Re: What would be a truly relational operating system ?
From: paul c <toledobythesea_at_oohay.ac>
Date: Sat, 14 Nov 2009 03:51:30 GMT
Message-ID: <6PpLm.51936$Db2.40145_at_edtnps83>
> It is reasonnable to assume that not only a driver needs to be
> developped but also a filesystem that represents the relation. As the
> result of the last 10 years researchm I do have a promising filesystem
> but I am studying the opportunity of building on the top of a driver
> as opposed to the host OS.
Date: Sat, 14 Nov 2009 03:51:30 GMT
Message-ID: <6PpLm.51936$Db2.40145_at_edtnps83>
Cimode wrote:
> On 11 nov, 05:39, paul c <toledobythe..._at_oohay.ac> wrote:
>> Cimode wrote:
...
>> plus de détails »
> It is reasonnable to assume that not only a driver needs to be
> developped but also a filesystem that represents the relation. As the
> result of the last 10 years researchm I do have a promising filesystem
> but I am studying the opportunity of building on the top of a driver
> as opposed to the host OS.
If I understand what you mean, I have to say that it would be a mug's game to try that because you'd have to interface to the many arcane aspects of the typical OS's internal interfaces. Maybe I have mis-interpreted but if not, I'd say that it is more important to map to various typical filesystem interfaces. I realize this is another layer, but to me that's really the name of the game. One of the Xerox Parc notables remarked something to do with recursion in programming concepts being omni-present and I think similar is true of indirection. Received on Sat Nov 14 2009 - 04:51:30 CET