Not-So-Mith'd Opportunity: Random Banter # 222

[Peelee]

I remember hearing once that the average for fully debugged and documented code was something like 4 lines per day.

I don't think I've ever come across anyone that slow, but I don't think I've ever come across fully debugged and documented code either, not that I've seen a lot of code.

Dangit, I wanted to make the "there's no such thing as a fully debugged code" joke!

[Anonymouswizard]

Dangit, I wanted to make the "there's no such thing as a fully debugged code" joke!

Of course there's such a thing as fully debugged code. You just don't know your full feature list until the user's tried it out.

[Teddy]

I remember learning that the ideal coding speed is about 2-3 lines every hour, because of everything else you have to do (in my experience that's getting it to actually compile, but I don't code for a living).

It's tricky, because it's extremely context dependent. Some days I produce several thousands of lines of code, but that's because I've set up a framework in which I copy-paste-search-replace a lot of very verbose code, and then use that to write minimalistic and easily maintainable code. I can probably still produce a thousand lines of code in a day if I've just got the ball rolling and I don't have to reference other's jobs to make my stuff work. And then there are those weeks when I do nothing but Alt + Tab between someone else's code and the documentation, and at the end of the Friday I erase half the code and write // TODO instead, because none of that code matches what I have to do.

So I rewatched Watership Down yesterday, and it's still a great film. The animation is beautiful, the story is streamlined, and scenes aren't actioned up compared to the book (I'm looking at you BBC rivet crossing!). Plus the casting is great, everybody sounds right, and the bunnies are properly bunny-like. And Bigwig is his calm, Hazel-supporting but experienced self.

I've only read the book, but I wanted to chime in as another member of the Bigwig fan crowd.

That's just what survived my numerous "I'l just purge this section and re write it."

Considering I only started serious coding in January, (grad course requiring software to be built), I really appreciate this response.

Yeah. Good experienced programmers can be really fast at solving problems they're familiar with, but that usually just leads up to them being assigned harder problems. You might not feel like it yourself, but even when you feel you're barely competent enough to make something compile, let alone actually do its job, you're still valuable enough for someone to make money of you.

I always wonder how the world would look if we kept everyone at the level where they're still complete wizards, but then there wouldn't be anyone to solve the really hard problems...

I remember hearing once that the average for fully debugged and documented code was something like 4 lines per day.

I don't think I've ever come across anyone that slow, but I don't think I've ever come across fully debugged and documented code either, not that I've seen a lot of code.

I can stretch myself believe this. Proving correctness of code or creating test cases comprehensive enough to say with confidence that everything works is really time consuming. It's kind of like when your connection cuts out right in the middle of a download, and the downloader still tracks average download speed...

Dangit, I wanted to make the "there's no such thing as a fully debugged code" joke!

I actually think some of the kernels in banking and other critical financial software come as close to this as you get. Given, the banks still don't dare to run their software if they don't have their almost-retired COBOL expert in the house, so there's that...

[Bartmanhomer]

I see that truemane and Peelee ascended to moderator status now. I guess Roland really trusted you two new mods.

[factotum]

Dangit, I wanted to make the "there's no such thing as a fully debugged code" joke!

We can say that any program can be one line shorter, and any program has at least one bug in it. We therefore can state that any program can be reduced to one line of code that doesn't work.

[enderlord99]

We can say that any program can be one line shorter

Of course. Just gotta remove a linebreak!

[Algeh]

It's tricky, because it's extremely context dependent. Some days I produce several thousands of lines of code, but that's because I've set up a framework in which I copy-paste-search-replace a lot of very verbose code, and then use that to write minimalistic and easily maintainable code. I can probably still produce a thousand lines of code in a day if I've just got the ball rolling and I don't have to reference other's jobs to make my stuff work. And then there are those weeks when I do nothing but Alt + Tab between someone else's code and the documentation, and at the end of the Friday I erase half the code and write // TODO instead, because none of that code matches what I have to do.

I am pretty sure the last time I produced hundreds or thousands of lines of code in a day was in 7th grade, and they were in line-numbered BASIC. They were also mostly just PRINT statements designed to either troll or otherwise antagonize the other 2-3 kids using those terminals, because 12 year olds make choices in a very 12 year old way.

