Skip to Content

Premiers Pas

First Steps

It has been about 7 weeks into the launch, and a lot has happened. We’ve made progress on growing our community, as well as important, encouraging, and enabling first steps in many projects.

Codecs

The zc/codec project has been quietly gaining traction. A general interface which allows pluggable sound codecs to easily interoperate with I/O and processing and core sound processors has evolved to meet demands of codec authors. Moreover, the project supports 3rd party codecs and so can be used in conjunction with codecs like AAC which are out of zc scope for licensing reasons.

Although hidden in zc/ext, we’ve grown our codec collection to support FLAC decoding and ogg/vorbis decoding. More are on the way.

Sound I/O

Sound I/O has traditionally been quite problematic for Go, and progress here has been characterised by alpha chaos. Fast forward motion and volatile code, as difficult as it is to follow, have enabled several key advances

Host Structuring

An portable code layout enabling multiple host specific implementations has been put in place, as well as a means for 3rd parties to write and distribute interfaces to host drivers independently. Although the code is still alpha and some details of the interface are still likely to change over time, the big picture is a huge step toward the goal of effective sound I/O with host only dependencies and the overall organisation of the host structuring is unlikely to change in the near future. This work comes with a new porting guide and so the project is well positioned to become a widely ported and used sound I/O package for Go.

Reliabality

Go, as well as any non-C based language, has a terrible reputation for reliability in the case of mass marketed low latency applications. This is the case even when using very seriously engineered and established projects like portaudio via cgo. ZikiChombo has pushed forward on this issue, engaging portaudio and golang developers to help clarify the root causes, and we have arrived at some proposed solutions. For example, the addition of rb is a first draft design of a means to bypass the various problems behind low latency reliability on un-dedicated hardware for sound i/o with Go.

DSP

The DSP code base has grown some documentation and clarity, and the first dsp-star labelled issue has been fixed, giving zc/dsp its build badge.

The DSP code base has also grown a resample package which enables monotonic continuous time views of PCM data, dynamic resampling, and control over resampling quality.

Some progress has also been made offline on the infamous half-complex issue, So we’re looking forward to that becoming a non-blocker for advancing on some higher level dsp functionality in the near future.

Mel Support

The zc/sound project also grew support for the Mel frequency scale interoperating with the frequency type. This is a simple addition that is also key to enabling higher level music dsp functionality down the road.

Licensing and Community

In the first steps of our first weeks, it has become evident that a plethora of open source licenses and associated repositories are the norm. As an old guy, I had thought things would be more readily organisable by a single shared author license. I was wrong; and ZikiChombo has made several procedural steps to build the community and enable a simple licensing solution for consumers (That means no vendering).

First, zc has arranged code so that in many cases, import direction can work both ways between collaboraters. This is a valuable asset for code organisation, allow direct code contributions and 3rd party distributions.

Second zc has created a special zc/ext module which can import both 3rd party code and the rest of zc. This is a playground which can facilitate collaboration.

Finally, zc has decided to relicense its code base under the Apache 2.0 license to work with more other projects without introducing licensing multiplicity, and to afford some standard open source patent protection as well.

Conclusion

Our first steps are indeed exciting. We’ve got a long way to go, and a lot of work to do. But major progress has indeed been made on many fronts.

Thanks to all the ZikiChombo members, collaborators, discussions and the general culture of mutual respect between open source projects, without which probably none of our premiers pas could have been made.