More than a decade ago, I published Effective Enterprise Java, and in the opening chapter I talked about the Ten Fallacies of Enterprise Computing, essentially an extension/add-on to Peter Deutsch’s Fallacies of Distributed Computing. But in the ten-plus years since, I’ve had time to think about it, and now I’m convinced that Enterprise Fallacies are a different list. Now, with the rise of cloud computing stepping in to complement, supplment or replace entirely the on-premise enterprise data center, it seemed reasonable to get back to it.
At first, it was called “DLL Hell”. Then “JAR Hell”. “Assembly Hell”. Now, it’s fallen under the label of “NPM-Gate”, but it always comes back to the same basic thing: software developers need to think about their software build and runtime dependencies as a form of Supply Chain Management. Failure to do so—on both the part of the supplier and the consumer—leads to the breakdown of civilization and everything we hold dear.
tl;dr Patterns, 20 Years Later: The Abstract Factory pattern is often “combined”, conceptually, together with Factory Method into a sort of uber-“Factory pattern”. The two are distinctly separate in the Gang-of-Four literature, however, and for some pretty good reason, as the intentions are different. Subtly so, in some ways, but still different.
tl;dr Creational patterns specifically deal with the creation of objects/entities in the code. They abstract the instantiation process (meaning that most of the time, they provide an abstraction layer above the use of the raw language facilities to construct an object). They help make a system independent of how its objects are created, composed, and represented.
tl;dr Patterns, 20 Years Later: The Builder pattern has enjoyed some success within the O-O community, particularly among the crowd that sees it as a way to build fluent APIs (APIs which read, more or less, like a natural language, a la English). Builder has a few tricks up its sleeve beyond just fluent APIs, however.
Let’s talk a bit about the various pattern implementations, the languages in which they are being written, and what to “get” out of them.
tl;dr Patterns, 20 Years Later: With the prevalance of libraries and tools that can peer past access controls (such as Reflection on the JVM and CLR, or the Mirrors facility in Swift), trying to encapsulate private details away from prying eyes can be increasingly difficult. Other languages lack access controls entirely, or the concept of objects. In any of these languages that offer closures, we can hold state within a closure yet outside of the object, rendering it almost entirely inaccessible to outside parties.
tl;dr Patterns, 20 Years Later: A Constructor Function is a function designed specifically to construct instances of entities (typically objects, although in languages which do not support objects as native types, this will typically be something that masquerades as an object). It is often seen as a variation on a Factory Method, though there is enough variation on the intent that it is worth calling this out as a standalone pattern.
tl;dr Patterns, 20 Years Later: Factory Method is a pattern that is often called by a simpler name hinging on the word “Factory”, a la “the Factory pattern” or somesuch. The GOF language actually has two patterns which each could qualify under that moniker, this one and the Abstract Factory, depending on the intent and the desired consequences.
tl;dr Patterns, 20 Years Later: Let’s start with everybody’s favorite (and most despised) pattern, the Singleton. Everybody loves the Singleton because, conceptually, it seems the easiest of the lot to understand and (in most post-1995 languages) the easiest to implement. But everybody hates it because its singleton-y nature means it is a natural target for concurrency problems up the wazoo. (And then there’s that whole “Singleton instance vs static methods” debate that goes on.)
tl;dr 20 years ago, the “Gang of Four” published the seminal work on design patterns. Written to the languages of its time (C++ and Smalltalk), and written using the design philosophies of the time (stressing inheritance, for example), it nevertheless spawned a huge “movement” within the industry. Which, as history has shown us, was already the hallmark of its doom—anything that has ever become a “movement” within this industry eventually disappoints and is burned at the public-relations stake when it fails to deliver on the overhyped promises that it never actually made. It’s time to go back, re-examine the 23 patterns (and, possibly, a few variants) with a fresh set of eyes, match them up against languages which have had 20 years to mature, and see what emerges. (Spoiler alert: all of the original 23 hold up pretty well, and there’s a lot of nuance that I think we missed the first time around.)
tl;dr Peter Verhas asks a seemingly innocent question during a technical interview, and gets an answer that is not wrong, but doesn’t really fit. He then claims that “Sometimes I also meet candidates who not only simply do not know the answer but give the wrong answer. To know something wrong is worse than not knowing. Out of these very few even insists and tries to explain how I should have interpreted their answer. That is already a personality problem and definitely a no-go in an interview.” I claim that Peter is not only wrong, but that in addition to doing his company a complete disservice with this kind of interview, I personally would never want to work for a company that takes this attitude.
tl;dr I’ve been asked a number of times over the years how, exactly, I approach learning new stuff, whether that be a new programming language, a new platform, whatever. This is obviously a highly personal (meaning specific to the individual offering the answer) subject, so my approach may or may not work for you; regardless, I’d suggest to anyone that they give it a shot and if it works, coolness.
tl;dr My first (!) course is up at Pluralsight: “On Polyglot Programming”.
tl;dr It’s been a few years since I did this particular routine for the NFJS shows, but I found a sequence of demos/explanations that really demonstrated clearly why Java (and other classic O-O) developers should learn a little functional programming style, even if they never pick up an actual functional language. And the key to that sequence of demos? “Collections are the gateway drug to functional programming.”
tl;dr A recent post on medium.com addresses the topic of technical debt; I had an intuitive disagreement with the thrust of the post, and wrote this as a way of clarifying my own thoughts on the matter. It raises some interesting questions about what technical debt actually is—and if we can’t define it, how can we possibly understand how to avoid it or remove it, as opposed to our current practice of using it as a “get-out-of-this-codebase-by-blowing-it-all-up” card?
It’s really starting to appear like the “technical monoculture” that so pervaded the 90’s and 00’s is finally starting to die the long-deserved ugly death it was supposed to. And I couldn’t be happier.