We'd all write "fake programs" to leave running on those terminals that were basically designed to make passerby who hit a key worry they'd done something wrong. These would generally have "press any key" as their starting point, and then launch into a false narrative about all of the things the computer was now doing since you foolishly pressed a key. I think the most impressive-looking one was a "codebreaker" program that consisted entirely of slightly-time-delayed print statements to look like it was gradually "locking in" more and more digits of a number with each new line.

We would also write terrible poems to/about each other and leave them printing in a GOTO loop. This is one of many times when it has been convenient that my name does not rhyme easily. We'd also write programs that basically consisted of in-program menuing systems where all of the "good" options were "secret" keystrokes that were not listed as one of the options in our menu. Some would be several layers deep in this, to the point that some of these programs spanned multiple files.

As far as code that accomplished something, many fewer lines per day.

I miss having time to spend entire days writing sprawling, pointless programs with no clue about best practices. I don't think I could write a mess like that now even while drinking. (I wish I still had copies of these programs, but they were all on this weird, ancient mainframe and terminal system and I didn't think to print out copies or copy things to my home machine while dialed in from home, so I left them all behind when I left that school.)

[LaZodiac]

Every time people talk programming to me, I feel like a peasant in a fantasy setting listening to a wizard discuss magi-theory.

[Rawhide]

if ( talkProgramming == true ) 
{
   zodi = confused;
}

and

if ( talkProgramming == true ) { zodi = confused; }

are the same. If you're being judged based on lines added, write the first!

[Mith]

Every time people talk programming to me, I feel like a peasant in a fantasy setting listening to a wizard discuss magi-theory.

I get that. I'm not too far removed from that state to have forgotten what it was like. Even now, I look at what I want to do to improve my work, look at the explanations, and just say "Maybe another day."

But like medival fantasy wizardry, you can at least read up and become a practitioner. It's not sorcery!

[Anonymouswizard]

Every time people talk programming to me, I feel like a peasant in a fantasy setting listening to a wizard discuss magi-theory.

I've built at least one setting where mana followed similar rules to electricity. I simplified it quite a bit, but left in enough to allow you to theoretically create a magical computer (simple ones did exist in the setting, but were generally missing standing stones). I think Carbon and Silicon conducted magic well, while iron acted as an insulator, most varieties of stone and a good number of gems worked well enough as conductors but I never bothered to be too consistent about elements (diamond was am exceptionally useful material and incredibly valuable).

Never worked out how negative mana values worked though. You could still use it for spellcasting, because mana stood in for voltage, but I never worked out if the reversedata polarity would have additional affects.

[Teddy]

We can say that any program can be one line shorter, and any program has at least one bug in it. We therefore can state that any program can be reduced to one line of code that doesn't work.

Nonono, the true kernel of all programs is the zero line program. And it doesn't work.

Of course. Just gotta remove a linebreak!

Unless it is a language in which line breaks matter, like Python.

I am pretty sure the last time I produced hundreds or thousands of lines of code in a day was in 7th grade, and they were in line-numbered BASIC. They were also mostly just PRINT statements designed to either troll or otherwise antagonize the other 2-3 kids using those terminals, because 12 year olds make choices in a very 12 year old way.

We'd all write "fake programs" to leave running on those terminals that were basically designed to make passerby who hit a key worry they'd done something wrong. These would generally have "press any key" as their starting point, and then launch into a false narrative about all of the things the computer was now doing since you foolishly pressed a key. I think the most impressive-looking one was a "codebreaker" program that consisted entirely of slightly-time-delayed print statements to look like it was gradually "locking in" more and more digits of a number with each new line.

We would also write terrible poems to/about each other and leave them printing in a GOTO loop. This is one of many times when it has been convenient that my name does not rhyme easily. We'd also write programs that basically consisted of in-program menuing systems where all of the "good" options were "secret" keystrokes that were not listed as one of the options in our menu. Some would be several layers deep in this, to the point that some of these programs spanned multiple files.

As far as code that accomplished something, many fewer lines per day.

I miss having time to spend entire days writing sprawling, pointless programs with no clue about best practices. I don't think I could write a mess like that now even while drinking. (I wish I still had copies of these programs, but they were all on this weird, ancient mainframe and terminal system and I didn't think to print out copies or copy things to my home machine while dialed in from home, so I left them all behind when I left that school.)

