On 16 July 2018 at 10:22, cyg Simple <[email protected]> wrote: > On 7/16/2018 10:02 AM, Stephen John Smoogen wrote: > >> >> Most of the rsh usage is going to be legacy hardware and systems which >> various places still have in good numbers. Various industrial and lab >> components might have been built in 1995 and is slower than your >> iphone but the replacement costs tens or hundreds of millions of >> dollars... (and still uses rsh for backwards compatibility). Payroll >> systems in other places use rsh and rcp and cost large amounts to >> 'upgrade'. The people running these don't show up mailing lists >> because they may not even know that the system uses rsh/telnet or some >> other obscure thing.. they just run a script on a Windows desktop that >> someone wrote years ago. They only show up when stuff stops working. >> > > But are those scripts Cygwin? I doubt they are. Do you have proof of any? >
I don't have any recent proof. Most of mine is 6-8 years old so I can't say if they are still in production. I am extrapolating of what people ask on IRC and mailing lists at times and going by experience of where that software would have been used and why it takes forever to get rid of it. [I don't like rsh/rlogin/rcp and would prefer it was gone. I just have found that fighting that battle has a very very long tail of "oh we still use that... yeah we are running a RSX11 controller now on an ARM chip but its embedded software only talks rsh so lets put an SSH proxy on this windows box and have it rsh to it." ] I also don't have any recent proof that 90-95% of the software in the Cygwin distribution is used anywhere. I just have hearsay that people use X and various network servers on it.. so if my proof is needed to keep stuff in.. I would expect a lot smaller distribution soon. > -- > cyg Simple -- Stephen J Smoogen.
