You said it -- a VS65 is not a yardstick of performance. The point, though, is that there is evidence that there was a U.S. mil spec ADA compiler, certified by the Wright-Patterson Air Force Base certification center, and it may be that there are still U.S. goverment projects in which ADA is required. Now that the VS intellectual property is once again in the hands of a U.S. company and not a "foreign" company, government projects may be possible.
At the moment no one knows where the ADA compiler is. No one knows who developed it, nor where the source or object may be. I'd like to get a handle on it.
If the VS ADA compiler had been developed in the normal manner in the normal course of events, the people in Tewksbury would know of it and would have the source and object. At the moment it appears they do not. That suggests that the VS ADA compiler was not developed in the usual manner, or by the usual people. I would like to hear from anyone who has any knowledge of it, to put the pieces of this story together and possibly gain access to the compiler. It may open some new doors to new business for the New VS.
Can't help you on the ADA Compiler, however Casey Jones ( WANG VS OS Support in Australia and New Zealand ) may know of it's whereabouts/origin. I think Casey is living in New Zealand but I don't know where. Maybe one or another of the WANG VS members may have contact details for him. He may have at least an object/binary version of the ADA Compiler.
Also Mil-Spec ADA may require "The Orange Book" Mil-Spec version of the WANG VS Operating System and Tempest standard Hardware/Software ( eg Faraday external RFI shielding ).
The Tempest ( or US State Department Restricted Exports ) was filtered on a priority basis ( those Trusted Countries/Subsidiaries ) who would not on-sell Tempest standard products to black-listed countries/organisations.
This list was basically in this order of controlled purchases:
USA Inter & Intra-National Security Sites ( eg USA Embassies )
USA Secure Commercial Sites ( eg Lockheed, Boeing, Raytheon )
USA International Allies ( CANAUSUK - Canada, Australia, USA & United Kingdom )
USA Military Treaties ( NATO & ANZUS Countries )
USA Restricted Military Agreements ( Israel, SA, Japan, South Korea and SEATO )
Non-Warsaw Pact European Countries ( eg Switzerland )
Obviously since the warming of the "Cold War" the whole geo-politics has changed and the USA State Department would have to be consulted on the current rules of play.
I have been associated with Casey Jones since late 1999. I worked with him (and still do, to some extent) on reviving migrations to COBOL ReSource. I have written to Casey asking him about the VS ADA compiler.
Succinctly put, dried up creek this end too. My only other thought is the old US Federal Systems Division (FSD) in Washington, DC. I don't even know of they exist any more and I don't know what oversight department ( State, Interior or Treasury??? ) they came under at that time.
I will wait till Howard Mackrory returns after 4th July ( your timezone ), he may have some history of ADA's status/whereabouts/origins etc. The Countess of Lovelace is of greater mystery in life as she is in death.
I tracked this down, to a degree, this afternoon. VS Ada is a port of Alsys Ada. Aonix, the current incarnation of Alsys, has confirmed that they were not involved in the project beyond providing the source to Wang. Compucom will continue to look into this.
I think WANG Sales didn't fully understandard that ANSI/ISO Compiler Standards had a cultural as well as commercial profiles. I was told in days of yore, that WLI had Engineering and Sales and didn't have or comprehend Marketing. Hence if there was a large enough customer who wanted something ( eg ADA, Ada ) they got it, irrespective of overall integration relevance. WLI was obsessed with proprietary ownership of its products and consequentally made strange, if not commercially damaging decisions.
WLI had products not markets, some were great eg WP, WANG VS, others were awful eg ADMS, PACE, WANGNET.
The big problem that I think I've seen was that while Wang was an IBM wannabe and did a lot of very good development across a lot of product lines, they never revisited some of them to keep them current or improve them. Wangnet eventually died because it remained at the level of analog CATV networking. The 3480 tape drive became a white elephant because it was never revisited, while IBM and others advanced through 3480a, b, c, d, e, f, g, etc., to 3490, 3490 a, b, c, d, e, etc., and capacities like 20 GB per tape while Wang's 3480 remained at 150 or 250 MB per tape. 802.3 and TCP/IP were poorly done to begin with, TCP/IP ported by outsiders, and were never revisited to fix them in any major way. VS TCP/IP is an utter pig, burdening the VS CPU with all the stack overhead and has limited connecitivity making it impossible to use for any large-scale TCP/IP applications like a public web server. Workstation microcode loadability, originally conceived to offload work from the CPU, devolved to being just a mechanism to preserve proprietary control. Microcode eventually became a bad word at Wang while in other areas of engineering people were mastering and doing marvelous things with microcode loadability.
There are quite a lot of us, though, who love PACE. PACE 2 was ahead of its time when it was released and remained ahead for quite a long time. Oracle didn't have a shred of referential integrity integrated into their DD until at least V7 or so... before that, referential integrity in Oracle was a bad joke made up of hundreds or thousands of triggers impossible to maintain. The only way to have referential integrity in Oracle was to use their CASE tool, which generated the SQL*Forms triggers, and never touch the resulting generated forms code, always going back to the CASE tool to make changes. Meanwhile, PACE had bulletproof referential integrity from the get-go.
Connectivity was another area that lacked attention, but Wang wasn't the only one to have made that mistake. As PCs became more of a force, none of the major computer vendors understood that the PC was a ready-made co-processor they could have co-opted and used to expand the capabilities of their systems. All they did, Wang included, was grudgingly develop limited workstation emulation to allow PCs to pretend to be workstations. It was mostly third parties like Lightspeed who did the major work and integrated file transfer into their workstation emulators and advanced to TCP/IP connectivity as Novell's IPX/SPX fell out of favor. But the promise of a PC being a robust, microcode-loadable co-processor never materialized. Computer vendors could have treated the PC as a very useful device entirely engineered and funded by third parties, and could have co-opted the PC to integrate it as a co-processor entirely loaded by and controlled by the VS. That could have led to a whole world of amazing functionalities that could have vastly expanded the usefulness of the mainframes. As far as I know, no major vendor did that, and the only good workstation integration I ever saw was on a DEC VAX, done by a third party in the late 1970s or early 1980s.
As large as Wang once was (about 35,000 employees), they never showed the depth and breadth necessary to effectively compete with IBM. Wang developed products such as those I mention above but didn't keep them alive as ongoing, advancing products. Many were more like one-shot developments, ignoring the fact that nothing technological remains the same, no solutions serve forever, etc. To remain relevant, a product has to continue to be active and under development.
What we're trying to make work now is the fundamental nature of the VS in the form of the New VS... that it provides the best data processing environment ever developed, usable with much more modest effort and staffing than any IBM mainframe system, and with much more rapid development than systems not built from the ground up for data processing, such as PCs using Microsoft tools or Unix tools. The VS, like IBM mainframes, has a data processing instruction set and many other features that lend themselves to efficient development and execution of data processing functions. The business of business remains, after all, characters, and the VS processes characters very efficiently and gives us tools that allow us to build applications effectively and efficiently.
The long-term survival of the New VS will depend on our ability to market and sell it to people who have never had a VS of any kind before. ADA is just one small piece of that puzzle, a very small piece that could open the door to some government work here. More important would be packages that have fallen out of use but could be revived. It's still the case that good packages can drive platform sales. We'd also like to figure out ways to interest up and coming software development shops in going the VS data processing route. We just haven't yet figured out how to do that, and marketing funds are scarce.