IBIS Gets a Tune-Up
I thought I would wrap up Intertwingler by the end of 2023, but I got diverted by a request to do something that ultimately turned out to be equally important.
I thought I would wrap up Intertwingler by the end of 2023, but I got diverted by a request to do something that ultimately turned out to be equally important.
Recounting an impromptu episode last week where I helped Venkat Rao fix his blog, Ribbonfarm dot com.
This is the definitive write-up of the conceptual framework I am calling the Specificity Gradient.
I think I have finally identified what bugs me so much about spreadsheets: they come right up to the line of being useful for SO many other things, but stop short. I am almost mad enough to do something about it.
Lots of people have opinions about testing software, so I get to have some too.
À propos of nothing, I decided to do a roundup of all the programming languages I have used more than zero.
Last Friday I tempted fate by changing my work setup. Let's just say I learned a lot about the state of the art of networked file systems.
The lack of a lightweight, efficient, directly-attached persistent storage mechanism, that can be readily shared between programming languages and frameworks, has frustrated Semantic Web development. I intend to do something about it.
I find this matrix to be extremely useful when considering Resources, as understood by, e.g., Roy Fielding's REST dissertation.
After 11 years as a staunch Mercurial user, I am finally capitulating to Git.
It doesn't even involve aesthetics.
“I don't really see a difference between prototype and production code.” OK what the hell do I mean by that?
I need a separate place to put my technical writing. As an homage to (French) LogoWriter, I am calling it the Verso.
This is a draft of a protocol I designed for embedding RDF statements in plain HTML forms, enabling quick-and-dirty Semantic Web applications.
A design principle I've been developing, for a good six or so years, finally gets a name.
A pull request ensures a happy ending.
My roots in information security have long given me the “no” feeling when it comes to the increasing dependency on JavaScript to get basic things done on the Web. This is an idea to fix it.
This is just an idea at this stage, and I'm confident I'm not the first to consider it: using flags and/or heraldry to make long, generated identifiers more memorable. (Disclaimer: no examples yet!)
If you are blissfully unaware of the ins and outs of Linux system maintenance and its many failure modes, you can ignore this piece. Otherwise, put on your sanctimony hat.
This is the second installment of my observation of the state of Web application development, along with my vision for the kind of system I would like to use.
This is an attempt to articulate my understanding of the state of Web development and how it came to be. It is not meant to be a completely accurate account. Plus, it wouldn't be right to post something on the Web without a healthy dose of editorial. It is the first of a two-part series, the second consisting of what I'm doing about it.
In order to promote my principle of expedient desirable products, I figured it would make sense to begin providing some.
When it comes to information security, user experience is often an afterthought. The non-paranoid rarely understand for themselves the principles that keep them safe while simultaneously bringing their offspring online. What can we do to help them?
Every business problem can be imagined having a corresponding minimum viable product, which represents the most bare-bones solution somebody will buy. But how much of our relative effort do we want to spend finding it?
Looking back on a particularly challenging episode of my career, I consider the value of conceptual integrity and how it affects the bang-to-buck ratio of writing code.
Reverse Polish Notation is an extremely economical way to make sense to computers at the cost of making sense to people. But what about applying the same principle to arrange language in a way that is most useful to people?
With the advent of Agile process models and the increasing influence of user experience design, iterative development promises value early on — but does it deliver?
A conversation on Twitter this summer with a lead at an agency led to this screed about features as a metric for software projects.
I drew inspiration from an annoying software misconfiguration left untouched for an age to pen a screed about the value of ancillary and maintenance-oriented knowledge work.
This is a sketch of an idea for naming projects and other processes and properties within an organization by way of randomly-generated cryptonyms. These cryptonyms serve as intentionally meaningless handles to ultimately decouple projects from products and minimize the psychological implications that meaningful names may evoke.
What if, instead of (or in addition to) syndication feeds for new articles, we made feeds for new links?
This document explores the relationship of Web syndication feeds to conventional hypertext documents.
This note is a cursory, non-scientific inquiry into the application of the concept of crop-rotation, an ancient agricultural technique for preserving the fertility of land, into the implementation phase of a software project. No research as of yet has been performed to ascertain if this adaptation has been attempted in similar environments. This note is one of an upcoming series on software…
This note is a cursory, non-scientific inquiry into the application of the concept of crop-rotation, an ancient agricultural technique for preserving the fertility of land, into the implementation phase of a software project. No research as of yet has been performed to ascertain if this adaptation has been attempted in similar environments. This note is one of an upcoming series on software…
This note is a cursory, non-scientific inquiry into the application of the concept of crop-rotation, an ancient agricultural technique for preserving the fertility of land, into the implementation phase of a software project. No research as of yet has been performed to ascertain if this adaptation has been attempted in similar environments. This note is one of an upcoming series on software…
Avoid using characters in URIs your users don't know how to pronounce.
No, wait: THIS is by far the most popular page on this site.
In order to save a file you typically have to come up with a name for it. Here I advocate separating the naming part from saving part (from the publishing part).
Making the case for a mechanism for preserving URI history to track renames and deletions.
Some Web resources really are just files.
Policy guidelines for Web resources proper.
TL;DR there are a lot more available than you'd expect.
TL;DR you don't need 'em.
Policy guidelines for how URI parameters in both the path (underutilized) and query components ought to be interpreted.
Web resources are much richer objects than files, and this fact is rarely taken full advantage of.
This is, by far, the most popular document on this entire website.
Hyphens-rule, underscores_drool.
General guidelines for choosing a domain name.
The standardized constraints on URI syntax are a lot looser than you would expect them to be, but it behoves us to come up with artificial constraints that create behaviour we can depend on.
You cannot depend on the CaseMixing of your domain name to be preserved.