Heh, that sounds like a fun time.

if ( talkProgramming == true ) 
{
   zodi = confused;
}

and

if ( talkProgramming == true ) { zodi = confused; }

are the same. If you're being judged based on lines added, write the first!

Given, nowadays you usually use a common formatter in any team or company setting which makes your personal choice more or less irrelevant. That said, I personally prefer K&R style indentation over Allman style.

ION:
I finally got my vacation, and am now back up at Ancestral Farm for the first time since Christmas. You can see that last year was tough on the vegetation, because everything is growing and blooming like crazy to catch up. It's a bit cold, though, so I haven't felt like testing the water yet. Hopefully the weather will warm up over time (the prognosis looks so-so).

Tomorrow we're heading up to the mountains for three-four days of hiking. It's probably going to be even colder, but I'm looking forward to it.

[DataNinja]

Nonono, the true kernel of all programs is the zero line program. And it doesn't work.

Sure it does. It does exactly the nothing that you expect it to do.

[Bartmanhomer]

I just have a wonderful dream last night:

Spoiler: Last Night Dream

Show

Last night Roland St. Jude assigns me as a moderator in this forum. I was so nervous and excited. I was in charge of D&D 3.5 Sub-Forum. Members were discussion and debating tiers and stuff. There was a heated debate where everybody discussed If Pun-Pun a dragon or not. I read the post and there were a lot of name-callings. So I went in mod mode and trying to calm everybody down and tell them to be civil and tell them that my knowledge that Pun-Pun is a dragon wrought Kobold so he's a dragon in a way. I give them the sources and everybody agrees. That was it for my dream.

[LaZodiac]

I've built at least one setting where mana followed similar rules to electricity. I simplified it quite a bit, but left in enough to allow you to theoretically create a magical computer (simple ones did exist in the setting, but were generally missing standing stones). I think Carbon and Silicon conducted magic well, while iron acted as an insulator, most varieties of stone and a good number of gems worked well enough as conductors but I never bothered to be too consistent about elements (diamond was am exceptionally useful material and incredibly valuable).

Never worked out how negative mana values worked though. You could still use it for spellcasting, because mana stood in for voltage, but I never worked out if the reversedata polarity would have additional affects.

My own novel's magic system, since we're bringing it up, is based on art (with a little math/science thrown in since I do respect that even if I don't understand it). Study something long enough, see the inherent beauty and flow of it, learn to understand it, and you'll be able to make your own spells. But once they're made, anyone able to writing a magic circle and who has the capacity to spellcast can cast it. And that's not even getting into the finer details.

I'd think negative values would probably reverse the spell. Fireball becomes cold, blue fire that freezes on impact, creating an implosion instead of an explosion. As an example.

[Anonymouswizard]

My own novel's magic system, since we're bringing it up, is based on art (with a little math/science thrown in since I do respect that even if I don't understand it). Study something long enough, see the inherent beauty and flow of it, learn to understand it, and you'll be able to make your own spells. But once they're made, anyone able to writing a magic circle and who has the capacity to spellcast can cast it. And that's not even getting into the finer details.

Surely to understand something you have to break it down, take it apart,and learn how it functions. To find out the relationship between every part. Definitely sounds like science to me. In all seriousness though, sounds cool.

Although that's essentially how Divination magic was discovered in my setting. Nexuses essentially perform complex calculations to regulate the world, somebody worked out a way to replicate this and get somewhat correct answers about the world (if only to binary questions). Unfortunately somebody then and made a stone tablet that made this easy to do...

I'd think negative values would probably reverse the spell. Fireball becomes cold, blue fire that freezes on impact, creating an implosion instead of an explosion. As an example.

Not really consistent with this whole 'electrics' theme I've got going on. I think I might end up ruling that complex spells don't work in a negative mana zone unless intentionally designed to, and good luck getting a standard Divination Tablet to function.

[Honest Tiefling]

Every time people talk programming to me, I feel like a peasant in a fantasy setting listening to a wizard discuss magi-theory.

Just start throwing cats at people's faces and see how long it takes for one to die. Nothing says lack of player class levels quite like a lack of HP.

