Live data from Hacker News

The Metaprogrammer

codersnotes.com

141–150 of 164 posts

Re: The Metaprogrammer

#141

This effect is observable not only in the programming community, but in many others. Best example would be photography, where people spend much more time discussing their equipment and the newest lenses instead of going out and actually take pictures. I had to experience this for myself when searching for new hobbies. With most of them, I spent most time with the meta, e.g., which fountain pen to use for my writing o…

Classic wet shaving is a bit like that. I'm not sure if it's exactly the same as your examples though, as a lot of people into this particular hobby definitely do use and enjoy the different products. "Shave of the day" posts on forums typically show posters different set ups, with the same users having a vast rotation of gear. I guess it'd be like having lots of different pens or skateboards. But I've also seen many…

The problem with "just getting good at the game" is that it may take even more time than theorycrafting that has some objectively better result (the usual term that WoW players used when I played at least). With action games there's a certain point when you can play more but you're at a plateau because you're limited by skill and have few means to even know what to improve. Focused practice matters for any activity involving skill, but I dunno what black abyss of learning exists between me flailing my mouse like an idiot and the dudes that no scope headshots on 20 pixels reliably. Industry marketing likes to tell me I need a new mouse (some 4000+ dpi thing), new monitor (144 Hz+), but I think I'm missing something more basic like basic hand-eye coordination that gear will not solve.

Re: The Metaprogrammer

#142

Earlier quoted context omitted.

I have a friend like this. I've been in bands in the past and worked in (even built my own) studios. I seen and partaken the 24/96 vs. 16/48 debate (spoiler: that only matters in the studio). I've worked with high end mics, pres, dacs, boards, tape, etc. The big thing we tangle on is always digital vs. analog. Oh boy.... Zzzzz... I've cut tape and editing on analog sucks. Noise on analog is not ideal. Compression and…

My theory about audiophiles is that a lot of them grew up in the 1970s-1990s where there really was a huge difference between "high end" and consumer grade audio. That gap has narrowed considerably. I think there's also a lot more people these days out there actually explaining the engineering and science in good audio (in the old days technical details were considerably harder to look up), this makes it hard to push…

Modern "audiophile" gear is much better today than in the 90s, especially for the same price points, but my observation is that there's a divergence between hipsters that prefer gear older than themselves looking for something unique and those that seek technical specs religiously and know where to optimize for best technical sound. I just want the best sound I can get within a reasonable budget even if I could tell the difference if I spent $1000+ more on a better amp.

Most of the anti-digital people I know fall into two overlapping camps - technophobes that don't know how basic DACs work (believing that digital audio means output of literal square and stepper waves into the air) or those that simply prefer older recordings from a point of nostalgia or artistic taste. Showing conclusively that some of their favorite records were mastered digitally before being compressed and filtered for LP is similar to watching climate change deniers try to reason via strawman and red herring arguments.

Re: The Metaprogrammer

#143
post #137

Earlier quoted context omitted.

I personally disagree. That requires that you use vim or emacs, which is not feasable (or even desirable) for many programmers. It also requires a stable and low latency network connection; something which is not always available. Anecdotally, I had to work this way once for an employer. It worked rather well, but our "development" machines were on-premises, and when they went down, everybody was down. I was also ext…

Use vim or emacs not feasible? not desirable? If most of the greatest programmers use them, if you're a wannabe programmer how would you not use these tools? It's like willing to be a cowboy but for thanks to the horse and the gun...

Cowboys have evolved. Many use ATVs and only infrequently have to use a gun. The hat has stuck around, though - great protection against the sun.

I've attempted to write Java using VIM. I've even been moderately successful. But it's still easier to do with Eclipse or Intellij.

Re: The Metaprogrammer

#144
post #93

Earlier quoted context omitted.

Why would you do that? Two keyboards sound a waste of time

I'm just used to it, there's no good reason. First, I had a desktop, then bought a laptop, kept the desktop in front of me, the laptop on the left. At some point the desktop left my desk and I only had a laptop, then after many minor changes over time [skipping a lot] I found myself in front of two desktops. But it doubles the power! Isn't it how this work?

