XML and JSON in 2026
The best thing about long-lived incumbent technologies like JSON and XML is that nobody really has to think about them much any more. Except for, I do occasionally, because while I’m not the inventor of either, my name’s on the front of both official specifications. Hey, it’s JSON’s 25th birthday, what a run! And what ever happened to XML? Let’s shake off the dust and have a look.
JSONiana · RFC 8259 is now nine years old and, like all the RFCs, is immutable. And, as is usually the case, a list of errata has built up over the years.
Until a few days ago, many of the errata apparently hadn’t ever been looked at for a period measured in years. Now they’ve all been rejected or accepted. Despite a couple having been marked “Held for Document Update”, nobody is interested in writing a superseding RFC. There are already enough other JSON specs [1][2] but fortunately they all say the same thing.
Which is to say, JSON is what it is and will never be improved or changed in any way. Among other things, there are literally billions of instances of JSON-reading software out there, most of them embedded in dumb low-rent devices that will never be updated.
Granted, it’s irritating that JSON doesn’t have comments (ProTip: Add a “comment” field to your messages) and makes it hard to get the commas right and doesn’t distinguish between the different flavors of numbers and doesn’t have date/time literals and allows junk Unicode. Not gonna be fixed. Which is OK because empirically, it’s good enough. Probably a few megabytes of JSON will have flowed back and forth between your phone or computer and the Net while you’ve been reading this.
Of course, there’s YAML and TOML and CBOR and Thrift and Avro and Protobufs and Markdown and more. Maybe for your app one of them is a better choice than JSON.
Oh wait, I forgot, there is a new thing: Work is under way to write an RFC specifying JSON Schema, which is quite widely used but not well specified. Good luck to the people working on that; I’m not one of them.
The best thing about JSON is nobody really has to think about it any more.
XMLitude · Last month, on the “xml-dev” mailing list, Elliotte Rusty Harold remarked, on the subject of XML generally: “Count me as one of the people who thinks it’s mostly obsolete and ultimately a failed experiment. People don’t want or need markup that’s designed to make documents easier for computers to read but harder for humans to write.”
I replied and here’s an expanded version of what I wrote:
Irrespective of the current uptake, and seen as an experiment, XML has been a success. It proved that:
You can have a data interchange format that is radically independent of your computer architecture, operating system, programming language, and application.
The only sane text standard for modern computing is Unicode, which in practice is affordable and reasonably straightforward to use.
Prior to 1996, neither of these things were widely believed. The only “interoperable” data format was ASN.1, which is horrible and lacked quality software support. The resistance to Unicode was significant and widespread, and adoption was disappointing. Today, #1 and #2 above are the (low) bar to entry for any data packaging technology.
As for current use, I guess “office" document formats are XML for the long haul [3] [4], but relatively few developers ever have to look inside them (thank goodness). XML remains a de-facto standard for text-oriented humanities computing [5], and for legislative data processing [6][7][8]. At one point it dominated things like aircraft maintenance manuals, don’t know if that’s still true. RSS and Atom aren’t what they once were, but are far from gone; they’re how I drive my own personal news-reading. Then of course there’s EPUB; do you read books on screens? And are XBRL and UBL still things?
It is true that there are few-to-no new applications that I know of that have much reliance on XML.
Eh, it’s OK, it had a good run and moved the needle. It’ll keep a few folks employed for the foreseeable future.
Like JSON, the best thing about XML is nobody has to think about it any more. Oops, if you got here I guess you just did. Sorry bout that.
Comment feed for ongoing:
From: Yannis (Jun 02 2026, at 10:43)
Great post, I've long wanted your thoughts on this actually and good to see you are at peace with XML's relative decline. I do think it was misapplied, and that an edge-labelled graph (which JSON is closer to) is a better representation for most things in the world.
I would love to see the post-mortem on the xml-dev mailing list if you would be able to link that. Couldn't find it from a web search.
[link]
From: Gareth (Jun 02 2026, at 14:05)
For the record, XML is still alive and well in a number of industry sectors beyond legislation. Aerospace & defence documentation uses S1000D, technical writers use DITA, scholarly publishers use JATS and BITS, and codes & standards publishers use NISO STS. Of course peak XML is long past, but there is no replacement on the horizon for these use cases. Whatever comes next needs mature validation, query, and transformation tools and technologies coupled with the ability to handle rich text. And some unique benefits beyond what XML already offers.
