On Monday, one of our students at $DAYJOB asked me what projects I do in my freetime. After I infodumped on her for half an hour, she asked in disbelief “And you do these in your freetime, without being paid?”.
Like, mate, did you not listen how feckin’ excited I got just then? Of course, I do these in my freetime.
To be fair, though, the last project I told her about is very dry. It’s a library to help automate CI builds. And the thing I’m thrilled to build is a compile-safe API for accessing the packages in your workspaces. Like, yeah, it does take a special kind of nerd to get excited about that…
I think the trill comes from the feeling that “I can do it better” if it is already done before or “I finally understand this in enough detail to build a nice abstraction” if not
Yeah, the latter is certainly a big part of it. The way to make it compile-safe is to use macros to generate code, so that my users can write e.g. Package::my_frontend.version and that gives them the version of their frontend package.
Writing such macros, i.e. writing code to generate code, is certainly something I haven’t done a ton of yet, because you practically cannot justify doing that in an application codebase, only in a library, so it is new stuff that I learn.
But well, you did already call it a “nice abstraction”, which is another big part where my excitement comes from and where I think, the special nerdery is necessary.
Others might build projects which are visually tangible, like a sexy GUI, or which do something tangible, for example a colleague (who I will absolutely not deny his own special nerdery) is currently building a driver for a motor. If that driver works, you can see a motor moving in the real-world. Even non-nerds can at least tell that something is happening.
But with my project, my success is that you can write Package::my_frontend instead of Package::from_str("my_frontend")?. And that if you rename the package to super_duper_frontend, that the compiler will tell you to fix the code rather than it only breaking once you actually run the build code for the frontend.
No chance of explaining to non-coders why this is exciting or even just when you’re successful.
I get what you mean, I had a similar kind of feeling when I did maths in my PhD too. I think that is an aptitude for mathematical style of thinking. I also love it when things that seem dissociated from each other suddenly becomes part of a single concept when you do a nice abstraction. I guess that is one of the points of abstraction anyway (reduce the number of building blocks as much as possible while keeping the same richness in end results).
On Monday, one of our students at $DAYJOB asked me what projects I do in my freetime. After I infodumped on her for half an hour, she asked in disbelief “And you do these in your freetime, without being paid?”.
Like, mate, did you not listen how feckin’ excited I got just then? Of course, I do these in my freetime.
To be fair, though, the last project I told her about is very dry. It’s a library to help automate CI builds. And the thing I’m thrilled to build is a compile-safe API for accessing the packages in your workspaces. Like, yeah, it does take a special kind of nerd to get excited about that…
I think the trill comes from the feeling that “I can do it better” if it is already done before or “I finally understand this in enough detail to build a nice abstraction” if not
Yeah, the latter is certainly a big part of it. The way to make it compile-safe is to use macros to generate code, so that my users can write e.g.
Package::my_frontend.versionand that gives them the version of their frontend package.Writing such macros, i.e. writing code to generate code, is certainly something I haven’t done a ton of yet, because you practically cannot justify doing that in an application codebase, only in a library, so it is new stuff that I learn.
But well, you did already call it a “nice abstraction”, which is another big part where my excitement comes from and where I think, the special nerdery is necessary.
Others might build projects which are visually tangible, like a sexy GUI, or which do something tangible, for example a colleague (who I will absolutely not deny his own special nerdery) is currently building a driver for a motor. If that driver works, you can see a motor moving in the real-world. Even non-nerds can at least tell that something is happening.
But with my project, my success is that you can write
Package::my_frontendinstead ofPackage::from_str("my_frontend")?. And that if you rename the package tosuper_duper_frontend, that the compiler will tell you to fix the code rather than it only breaking once you actually run the build code for the frontend.No chance of explaining to non-coders why this is exciting or even just when you’re successful.
I get what you mean, I had a similar kind of feeling when I did maths in my PhD too. I think that is an aptitude for mathematical style of thinking. I also love it when things that seem dissociated from each other suddenly becomes part of a single concept when you do a nice abstraction. I guess that is one of the points of abstraction anyway (reduce the number of building blocks as much as possible while keeping the same richness in end results).