It depends. How do you coordinate the work between the two computers?

Re: The Metaprogrammer

#145
post #106
post #89

Earlier quoted context omitted.

I should be able to run unit tests locally, and have it happen quickly. Added to that, I'm often travelling on the job and sometimes code on the go, massive multicore build machine may be out of my reach > As time goes on (16 years as a dev now) I find myself a lot less concerned about this stuff. That's the point, we shouldn't be concerned, and it's best achieved by giving us equipment that lets us take this stuff o…

I think the point is that as you gain experience you realise how unimportant your local terminal is. Engineers tend to grow out of it. Not that many coders 'often' travel on the job, as a proportion. Most are office-bound.

One more data-point: I've purposefully worked on a thin client. I had an employer that gave us access to powerful machines but not laptop, so I bought a cheap, under-powered laptop and did almost everything on the servers. My experience was that at first it was great and I appreciated the habits the constraints encouraged. Then I took a job where I was given a beefy laptop and immediately noticed a whole bunch of pain points that I had just been living with that disappeared entirely. The good habits from having a low-powered machine can transfer onto a high-powered machine, but there is are a whole lot of little irritations that completely go away when you have more computing power locally available.

Re: The Metaprogrammer

#146
post #93

Earlier quoted context omitted.

I'm just used to it, there's no good reason. First, I had a desktop, then bought a laptop, kept the desktop in front of me, the laptop on the left. At some point the desktop left my desk and I only had a laptop, then after many minor changes over time [skipping a lot] I found myself in front of two desktops. But it doubles the power! Isn't it how this work?

It depends. How do you coordinate the work between the two computers?

Currently, one of them mostly acts as a server, and I work with network shares and symlinks.

But that's how I work, I won't force it on anyone else.

Re: The Metaprogrammer

#147
post #40

This is funny because no-one bats an eyelid at an "executive" demanding a corner office, an expense account, a company Mercedes, to fly first class... But God forbid a humble worker should beg for scraps.

An executive is different from a salaried employee, unless you're a githuber or whatever. Get over it.

In what ways specifically, from a bottom-line point of view at a software company, is an executive different from a programmer that justifies the executive asking for a more pleasant work environment, but not the developer?

Re: The Metaprogrammer

#149
post #34

"If they don't have the shiniest Apple MacBook then they can't work. Never mind that their job involves typing letters into a text file, something you could have done on CP/M back in '78" I don't want an Apple at all, but I do want a beefy machine because my job doesn't only involve typing text you see - for instance I also need to build the result of all that typing, and if it takes ages, guess what, the time I wast…

I may be an old-timer and I typed my code into CP/M on a text-only display starting circa 1983 or so. I had to compile the code too and guess what ... it was fast!

But when you've only got 64K RAM there's a limit to how sophisticated a program can be. And when you're writing the type of tools we could write back then, your compilation cycles would be fast too.

Would I go back? Some of my CSS files are bigger than the hand-coded assembly language programs I wrote back then but I still love what computing has become. Many facets of today's systems seem almost unimaginable (except that I was an avid sci-fi reader). So I'll join you in demanding a fast machine so that I'm not waiting for code to compile. And who knows what we'll be building in another 30 years?

Re: The Metaprogrammer

#150
post #138
post #118

Earlier quoted context omitted.

> Building, editing, testing, all of that is better done remotely with ssh and gnu screen or similar. Why? I have a full command line suite running on my laptop, because I'm running Linux. Why would I log in elsewhere, when I can run stuff locally?

For many reasons. The remote server is in a data center, supposedly better maintained than your PC, it has backups, RAIDx all the stuff. If you automate some tasks, or run long-lasting scripts, it is much more reliable. Also, keeping it in the terminal forces you to use the right tools, e.g. vim/emacs instead of the good old Dreamweaver...

Just because I'm working on my PC doesn't mean my changes aren't pushed elsewhere frequently (admittedly the "trunk based development" crowd seem determined to make this less flexible than it ought to be...)

I'm personally an emacs user for most purposes, but I'm not going to claim it's the right tool for everyone, all the time.

Post reply on HN