[link]
From: Curtis Pew (Jun 02 2026, at 15:20)
The 3MF file format used in 3D printing (https://en.wikipedia.org/wiki/3D_Manufacturing_Format) is, like EPUB, a zip container where many of the interior files are XML documents, including the "model" files that describe the mesh that defines the object(s) to be printed. XML was hyped by many in the industry as a panacea for data exchange, and it clearly failed at that, but it's still quite useful and has its place.
[link]
From: Jon Bosak (Jun 02 2026, at 16:23)
We were congratulated for bringing XML in under the radar, but UBL wears the stealth technology crown. Unbeknown to just about everyone in the high tech world, UBL (ISO/IEC 19845:2015) is slowly and silently becoming just what it was intended to be: the universal business language. It is already the data format used for all public procurement in over a dozen countries, most of them in Europe, but also including Singapore, Australia, and New Zealand. All large taxpaying businesses in India are required to use invoices, export invoices, credit notes, and debit notes based on the UBL schemas. UBL is mandated for tax purposes in Japan, Turkey, Colombia, Peru, Panama, Nigeria, and (by 2028) in Spain, and it is the single recommended format for electronic business documents used by members of the Business Payments Coalition, which includes many banks, major institutions, and global corporations.
So yeah, UBL is still a thing. But you can just ignore it.
[link]
From: Tony (Jun 02 2026, at 18:50)
JSON superceded COM, so that's one good thing it did. (I still use it for client-server comms)
The last full time job I had involved setting up a railway signalling network.
It used XML to describe the various components: track sections, signals, and other stuff.
The project ultimately folded, but not due to the use of XML.
[link]
From: Andreas Jung (Jun 02 2026, at 20:54)
The problem is not XML but the way how it has been mis-used in various ways - for complex configuration languages, for complex, hard-to-manage DSLs and so on. At some point in the 90s, XML turned into the Everything-must-be-XML religion - a wide-spread misbelief. Then the failed adaptation of XML in the browser environment. Ok, there are a few people still using XForms in the browser or XSLT inside the browser (which is completely insane in my opinion...yes, they can do it).
POV: XML is a perfect data format for content and data but it sucks completely at the point when we use XML for configurations, languages, business logic.
Regarding JSON: not a perfect format but things like validations, transformations can also be done with any language and any popular toolchain by junior developers. XML does not play any major role in AI these days, JSON does.
Let's keep XML for data and content but stop misusing it for any other purpose that is not manageable and maintainable by humans and mortals rather than by #XML gods and their apostles from the #XML parallel universe.
[link]
From: Tim (but not THE Tim) (Jun 02 2026, at 22:31)
Related: I still find XSLT useful, which was brought to use by XML. I can convert XML documents of course, but I've also used it to for other things such as converting iCalendar .ics files to xCalendar
[link]
From: Ulf (Jun 03 2026, at 02:49)
The worst thing about XML was that too many people assumed it should be human-editable, just because it is human-readible. My memories of Struts config files, Ant build files and Java's web.xml are not too fond.
[link]
From: Nathan (Jun 03 2026, at 10:34)
It seems trendy these days to hate on XML or to talk glibly about its decline and / or death without consideration of the major changes that it brought about. I'm just barely old enough to remember when XML started to become A Thing, and it was a revelation.
Human-readable? Glorious, in a world of hand-rolled binary formats or cumbersome serialization frameworks that broke constantly. Unicode? Meh, who cares, latin-1 is good enough for everybody...until I wrote a post title in Japanese and the RSS feed Just Worked and it didn't occur to me until much later how shocking that was. A public standard that's free to use in the midst of proprietary, corporate mystery meat? Refreshing.
It's like people hating on MapQuest and its printed directions for being old and clunky and obsolete because they've never had to pull a folded map out of a glove box to figure out how to get somewhere.
XML was the right technology for the right time and the world is better for it. Thank you for the part you played.
