On Tuesday 06 February 2007 14:04, Anant Narayanan wrote:
And how does that become a standard?
What we have here is our own interpretations of the word "standard". If I were to make a fictitious language of my own; and if I would write documentation describing the semantics of that language; that's a standard. It may not be popular in the sense of other people adopting it; but any compiler that conforms to my semantics automatically conforms to my "standard".
We are talking of documentation of some technology which has substantial peer review and multiple implementations which can interoperate. You can write an arbitary set of rules to do something and over time peer review and multiple implementations may happen at which point you can call it a standard. Your initial efforts are not a standard but just a proposal or description with a bench mark implementation.
OOXML essentially describes how to enclose
binary blobs, while saying nothing about the blob itself, which is the center of the interoperability problem.
I never said OOXML wasn't crap. I only said that .NET is a proper standard.
I never said that .net is not a standard but that it is encumbered and hence not open. You cannot implement a patented software tech by reading the standard because the standard substantially describes the patented tech and would imply wilful violation. Standards bodies clearly state that encumbered standards are available on RAND terms - tech like the GSM specs - to be read as pay a fat sum for the privlege of not getting sued. And in the light of the microvell deal to be avoided like the plague. This was in response to your "nobody is going to sue anybody"
The ECMA .NET specification, unlike OOXML, describes in detail the language itself and provides all the necessary information to create working tools. Which is why Mono was made possible in the first place.
Agreed
Again, .NET is an ECMA standard; complete with a reference implementation. If you consider JavaScript to be a standard, there's no reason why .NET isn't.
Agreed again. Except for the patent part and Micovell deal tactitly acknowledging that patent encumberances are present in some Novell stuff. Without knowing what exactly this is your only chance is avoid unpaid Novell stuff like the plague - unless you are a great gambler or love the lawyers.
There's only one proper implementation of an XHTML 1.1 based browser. It's still a standard, is it not?
It's not. Not until someone writes an implementation as per the standards documentation.
As far as I am concerned it still is. Because I can make web pages that comply to XHTML 1.1 and validate them with W3C's validator. Whether or not the end-user will be able to view the web pages as XHTML 1.1 intended them to be viewed is not my problem.
Hope it's good for you that others view something entirely different than what u intended.