(see here for the full Agile Organizer series) Retrospectives If you do Agile, it’s probable that you do something called a retrospective. At least in the ones I’ve been involved with, there’s usually a part where we go over what’s going well, and what could be changed. With the agile organizer, I’ve found that periodically doing retrospectives about the way I am using it is useful and worthwhile. If you use it, I hope you would too.
(see here for the full Agile Organizer series) In The Agile Organizer, I described the basics my organization system. In this second part, I go into some ways that I use it that weren’t covered in the first post. The Dumping Ground Sometimes, when I’m working on an idea, I just want to get the bits down, but don’t want to think, necessarily, about how they fit just yet. In such a case, I’ll have a page where I just stick the stickies to a page wherever they’ll fit, with the idea that I’ll come back to it later to sort it all out.
In looking at functional languages, I’ve often been impressed with the clarity of implementations of Quicksort. Now that Java 8 includes lambdas, we can now do one that, while not as nice as an implementation in, say Haskell, is still quite nice. While I wouldn’t use the below code in production, as it’s choice of pivot point is naive and would go kaboom on a large sorted list from stack exhaustion, it is a nice illustration of the utility and clarity that Java 8’s lambdas when mixed with Guava functional stuff give.
As far as programming languages go, my truly favorite language is a Lisp. Scheme or Common Lisp are fine. I have a few hesitations about Clojure (the way it does recursion is offputting to me, but perhaps a few macros might clean it up), but for what I’ve done with it when working with Riemann, it’s been a pleasure, even if it’s slow going because I don’t do it frequently enough to have the needed bits already in my head.
In one of my previous posts, I wrote about how I’d written the website software for a website I maintain in Common Lisp. There were a few macros I wrote while building it that I thought were potentially useful and illustrative of the kinds of things you can do with macros that make Lisps cool.
In this post (there may be others later), I’ll be talking about a macro I wrote to deal with database queries. When dealing with database queries, making sure that you write/generate the queries properly with respect to escaping and bind variables, and connection caching, etc. can be a real pain, and I’m not a big fan of ORMs as they never really do what I want.
In the course of reading various design patterns and refactoring books (like Refactoring), I noticed while flow control is sorta mentioned here and there, it’s not addressed especially well. It mostly encourages things like replacing conditional with polymorphism and replace nested conditional with guard clauses and the like, rather than finding ways to straighten out and clarify code that contains flow control and conditionals. One of my thoughts on flow control and conditionals follows one of the rules of the Zen of Python:
Given that people can get really touchy on the subject, let me state this up front: I’m not saying we should tolerate lousy code. Over the course of my career, I’ve seen something that I will call “the cult of abstraction.” On the surface, this sounds like a good thing, until you need to debug their code. While on the surface, it’s all nice and pretty, and uses classic design patterns, when you actually want to find out what it does, you discover that:
(see here for the full Agile Organizer series) For a long time, I’ve tried a multitude of personal organization systems, from the normal calendars, to online calendars, to Evernote, org-mode, and so on and nothing has ever truly worked for me. The main reason is that I’ve got a few main needs: Capture - if it never makes it in, it’s gone Ideation - have to be able to brainstorm with it Reference - for things I need frequently enough, but not so much that I keep it in my head Planning - what am I doing?