I am having issues compiling the [dssi~] external on an up-to-date Arch Linux (i386) box, also on Arch Linux Arm. Dependencies listed in Readme are all satisfied.

Code:
cc: error: suppress: No such file or directory
cc: error: unrecognized command line option ‘-bundle’
cc: error: unrecognized command line option ‘-flat_namespace’
makefile:48: recipe for target 'src/dssi~.pd_darwin' failed

The Makefile for dssi~ is unlike the Makefiles for [hid] and [plugin~].

dssi~ seems perhaps deprecated. Did [pluginhost~] actually materialize? Is there a preferred object to host dssi plugins?
`]

The makefiles for [dssi~] and [pluginhost~] both seem a little messy for Linux compilation. For now I'm returning to ghostess for dssi hosting. As I have time I'll tinker with creating a new makefile for [dssi~]. This is admittedly of little interest to most users, but internal dssi hosting is a worthy option so I'll update here if I sort it out. Thanks everyone who's taken the time to view this.

internal dssi hosting is a worthy option so I'll update here if I sort it out

That would be highly appreciated, thanks for tackling this!

After some trial and error I was able to get dssi 0.96 to compile and run on my Linux machine (Lubuntu 15.10). Here are the changes I made to the files:

• In the Makefile: changed "LIBDIR=/usr/local/lib" to "LIBDIR=/usr/lib"
• In the Makefile: changed "current: pd_darwin" to "current: pd_linux"
• in dssi~.c: added "#include pthread.h" on line 31

The first change might be optional, depending on how your filesystem is set up.

The help patch and output~ abstraction in the doc folder also needed a few changes. (The first thing I did was to cut the connection from the sine plugin to throw~, in Instance #2.) I got Hexter to load by changing the dssi~ object in Instance #1 to "dssi~ hexter"; it didn't seem to need a path. And it needed to be connected to throw~.

After that it seemed to work fine. Selecting channel 1 (it's at the bottom right of Instance #1) got sounds out of Hexter.

A few things to look out for -- clicking on "reset 1" or "dssi show 2", or trying to create a plug~ object, crashed Pd. And I had some occasional problems with Pd hanging until I killed the hexter_gtk process, usually with that big help file. A fresh patch with a dssi~ object doesn't seem to crash as often.

I poked through the dssi~ C file and saw something like a "deactivate plugin" signal you could send to it, but can't figure out how to make it work.

Anyway - hope this works for other folks. Have fun!