[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Standards & RFCs
Oleg Goldshmidt wrote:
> Yuval,
>
>
> Yes, you can write code that is not standard. You can write C code
> that will compile under gcc but not with other compilers. Anyone who
> doesn't use gcc won't be able to use the code. It is widely known (and
> resented) that you can create a "Netscape-enhanced" web page that won't
> confirm to the http standard and anyone who uses another browser won't
> be able to see it. In those cases, not conforming to standards is
> mostly your problem. If you write a piece of code that sends silly
> emails to anybody who doesn't put [Nu] in the subjects of his emails,
> you create problems to a lot of people.
>
When I was complaining of Netscape (and the same by the way applies to the
Explorer), I wasn't complaining of it accepting non-standard pages, I was
speaking of its own conformance to the standards in the RFC. Check out the
definition of User agent in the standard as opposed to what the communicator
sends (different character set altogether), check the way it encodes user input
from forms as opposed to the HTML standard (not all characters which are
supposed to be encoded are encoded in fact). I'll ignore the cookies for which
Netscape has a different standard altogether (which apparently everyone use
instead of the one proposed in the RFC). I've got nothing against it, as the
only true standards are the de facto standards. BTW, I was lying, I do have
something against it as I had to patch good programs which did conform to the
standards, but that's the real world.
--
Yuval El-Hanany | Kawasaki GPZ500 '97 |
qwmiwn@geocities.com | | Have backpack,
Home : 972-3-5243488 | Debian Linux Inside | will travel
Work : 972-9-9586077-12| |