[LaZodiac]

Surely to understand something you have to break it down, take it apart,and learn how it functions. To find out the relationship between every part. Definitely sounds like science to me. In all seriousness though, sounds cool.

Although that's essentially how Divination magic was discovered in my setting. Nexuses essentially perform complex calculations to regulate the world, somebody worked out a way to replicate this and get somewhat correct answers about the world (if only to binary questions). Unfortunately somebody then and made a stone tablet that made this easy to do...

Not really consistent with this whole 'electrics' theme I've got going on. I think I might end up ruling that complex spells don't work in a negative mana zone unless intentionally designed to, and good luck getting a standard Divination Tablet to function.

I suppose it can be read as science, yeah. My head goes to art because that's how my thinking tends to go. And since there's various "styles" of magic, but all of which are just magic at the end of the day, you can probably make use of math, science, or art to learn spells and such. Everyone's mind works a little differently, after all, so people are going to learn differently as well.

Incidently if you wanna read my book I can send you the latest draft and you can become a beta-reader .

I'm absolutely tank at math so I have no idea what negative magic would actually mean by that terminology so I was going off my understanding.

Just start throwing cats at people's faces and see how long it takes for one to die. Nothing says lack of player class levels quite like a lack of HP.

The bane of all hedge wizards; a house cat.

[halfeye]

I'm absolutely tank at math so I have no idea what negative magic would actually mean by that terminology so I was going off my understanding.

Does it have to mean anything? Temperature has an absolute limit, there is no temperature below absolute zero. Pressure is much the same, there is no pressure below zero. There are systems (electricity for one) where negatives make sense, but it isn't a rule that things have to be that way, and it might be easier to have no negative.

[factotum]

There are systems (electricity for one) where negatives make sense, but it isn't a rule that things have to be that way, and it might be easier to have no negative.

Even there, though, the only thing that really can go negative in electricity is voltage (a negative current or power would make no sense at all), and that all depends on the reference you're measuring it from, so a negative voltage isn't really significant--you can remove it simply by finding the lowest voltage in your circuit and measuring everything with respect to that. The only thing that prevents you doing so is the convention that "ground" or "earth" must be zero volts.

[Thufir]

So I rewatched Watership Down yesterday, and it's still a great film. The animation is beautiful, the story is streamlined, and scenes aren't actioned up compared to the book (I'm looking at you BBC river crossing!). Plus the casting is great, everybody sounds right, and the bunnies are properly bunny-like. And Bigwig is his calm, Hazel-supporting but experienced self.

I give it four out of uncountable stars.

The film of Watership Down remains one of the best book-to-screen adaptations I can think of.

[DavidSh]

On the contrary, a negative current makes perfect sense in the context of solving for voltages and currents in circuits more complicated than a single loop. It just means that the direction of the current over the circuit element is the reverse of the direction you defined to be positive.

[Teddy]

Never worked out how negative mana values worked though. You could still use it for spellcasting, because mana stood in for voltage, but I never worked out if the reversedata polarity would have additional affects.

Wouldn't it be simpler if mana stood for charge, and voltage instead became some sort of spellpower? That way, reverse polarity would basically mean you were draining magic from the surroundings. Obviously you wouldn't be able to output magic under those conditions, but it could be its own school of magic to manipulate that reverse current of magic.

Sure it does. It does exactly the nothing that you expect it to do.

Yeah, but it doesn't work, because doing nothing isn't work.

Surely to understand something you have to break it down, take it apart,and learn how it functions. To find out the relationship between every part. Definitely sounds like science to me. In all seriousness though, sounds cool.

I'm not sure that's how you understand art, necessarily. Or at least not everyone would agree that's how you do it.

[Peelee]

The film of Watership Down remains one of the best book-to-screen adaptations I can think of.

Shawshank Redemption. Only movie I've ever seen where it was better than the book.

[LaZodiac]

Does it have to mean anything? Temperature has an absolute limit, there is no temperature below absolute zero. Pressure is much the same, there is no pressure below zero. There are systems (electricity for one) where negatives make sense, but it isn't a rule that things have to be that way, and it might be easier to have no negative.

