There are obviously much fast formats for that type of thing, but they have their own drawbacks which are significant. Protocol Buffers are crazy fast but aren't self describing or human readable. JSON is really just a different representation of XML on many levels and certainly doesn't deal with binary data any more elegantly. (base64 here we come!)

If you have binary data and plenty of CPU on both ends, then gzipping XML gets you pretty close to the native size actually and can be a good solution. You keep the human readability and the compression can be done on the transport layer if it is HTTP.

If you don't have much CPU and want easy parsing, it is very easy to write a binary format for XML. I've done this for products that are sending down data to cell phones (including images). The native format is XML, because it is easy to author in templates and easy to debug, but with an extra parameter it gets thrown into a denser and very fast to parse format for the other end to deal with. This costs a bit of CPU on the server, but that generally isn't the problem.

Anyways, I would agree that the main author's points are pretty ridiculous. They either point to him not really understanding XML or never really considering the pros and cons of the alternatives. XML is great for what it is, and it has its drawbacks, but as a universal transport it is pretty rad.

Why does that matter? The fact that they aren't self-describing more or less requires that you use a published message definition, which solves one of the major problems of XML schemas -- nobody actually uses them, and many that do use them incorrectly.

XML is great for what it is, and it has its drawbacks, but as a universal transport it is pretty rad.

Having wasted far too much of my life on parsing XML documents, I simply can not agree.

XML as a universal transport just sucks compared to the modern alternatives.

Having just spent the past two days reverse engineering an API that used protocol buffers, I can tell you I would have much rather done it against XML. :)

As for parsing, I guess I don't get what the big deal is. I've long since given up on SAX parsers and occasionally use pull parsers, but for the majority of applications using a DOM parser gives you a very OO friendly approach to turning that XML into your objects, and it is dead simple. Sine the generated Protocol Buffer objects are immutable and can't be used as base classes, I suspect the amount of code to turn a 'message' to your domain objects and vice versa is very similar between protocol buffers and a DOM impl. At least it has been in my experience, neither is hard.

Why didn't you have the descriptor files? Optimizing for ease of external reverse engineering seems rather counter to the authoring organization's actual goals.

If you want third parties to be able to interoperate, give them the descriptors you use.

As for parsing, I guess I don't get what the big deal is.

Writing lots of code to read and parse data types as text, especially when no schema is available, is time consuming.

Protobuf is necessarily validated. The input types are known-correct, conversion to the domain model is direct and straight-forward, and it's exceptionally easy to generate code or otherwise automatically deserialize to native representations.

Parsing XML doesn't include lots and lots of code, at least not for me. As I said, the amount of code to parse a Proto built class into my own datastructure is almost identical to me doing it from a DOM tree save for some type conversion.

The other plus is that proto buffers involve firing up those tools and incorporating them into your project. How often do you just want something really simple, say the current temp in a zip code? That's a few lines of code to pull out via XML, without the trouble of pulling down the proto buffer definition, building the classes for your language etc..

Protocol Buffers have a text format that is equivalent to the binary format.

Also, XML isn't that much more self-describing than Protocol Buffers. Sure, you can read the tag names in XML, but what is the set of all allowed tag names? What does the stuff inside the tag names actually mean? You have to communicate that stuff somehow, out-of-band, in a spec of some sort. In Protocol Buffer land, a .proto file is just part of that spec.

You can get pedantic about needing a spec for everything, but the web isn't really built that way. Do you really track down the full docs for every API you interact with? Say you are just pulling down some information from tumblr via their API or the like? Probably not.. you just throw it in a browser a few times, pull out the fields you need and get on with it. XML lets you do that, quickly and easily and most of the time will work just fine forever more. XML is great for the same reasons HTML is great. It is fast and loose and lets you so easily introspect APIs that you can get your work done quickly.

If I was pulling some information from Tumblr, I would at least read the introductory API docs at http://www.tumblr.com/docs/api (if you didn't do that, how would you even know what URL to hit?). If they were using Protocol Buffers, they would publish their .proto file on that page. In fact that page reads very much like an informally-specified .proto file; it contains a list of all the parameters and what they all mean.

I think that's the key point that everybody misses. XML was really just intended to solve the data interoperability and transport problems, not to function as a datastore. It just so happens that once you've made some XML, or received some from somebody, you may as well just deal with it as is rather than XSLT it into your local apps proprietary format (and as it also turns out, it's fairly easy for people to write by hand and debug in simple cases).

Encoding binary data (as, say, CDATA) and then shipping or storing it in the text-based syntax is every bit the lose you suggest it is. Binary XML formats, however, can answer your objection - the main problem being that no good binary XML format is yet widely adopted.

You say "`Some day' doesn't do us any good today." I half agree. It is true that, today, if you want to use XML for large binary data you have to place your bet on a suitable binary format and you face a rocky road. XML "sucks" in that sense. On the other hand, if you can possibly - for the problem you are trying to solve - make a good bet on a binary format for XML, then it is a fine choice because the data model for XML as interchange format is excellent. So using it for binary today is not an option in many cases and has an element of risk on other cases - but longer term is a winning bet. Your benefit for considering it today is that longer term pay-off. "Go long" on XML.

Meanwhile, Protobuf and Thrift work today. They're fast, they're size efficient, and unlike XML, they're simple.

It's time to stop making excuses for XML. It's an inexcusable ball of poorly designed unjustifiable complexity created by architecture astronauts run amok.

I'm disappointed at the upvotes on a whole page of platitudes on why XML isn't a bad format, it's just misunderstood.

XML is horrible for binary data, and the pedantry of parsers makes it worse.