The dark side of git...

-

Git is Great!

As a distributed source code tool, git is great. I love that when I'm on an airplane I can commit code without a wireless connection and have be able to unwind what I was doing. It was clearly designed with the "offline" model in mind. The idea I can create a quick branch, experiment, make massive sweeping changes, and just drop the whole thing if I realize it sucked...is AWERSOME! For a fact this puts it ahead of it's open source predecessors (namely...SVN, CVS, and RCS).

But perhaps a victim of it's success

What I observe, however, is that a lot of folks have taken up a development model where "everything is a branch" and we end up with roles like "pull request approval engineer" (not a real title, but if you end up doing that job, you'll know that your doing it). This problem happens when the number of public branches/forks reaches a count and complexity that far exceed any value they could have possibly served.

What is productivity?

I'm going to take a somewhat unpopular stance here, but in general my stance is that branches are antiproductive... before everyone gets their pitchforks out, let me explain my version of "productivity" for a software project. Productivity is producing software that accomplishes the business purpose at a marginal cost that provides positive value. While that might have been wordy or even technically incorrect, the overall equality formula I wan to use is: sum of all activities to product the software must be less than the value the software provides. In other words, if it costs 2 million dollars to build/deploy some software, but the business can only recoup 1 million dollars in value (via cost saving or new sales or whatever) the I would consider that a failure.

The branching use case

As a software engineer, I want to create a branch so that other developers cannot see my changes in their builds.

Well that sucks because:

  1. First of all, the activity of creating the branch, merging in everyone else's branch to your branch (through possibly a different branch) is all stuff that you would get instantaneously for free if you were all working on the same mainline.
  2. Second, you're deliberately delaying visibility of changes from the rest of the team...which means the whole notion of continuous integration is getting thrown out the window
Which brings me to a key question

Are you operating with agility or fragility?

I would contend if you're branching for every feature or bug and merging them back in, your codebase and/or process is more fragile than agile.

Your Thoughts?

Comments

Joe Meirow said…
Whoa cowboy! I think your complaints aren't about "branching" as such. Your last two complaints:

I don't want experimental garbage in the mainline. So perhaps the actual complaint is that people let branches grow too large for too long before merging (testing is assumed).

Your last complaint about visibility is perhaps a function of decentralization. If we were using a centralized tool such as Perforce, you could see my unmerged branches. Of course you would also lose all the benefits that arise from the decentralized model. Just my opinion 🙂
August 24, 2019 at 5:11 AM
Mike Mainguy said…
Yeah I realize I was picking on the pattern by using a specific tool as an example. The antipattern I'm really railing against is the notion of overbranching because "it's easy" and ignoring the effort of managing code conflicts between child branches and the mainline.
August 29, 2019 at 11:15 AM
Post a Comment

Popular posts from this blog

Push versus pull deployment models

-
There are two deployment models: push and pull. In a "Pull" deployment model, the individual servers contact a master server, download their configuration and software, and configure themselves. In a "Push" model, a master server pushes the configuration and software to the individual servers and runs commands remotely. While neither technique is right or wrong, they both have some specific advantages and disadvantages that you should understand when making a decision. In the ruby hosting world, an example of a "push" deployment is heroku , while Engine Yard is an example of a "pull" type deployment. In the Pull model, each server has information about how to obtain it's configuration, when it boots (or whatever the configuration triggering event happens to be) it can continue to proceed without intervention from the master server. In the Push model, because the master server is orchestrating things, it will typically need to have a c
Read more

the myth of asynchronous JDBC

-
Image
I keep seeing people (especially in the scala/typesafe world) posting about async jdbc libraries. STOP IT! Under the current APIs, async JDBC belongs in a realm with Unicorns, Tiger Squirrels, and 8' spiders. While you might be able to move the blocking operations and queue requests and keep your "main" worker threads from blocking, jdbc is synchronous. At some point, somewhere, there's going to be a thread blocking waiting for a response. It's frustrating to see so many folks hyping this and muddying the waters. Unless you write your own client for a dbms and have a dbms that can multiplex calls over a single connection (or using some other strategy to enable this capability) db access is going to block. It's not impossible to make the calls completely async, but nobody's built it yet. Yes, I know ajdbc is taking a stab at this capability, but even IT uses a thread pool for the blocking calls (be default). Someday we'll have async databa
Read more

Installing virtualbox guest additions on Centos 7 minimal image

-
I've just spent some time setting up a bare Centos 7 image to support development stripping out as much as possible. While CoreOS is probably a better choice, we run redhat in production at this point and centos is a better fit for the time being. The problem I've found is that many of the instructions available via google were written against prerelease versions or espouse manually installing random rpms instead of just using yum. While I get this "works", I'm not a huge fan of this approach and would rather do everything with the package manager. After a LOT of scouring and trial and error, I finally found the "magic" combination. The winner is courtesy of http://www.pc-freak.net/ . I've taken these instructions and tweaked them slightly for my purposes. The original poster did this on a more "full fledged" version of centos with a windows host and my instructions are for a minimal install using and OSX host (though I'm certa
Read more