>> has no interest in becoming one. While it's fine to have a
> >>> developer-friendly work flow (i.e. Github PRs) _available_ for preset
> >>> contributors, it cannot be required. There are already too many
> required
> >>> technical hurdles for non-developer contributors; we certainly can't
> >>> afford to add another.
> >> The XML mapping format is inadequate and any GUI that could be designed
> >> around it would be too. It may have made sense 8 years ago when MIDI
> >> controllers designed for DJing were just starting to be made and when
> >> the M-Audio Xponent and Vestax VCI-100 were decent controllers. But now,
> >> even cheap controllers like the Mixtrack Pro 3 and Pioneer DDJ-SB2 are
> >> complex devices designed to have multiple layers of functionality (and
> >> increasingly the LEDs make use of different colors).
> >>
> >> There is no good way to support such devices without a fully featured
> >> programming language -- unless you think Traktor's maze of mouse-driven
> >> menus or VirtualDJ's hacky scripting language are good solutions. I
> >> suspect extending the XML mapping format would bring us closer to those
> >> messes. I think the best way to move forward with mappings is to make it
> >> easier to use JavaScript well.
> >>
> >> Being free software and the only digital DJ program for GNU/Linux that
> >> can do anything more than vinyl control, Mixxx attracts plenty of
> >> technically-inclined people who could program mappings in JavaScript.
> >> I'm thinking of people like myself who have programmed a few little
> >> things here and there, may or may not have any formal training in
> >> programming, and probably don't know JavaScript well or at all. These
> >> are generally the people who have been making mappings. As evidence,
> >> consider that many mappings were previously scattered around the web in
> >> their own GitHub repository (that wasn't forked from mixxxdj/mixxx) or
> >> on a personal blog. For this group, learning the basics of git is not
> >> too much to ask. The mapping documentation should be written for this
> >> audience. It should go out of the way to explain some basic things about
> >> JavaScript and good coding practices rather than assuming the reader
> >> understands the example code with minimal explanation.
> >>
> >>> Remember that Mixxx users are our customers and this (like all
> >>> user-facing items) is a customer service issue. To that end, what the
> >>> average user needs most is a GUI preset creation system that can map
> >>> everything, jog wheels included. (I have plans to make this actually
> >>> happen even for HID controllers.) They are then free to share these
> with
> >>> us any way they are able, be it Github PR, forum post, E-mail
> >>> attachment, SD card on a carrier pigeon, etc.
> >> As explained above, we shouldn't expect good mappings without coding.
> >> There is a place for incomplete mappings. That place is the forum. But
> >> those mappings should not be included in Mixxx. No matter what
> >> disclaimers we say, if someone buys a controller and tries a mapping
> >> included in Mixxx but it doesn't really work, or it works in an awkward
> >> way, that reflects poorly on Mixxx.
> >>
> >> On that note, "Community Supported Mappings" has been a misleading term.
> >> A mapping author continuing to staying involved in the Mixxx community
> >> and supporting the mapping has been more the exception than the rule.
> >> Keeping mappers involved is also important for getting input on how to
> >> improve the mapping system. I don't think requiring mappings to go
> >> through code review on a GitHub pull request will totally solve this
> >> issue, but I think it is a step towards keeping mappers engaged. Towards
> >> that end, I just put a note on the Contributing Mappings wiki page
> >> encouraging mappers to join the mailing list.
> >>
> >>> Again, thank you for your time and interest in improving Mixxx. We
> >>> appreciate the occasional whip-crack to keep the documentation up to
> >>> stuff. (It's important for fellow developers as well as users!)
> >> Thanks for making the only mapping system for DJ software that uses an
> >> actual programming language. :)
> >>
> >>
> ------------------------------------------------------------------------------
> >> _______________________________________________
> >> Get Mixxx, the #1 Free MP3 DJ Mixing software Today
> >> http://mixxx.org
> >>
> >>
> >> Mixxx-devel mailing list
> >> Mix...@li...
> >> https://lists.sourceforge.net/lists/listinfo/mixxx-devel
> >>
> >
> ------------------------------------------------------------------------------
> > _______________________________________________
> > Get Mixxx, the #1 Free MP3 DJ Mixing software Today
> > http://mixxx.org
> >
> >
> > Mixxx-devel mailing list
> > Mix...@li...
> > https://lists.sourceforge.net/lists/listinfo/mixxx-devel
>
>
>
> ------------------------------------------------------------------------------
> Go from Idea to Many App Stores Faster with Intel(R) XDK
> Give your users amazing mobile app experiences with Intel(R) XDK.
> Use one codebase in this all-in-one HTML5 development environment.
> Design, debug & build mobile apps & 2D/3D high-impact games for multiple
> OSs.
> http://pubads.g.doubleclick.net/gampad/clk?id=254741551&iu=/4140
> _______________________________________________
> Get Mixxx, the #1 Free MP3 DJ Mixing software Today
> http://mixxx.org
>
>
> Mixxx-devel mailing list
> Mix...@li...
> https://lists.sourceforge.net/lists/listinfo/mixxx-devel
>