Folks are asking where you are?
I've wondered the same and hope to see David Korn back in action again. Thread's been quiet. Did some reasearch earlier and David Korn was hired by Google after departing from AT&T (http://www.unix.com/what-is-on-your-mind-/237119-david-korn-glenn-fowler-laid-off.html). No idea what's he's working on but his profile is:
And some chat about this is at:
I haven't seen much progress at all on KSH in the last 3 years to be honest. But I did see a general industry wide movement away from the traditional UNIX systems over to Linux where BASH, Python, NodeJS(Yes even for scripting on the Linux command line) is preferred over KSH. (This is a bit odd given that BASH is derived from KSH, if I'm not mistaken.)
Seems like the biggest factor in determining these changes in 2013 was IBM's, Oracle's, HP's lack of any major announcements for the AIX, Solaris and HP-UX platforms that was being used as the last remaining major UNIX versions. For example, there was supposed to be AIX 8.1 but they downgraded that to AIX 7.2 with a future release of AIX seemingly planned for 2019 which likely may be more Linux based then anything else. Perhaps the industry is interpreting the lack of updates to the UNIX platforms as 'The End' and so anything tightly coupled with it like KSH is considered deprecated?
In the least, if KSH is on it's way out as a mainstream scripting language, I would have expected KSH to move towards the newer realms of development areas such as in Configuration Management and the likes of Salt, Ansible, Puppet, Chef etc. We see alot of new languages sprout up that are about 20% as powerful or mature as KSH is but are all DSL's (Domain Specific Languages) so no one knows these languages nor has anyone considered the limitations of these languages yet nor portability of transfer ability of these languages to other DSL's. Vendors are very reluctant to make their DSL's cross platform as well, and for good reason.
While the industry is struggling a bit with a shortage of folks able to write code for Salt, Ansible, Puppet, Chef etc. a significant number of scripters stand by the sidelines with proven languages. The templating nature of these DSL's is supposed to make them better at transferring the knowledge over to new folks that join, which is true (less of a learning curve then the equivalent 500 line traditional shell script written by an expert) but at the same time the DSL's are leaving behind infrastructures that no one is able to support and code that has to be rewritten when they leave or code that is simply a mish-mash of the DSL's and traditional languages like Bash, Perl, Python, KSH etc.
Perl popularity isn't what I would have expected as well with Python moving up quite quickly.
Going back to my earlier comment, if KSH was given templates for common tasks across any flavor such as changing an IP, adding a DHCP, managing a config file or config entry etc. across these platforms it would certainly help here, if there's still apetite for that. Also having one platform, Linux, vs 4-5 makes automation way easier: I would not need to work with the lowest common denominator to support everything.
All in all, hope David has a few surprises left up his sleeve. :)
My 5c ….