[link]
From: Kira (Jun 03 2026, at 19:57)
Microsoft has just recently migrated one of the project organization file formats in Visual Studio to XML:
https://devblogs.microsoft.com/visualstudio/new-simpler-solution-file-format/#why-did-you-choose-xml-for-.slnx
So it’s still getting some new adoptions here in the 2020s.
Also, I recently tweaked an SVG by hand because I didn’t feel like downloading software just to un-squash some paths. I greatly appreciated that it was human-parseable.
[link]
From: Alan Painter (Jun 04 2026, at 01:15)
I feel obligated to offer an additional "take" on the matter of the perceived demise of XML which, as Mark Twain famously wrote, is "greatly exaggerated" both herein and elsewhere.
The ongoing and, indeed, growing ubiquity of XML has been largely understated. As mentioned here, important fields are dominated by XML, with financial, health, publication, aeronautics, digital humanities, administrative and governmental being but a small subset of those many fields reliant upon XML descriptions of interactions and exchanges.
These successes attest to the maturity and solidity of the XML Technology offering. They demonstrate that XML vocabularies, described by schema representations, especially but not limited to the XML Schema Description Language, are solid components for information interchange. XML databases contain these language-independent documents without requiring ORM (Object Relational Model) interfaces that make up the thickest and often brittlest layers of applications that use relational databases for persistence.
Furthermore, the XML technologies that power these infrastructures, notably XQuery/XPath/XSLT, are vibrantly alive, with implementation updates arriving regularly from several different projects, often open-source or at least openly available. New versions of the standards are in the works for sometime soon. The most recent standards versions, 3.0 and 3.1 from 2017, are quite successful in their implementations and adoptions. The "XML community" is very strong. Te implementations are strong.
Admittedly, the widespread conviction is that XML is "passé" and even some of its earliest proponents are today prepared to read the last rites to XML. I personally can see this as nothing more than the creeping kakistocracy, powered by AI, that has overrun the technology landscape. The XML alternatives are "successful" in that they have captured the attention of the masses without, by any stretch of the imagination, achieving the solidity and reliability and, especially, the reusability of the XML technology stacks.
In this fashion, the supposed demise of XML is the quite simply the demise of sound principle and practice, replacing it with brittleness and obscurity.
Attending this 2026 edition of XML Prague, I'm happy to be among those who still see the power, even the beauty, of intelligent, tractable implementations, attempting to further their reaches and continuing to stave off the onslaught of obscurity.
[link]
From: Rick Jelliffe (Jun 05 2026, at 19:44)
XBRL is going strong: see https://www.xbrl.org/the-standard/why/xbrl-project-directory/
I have not heard of any regulating adopters ditching it.
It has two main forms: XRBL which is XML, and iXBRL which is XHTML. (Its "taxonomies" are defined in a layer superimposed on top of XML Schemas, which may be surprising proof that XML Schemas had not quite reached the limit of hair-pullingness!)
[link]
From: Rinie Kervel (Jun 06 2026, at 02:49)
I am still trying to figure out the right words. But I feel that every text that cannot be rendered to pages, lines and files but insist on chapters, paragraph and sentences ignores what Gutenberg did.
A book is also a set of pages with and index. XML tied the envelope to the letter wheres html5 makes the header optional to the body.
URL has DNS as a resolver so the location of the resource may evolve.
My clumsy attempt at
https://rinie.github.io/2026/06/04/postman-reads-envelope/
[link]
From: Len (Jun 07 2026, at 16:54)
Works well for MCCF. The AI loves. LLMs need document models. Using json for MCP which is fine . Lifecycle is still the determinant. Markdown sucks and Python would be safer if it hadn’t followed that dismal indented road. I don’t mind as long as I don’t have to edit it. XML is easy.
Claude to Blender via MCP is a marvelous time saver. The AI revolution is irresistible.
[link]
From: Shiva Velmurugan (Jun 08 2026, at 14:35)
I have fond memories of the few meetings we had while I was early at SNS, and you were already moving on to other things.
It's an interesting take. I wonder what your take is on https://json-structure.org/? Have you heard about it? Clemens is trying to "fix" if at all possible JSON "any", which really does force folks to write flexible but very brittle software.
[link]