Re: [xml-dev] How long before services sending/receiving XML might need

[Norman Gray <norman.gray@glasgow.ac.uk>]

Steve, hello.

On 14 Nov 2021, at 12:43, Stephen D Green wrote:

> But if you try creating some XML with .NET
>  libraries then simply changing from UTF 8 to UTF 16 you should realise how
>  poorly these libraries work and how many ridiculous hoops you have to jump
>  through to avoid memory leaks. It is simply a matter of curtailed
>  investment, still stuck with original DOM-based processing.

Right, I see what you mean, I think.  I've been in that general situation before, but to me, it just sounds like some rather bad tooling (I'm partly echoing Ihe's response here).  I'd have hoped that the XML-wrangling and the serialisation would be sufficiently orthogonal that the XML-wrangling (which doesn't care about encoding) would simply send string to the serialiser, and it would worry about encoding them into bytes.  Fiddly to set up, but not a hard problem.

> You then get to see how immature these
>  libraries are and how little investment has been made in getting them to
>  work with each other and improving them with time. They are stuck in 2005.

I think it's quite possible that the batteries-included XML support in ecosystems such as .NET hasn't had much TLC in the last decade.  It might be that that support was good enough for the 'XML for everything' attitude that was regrettably common around then, but that anyone doing XML 'seriously' would either use (commercial?) third-party frameworks or just learn to live with it, and that this has meant that that batteries-included support has never really got better.  I make this remark with some diffidence: I'm not sure how true it is.

So I probably agree with you, Steve, to the extent that wrangling XML does still feel rather clumsy in most languages.  But I don't think it follows that XML is past its sell-by date (which I think is what you're basically suggesting?), since the sort of problems where XML is the right answer (which do not obviously include cases where the goal is sending very simple messages about), are ones where a bit of clumsiness in the serialisation is lost in the noise.

Parenthetically, it occurs to me that 'encodings' is becoming less of a headache than it used to be, simply because UTF-8 is becoming more ubiquitous.

Best wishes,

Norman


-- 
Norman Gray  :  https://nxg.me.uk
SUPA School of Physics and Astronomy, University of Glasgow, UK