[Info-vax] Free Pascal for VMS ?

[Stephen Hoffman]

On 2018-05-11 00:39:01 +0000, seasoned_geek said:

> On Thursday, May 10, 2018 at 10:54:39 AM UTC-5, Stephen Hoffman wrote:
>> 
>> We'll still be dealing with C and Microsoft Windows and Unix/Linux in 
>> twenty years.
> 
> We will be dealing with C for a very long time. Windows won't make it 
> another 10 years, let alone 20. Just like Apple doesn't make its own 
> OS, putting a GUI desktop on top of BSD, Windows will be a GUI front 
> end on top of Cannonical's Ubuntu.
> If Linux is still around, it won't be anything you recognize. The Linux 
> philosophy of complete anarchy with everyone rolling their own 
> transport layer security and opening whatever ports they want means 
> security is physically impossible to achieve. It will be forced to have 
> a TCP/IP software appliance and disable the existing network libraries 
> after a few more rashes of high profile security breaches.

You really don't like the current market and the current reality, do 
you?   Okay, Windows will be around for the foreseeable future as a 
billion systems don't go away very quickly — DEC was telling folks to 
get off of OpenVMS most of thirty years ago (1995) and here we are — 
macOS and iOS and tvOS is based on mach and parts of BSD and 
Apple-specific code and is a unique open-source kernel known as XNU, 
Microsoft can only get as far with porting and upgrades and API changes 
as they do because they're serving up OO APIs for most applications, 
there are far better kernels around if Microsoft is going to expend 
huge time and effort and disruption involved with a kernel transplant 
and for little apparent gain, TLS is not going to be supplanted by 
IPsec until the latter becomes far more widespread and until IPsec 
deals with the current and near-term Internet network and app 
implementations, and app-to-app security is not going to be rushing to 
migrate to designs that trust the entire host, and appliances are going 
to have to deal with both the expectations of OO interfaces by app 
designers and that the remote ends of the connections will not be 
running on OpenVMS appliances.  Better APIs are certainly needed for 
OpenVMS and better handling of networking and vastly better network 
integration and massively upgraded security, but then VSI is not going 
to be sufficiently relevant outside of the OpenVMS installed base 
anywhere near soon enough to be remotely relevant in the next ten or 
twenty years.



-- 
Pure Personal Opinion | HoffmanLabs LLC