Years go by and I realize the mockery most people make of friendship (including myself) and so as sober as one can be: is there any point in planned collaboration?
Many attempts I've made at collaboration. Then the project loses steam, or it loses purpose, or it just dies out. There's conflicting emotions also. Resentment and disappointment froth around the whole ordeal. It's just not enjoyable. Most people I've worked with drag their knuckles (rightfully so if they're trapped in their prison of misery, I understand) and I've dragged mine plenty too, I'm horrible to work with it seems.
Anyway, I'm diving deeper into new programming subjects and just want to hear some perspectives. To emphasize, I don't message anyone. Just these forums. No one to talk to about programming either. Most of my relations are about keeping up pleasantries, I suppose.
So is that the correct path? Just fuck off and do your own thing?
Talking to people is just a distraction... isn't it? I don't know.
I'm about 70% confident that all collaboration is procrastination and helplessness disguised, or something to that effect. Or it's falling in love with the idea of getting something done magically just because you band together (when history clearly shows otherwise).
It's just hard to deal with people and their expectations as well as your own.
The thought of having someone to message brings me a lot of anxiety, frankly. Hate IM/irc.
My current understanding is that thinking, at least at a "high level", is a very social process. Which is why rubber duck debugging exists, why people imagine talking to someone when they want to think through really hard problems, why most ancient philosophy was written as dialogue, and why you posted this thread here. Apparently Descartes was wrong and it is not "I think, therefore I am", but "there are others, therefore I think".
But I have social anxiety too.
There are plenty of programming projects that can be safely divided into multiple parts with multiple workers who develop their own parts simultaneously and independently of one another. A team who can work well as a team can deliver their projects much faster than an individual who burdens the whole development project on their own.
>>3 has a point.
Also it very much depends on if you click with the people you are collaborating with. I tend to hate it most of the time (pointless back and forth for little to no gain but a lot of wasted energy and slow down) but some people are just great to work with (intellectually rewarding, motivating, knowledge sharing).
>>1
If it isn't hard to post on here, why would it be hard to post on irc?
>>2
rubber duck debugging
I always found this silly, but supposedly it helps people.
Never done this before and doubt I ever will.
Programming is not a solo task. It would be the equivalent of sayings its a group task.
Programming just can be simply done in a group or solo.
It is possible to program everything solo, but depending on what you are programming you will have to face more troubles. An example is an operating system one of the hardest things to do.
There are people who work together in groups on Linux kernel and then you see people at osdev who have worked years on their own operating system.
You shouldn't force yourself to do programming alone forever. If programming solo works better than its pretty simple on what to choose.
Each way is going to have its consequences.
I worked at a cowboy shop and everyone had their own projects.
Yet I had to interface with others work through APIs and so on.
If you capsule it nicely, and everyone is on the same page concerning the input/output, you can delegate those parts to anyone slightly competent.
The bigger issue was egos of the bosses and generally feeling superior because they were a big firm, or doing things to beaurocratic and not talking to each other.
Some mathematicians are like Tao or Erdos, others Perelman, or Wiles. I'm personally found of 18th century style correspondents (asynchronous feedback but without full collaboration).
>>7
Are you within an academic setting or do you investigate these mathematicians alone?
>>6
It does seem to be an issue of competence and dealing with ego. Exhausting stuff.
>>5
True, just finding more and more the negatives of dealing with others.
Most software has multiple authors.
Most authors has multiple softwares.
>>7
Yeah thats why i love BBS. It's a whenever type of thing
I've thought about setting up a messageboard on my LAN at home to blogpost to myself about stuff I'm working on. Right now I have a notebook that I write in whenever I'm programming or reading a book of interest, but a private image/textboard would be good for organizing my thoughts I think.
>>13
Why image/textboard and not a text/doc file?
>>14
I tried that before with a bash script that added new entries by topic, but I never stuck with it. I feel like a messageboard would present a better UI for schizoposting to yourself.
>>15
Why don't you post on Twitter instead?
<16
Instagram's better duhh.
Pixelfed is even better.
Yall should try haskell
>>19
For myself, I'm more partial to Scheme than to Haskell. I like how Scheme is homoiconic allowing me to write domain specific languages for myself. This allows me to work in programming as a solo endeavor.
>>20
That's what appealed me to Scheme, too. But after some time of development and several finished projects I've found that I can't live in a debugger any more. So I turned to OCaml then to Haskell.
debug dub
Is programming a solo endeavor?
From Where Lisp Fails: at Turning People into Fungible Cogs. ( http://www.loper-os.org/?p=69 ):
Every innovative work of mankind has been the product of one - sometimes two, rarely three - minds. And never the work of a herd. No mathematical theorem, no enjoyable novel, no work of art of any importance, have ever been produced by a herd. I fail to see why innovative software ought to play by a different set of rules.
I agree very much with the "one, two, maybe three minds" take on this. I like pair programming and close collaboration but only when the total number of individuals involved is small.
>>24
I think the invention part requires only "one, two, maybe three minds", but the phase after that will require a much larger team. For example, writing a novel can be done by only one person, but actually publishing the book requires a much larger bureaucracy (e.g. editors, artists, lawyers, printers, distributors, marketers, etc.). I think it is similar for software. The initial idea and first implementation can be done by one or two people, but it will eventually require a whole herd of people to maintain and improve upon the original work. That herd of people could be composed of highly educated people (in the case of engineers responsible for implementing an inventor's plans), or composed of manual laborers (in the case of lifting stones for building a pyramid), or a mix of both.
>>25
Very interesting analogy. I see what you're saying, on one hand, but on the other: in a world where a novel can be put together by a skeleton team of (not uncoincidentally) three people: the writer, the editor, and the marketer, how many other members of the above "publishing value-add pipeline" can be classified as rent-seekers?
I feel software today is in such a bad state because of the rent-seeking barnacles that are always attracted when large sums of money are involved.
>>26
It's a societal problem that also affects programming. Having worked for large banks and the government for a few years I've come to the conclusion that the majority of office workers in general are not useful. I worked in a processing center for a large bank, and the majority of those jobs could be automated fairly easily. Not "pie in the sky" deep-learning AI neural-nets running on $4MM worth of GPUs, but fairly simple algorithms could make thousands of jobs redundant in this organization. I suspect it's not ignorance or stupidity on management's part in keeping those jobs around, but is done because "job creation" is seen as so important to the economy, so government grants, social status of executives and the tone of media coverage etc are largely derived from employee headcount.
I think this syndrome of rent-seeking will continue until society at-large understands that a large amount of jobs have already been made redundant by technology.
It seems like most good things come from building *interfaces* other people can use rather than directly collaborating. This minimizes the amount of theory you have to communicate and maximizes the opportunity for other people to build on your work.