I mean not my world and I'm not smart enough to contest that so *rotates away into the sunset*

The film of Watership Down remains one of the best book-to-screen adaptations I can think of.

I'd back this.

Shawshank Redemption. Only movie I've ever seen where it was better than the book.

I'd also incredibly back this. Reading the differences between the two made me realize part of why I don't actually like King's novels that much. He doesn't put as much thought as he should into them.

At the same time I can't be mad because I've also done something arbitrarily that people have been like "wow what a cool powerful intentional thing you did" and I'm just like "... yes, exactly, perfect. I'm the best" so I can kinda get how he feels. I never like, left OUT something though (like not connecting the warden's favorite passage of Exodus to his escape).

[Anonymouswizard]

Even there, though, the only thing that really can go negative in electricity is voltage (a negative current or power would make no sense at all), and that all depends on the reference you're measuring it from, so a negative voltage isn't really significant--you can remove it simply by finding the lowest voltage in your circuit and measuring everything with respect to that. The only thing that prevents you doing so is the convention that "ground" or "earth" must be zero volts.

Nah, negative currents are a thing. They're just currents going against the direction you've defined as positive. Which is obviously from the negative/ground terminal to the positive!

The thing with negative voltage is that going below your Ground level can cause problems if you've not set yourself up to deal with it (which isn't hard, just makes sure you're drawing your voltage from ground). When it comes to magic negative mana is more impactful,and thinking through it again I've worked out a lot of what it means (something that draws from the world rather than using it as it's ground, essentially bestowing light magic resistance while still using spells).

The film of Watership Down remains one of the best book-to-screen adaptations I can think of.

I'd get behind that. The story managed to go from being rejected by all the male publishers to a well loved film that I'm fairly certain all of my friends have seen, and it retained the heart of the story.

Reading the differences between the two made me realize part of why I don't actually like King's novels that much. He doesn't put as much thought as he should into them.

Having pretty much read The Gunslinger and nothing else, I'd agree. I'd heard people talk up The Dark Tower, and so picked up the first book to find a rather standards fantasy setup with swords replaced guns, and a protagonist who couldn't hold my interest. There was effort, but there wasn't enough thought.

At the same time I can't be mad because I've also done something arbitrarily that people have been like "wow what a cool powerful intentional thing you did" and I'm just like "... yes, exactly, perfect. I'm the best" so I can kinda get how he feels. I never like, left OUT something though (like not connecting the warden's favorite passage of Exodus to his escape).

I really need to get back to rewriting a story of mine. A friend pointed out that it was much creepier and more interesting during the red herring phase, and that I should just rewrite it to explore that red herring instead. Essentially they wanted me to go back to the idea of 'AI who runs on multiple machines as a gestalt' and build on how creepily alien it was. Which I'm happy to do as gestalt AI play a big part in that universe, in fact it's seemingly impossible for AI to exist on a single processor.

[Caerulea]

Sure it does. It does exactly the nothing that you expect it to do.

Yeah, but it doesn't work, because doing nothing isn't work.

Not only that, depending on the language it won't compile or run.

Spoiler: Images

Show

[IMages redacted]. The langauges were c and java.

(waits for people to point out something that makes me feel dumb).

I really need to get back to rewriting a story of mine. A friend pointed out that it was much creepier and more interesting during the red herring phase, and that I should just rewrite it to explore that red herring instead.Essentially they wanted me to go back to the idea of 'AI who runs on multiple machines as a gestalt' and build on how creepily alien it was. Which I'm happy to do as gestalt AI play a big part in that universe, in fact it's seemingly impossible for AI to exist on a single processor.

Hmm? Why in the world would you try running an AI on 1 processor? Parallise as much as you can. You don't train ANNs on 1 cpu, you use a bunch of gpus. I'd imagine any AGI would have to inhabit a datacenter, unless your cpu is stupidly powerful. Am I being silly?

[Bartmanhomer]

I'm very happy now. I used to be angry a long time ago but now I'm not. The reason why I'm saying it because I just wanted to get it out of my chest.

[Algeh]

Yeah, but it doesn't work, because doing nothing isn't work.

Aw man, doing nothing was my dream job! Now I gotta find a new dream job.

[Qwertystop]

On the topic of programming-is-magic – anyone else seen the stuff here?