[cap-talk] Concrete application, WebCVOS

[James A. Donald]

Jed Donnelley wrote:
 > For me the primary issue is putting together
 > environments (compatibility libraries) on top of the
 > WebCVOS that are sufficiently the same as the native
 > Unix[/Mac?] and Windows environments so that
 > application developers - perhaps with very minor
 > twists - can make their codes available to run under
 > WebCVOS.

Sounds like an excellent plan, but for Microsoft
software, probably requires cooperation from Microsoft.

In the old days, Microsoft probably would have
cooperated.  In more recent days, would have cooperated
in order to steal your idea, and if your idea had any
success, would immediately produce a competing product
based in substantial part upon the code samples and
design discussions you had with them, as happened with
file compression software.  These days, probably would
react with paranoid suspicion and threaten to sue you.

The necessary hooking seems comparable with that
required for on the fly file compression software, and
outsiders managed to do that, though with considerable
(and duplicitous) assistance from Microsoft.

I have often advocated a completely incompatible VMed
environment, running on top of both linux and and
Windows, but this accomplishes much the same effect as
memory safe interpreted capabilities languages, and we
already have memory safe interpreted capabilities
languages, which do not seem to be setting the world on
fire.  People do not seem to be lining up to rewrite
their stuff in such languages, so I have sadly come to
the conclusion that neither would they rewrite their
stuff for a safe environment.