Re: What is the logic of storing XML in a Database?

From: Daniel <>
Date: 30 Mar 2007 06:33:46 -0700
Message-ID: <>

On Mar 29, 11:58 am, "Marshall" <> wrote:
> I have heard all the arguments for text formats, and they
> are all completely without merit. I observe that http transfers
> more binary files with fewer problems than it does html
> files. Consider jpeg, a *complicated* binary file format
> which requires a math-intensive discrete cosine transform
> decompressor that very few people could implement correctly
> to work. And yet somehow there is no problem.
All streams of bytes, whether of characters or other data, require encoding and decoding at the endpoints. The distinguishing feature of text documents is that any given document typically has a single encoding. Encoders and decoders are ubiquitous for the popular charsets. HTTP and XML have discovery mechanisms for the encoding.

All the bytes in a jpeg file have the same encoding, jpeg binary representation, and encoders and decoders are also ubiquitous. But it does not follow from this observation that binary data is a good format for data interchange, particularly between loosely coupled endpoints. A more likely analogy would be to some data structure containing a mixture of png, gif, jpeg, etc.

Binary data structures for data interchange do not typically contain data in one representation, nor are encoders and decoders ubiquitous. The number of legacy binary formats is huge, some are proprietary, and representations differ on different machines.

The standards are not currently in place to have ubiquitous exhange of binary data. There are not even sufficent standards in place to support ubiquitous access to all data in the popular DBMS products through industry standards such as ODBC, and that is small compared to the broader data interchange problem.

In some sense, avoiding having to solve this problem is the entire reason for XML.

Best regards,
Daniel Parker Received on Fri Mar 30 2007 - 15:33:46 CEST

Original text of this message