We have a spectrum of interesting needs, and a corresponding spectrum of solutions. Here's one dimension: the structured-ness view.

1. Highly structured data.
2. Semi-structured data, or the "mixed content model". We can further subdivide this into content which is "naturally" hierarchical, and a good fit for XML, and data which is …

If we look at a spreadsheet as a document we can develop a parser that locates the user-supplied information in that document. Like any language, we'll have to treat some elements of the spreadsheet as pure syntax, little more than punctuation. In the case of embedded formulas, these are elaborate …

See "When a Column is Not a Column " for a rant on failed attempts to scale spreadsheets up into relational databases. Bottom Line: A spreadsheet column is rarely an RDBMS column. Often the SS column's label is a key value, just like the SS row's label, and the cell's value …

In one form, the question is "How do we handle the [X] transaction in the warehouse?" Another form of the question is "What do we do when [Y] changes?" The third form is less clear, but essentially the same: "How do we maintain [Z] in the warehouse?"

All of these …

Yes, it's an omission and, yes, it's a mistake. I should have said something along the lines of "risk is an uninteresting complication" or "risk is just a euphemism for bad management."

Why is risk uninteresting?

First, the seeds (Chad Fowler's "The Big Rewrite ", and CodeCraft, "To Rewrite or not …

In "It's Strategic -- but it's not -- but it was -- now it's a burden ", I rambled about the situation I encounter on a regular basis. Here's a Haiku that captures the situation.

Please Replace It

It's heavily customized,

We can't install an upgrade,

We don't dare touch it.

The maintenance costs …

First, I'm a pack-rat; I worry about deleting something valuable. Specifically, I worry about orphaned files because the application software is no longer supported or the media is obsolete.

For example, my resume is a Hypercard stack. Really. Hypercard is an OS 9 application, and future OS X's won't support …

Here's what I observed.

Customer: "Our business performance isn't what we want. We think some of our intranet is no good."

Us: "We'd love to look at it."

Customer: "Not so fast. You can't 'look at it'. You have to propose a measurable change in business performance, define a fixed …

To begin with, we're talking about the implementation's source code. Here's the situation: Ann says that Bob's code is no good, and she's provided an alternative implementation. How do we choose?

The first part of any scorecard is the SEI Quality Measures Taxonomy . It's marked "Legacy", but I haven't seen …

While there are lame reasons for using XML, there are good reasons also. The good reasons, however, aren't delightfully clear. They seem to be clouded by the lame issues. I'm trying to sort out the best and most logical reasons for using XML.

Here are 8 typically lame reasons for …