The tech thing that annoys me most frequently is when people insist JSON is better than XML. So much breath, so much time, is wasted by those setting technology standards on this non-issue!

XML is no harder to parse than JSON, whether you're writing your own parser or reusing something from your language's standard library. Especially when you add the schema validation & namespacing that standards organizations rely on.

I wouldn't be so annoyed by this if a significant chunk of every meeting wasn't dedicated to the supposed benefits of JSON. Which usually rings hollow for me.

I don't particularly care about whether we use JSON, XML, or whatever. Talk about what aspects of JSON you think reads nicer, sure.

But personally I'd find it just as easy to support both! Don't presume "developers prefer JSON", I don't care. Can we move onto the datamodelling please?



after reading about and the idea it could be represented in xml / json / turtle notation / anything,

I feel like the real objective should be making schemas that validate the same data structures in any format, and treating those schemas/vocabularies as the actual data format

@Valenoern @alcinnz

It's a nice idea, but my concern is that there's no coherent way to talk about 'actual data format'.

Even in a simple computer program, the same data is encoded differently at the physical level, the database level, the object level in application code, and the transport or syndication level. Once we start talking about the wider world of different database types and programming languages, this variability increases still. Which is the 'actual data format'? None!

Sign in to participate in the conversation

For people who care about, support, or build Free, Libre, and Open Source Software (FLOSS).