thevoidzero

joined 1 year ago
[–] thevoidzero 1 points 1 day ago

Depends on circumstances a lot. It's easy if you're in college/work with similar people. Otherwise it might be hard to start, especially if you don't have a lot of free time.

I moved a lot alone and had to make a new social life a lot. during school, for high school, for college, then jobs, then moved country. Except for last one where I knew a few people every other case I had 0 friends carried over. Hardest to have a social life was during the time I was working on jobs as the ability to meet new people decreased a lot.

So basically it is hard when you don't goto college and job where you are forced to spend time with people, but that can also sometimes makes it hard to hangout with the same people outside of work.

So far things that have worked out for me:

  • People with same interest that you randomly meet sometimes.
  • keep your social media connected and when you see stories of people doing things you like strike a conversation about it. Don't force to have full convo, just say your piece about that story and leave it be if it doesn't go any farther. Small talks just sharing some sentences are good starting points. If it happens a few time with same person you might find someone you have common things with.
  • try some group activities that doesn't have to have a lot of talking. Something you can be present there just doing your thing, it could be local recreational sports group, volunteering, library, etc.
  • friends of friends, statistically your friends in average have more friends than you, so just hangout with them in group activities, and try to make new connections. You have to start somewhere.
  • online friends, sometimes it just helps to have people to talk to, careful on who you're hanging out with, but fandoms and such online are good to make friends that you can talk to without responsibilities of maintaining a relationship. It'll help you be more open on sharing your interests.
[–] thevoidzero 2 points 1 day ago

You said you can type in markdown, convert it to PDF with pandoc and you like the results.

Now all you need is an editor that can open two file side by side (anything works here, I use emacs), and needs to auto reload PDF on file change. And a tool that can run your configured command each time markdown file changes (I have my own program for this, but it's a simple bash script as well if you want to write).

Now with those two all you do is write in markdown and every time you save it the command will run, get the pdf and it'll reload the pdf. Even if you don't have the same program to open text and PDF you can just use two with split screen.

[–] thevoidzero 2 points 3 days ago (1 children)

Thank you. I have a family to worry about so can't survive on ramen alone... But I'll look for other countries postdoc and such. I'm a bit scared of moving somewhere I don't know the local language now because of how things are going in the US, if something like this happens and I can't understand their language I'd not even know the dangers.

[–] thevoidzero 4 points 3 days ago

My understanding is this:

It's just the principle of AUR wrappers. Yes they are very useful, but anyone and their uncle can put a package in AUR name it whatever they want as long as it's not taken. AUR wrapper makes it easier to install things without knowing much, but manually searching for something, finding it, and installing it involves conscious choices. Arch cannot be responsible for people installing malware from a software they recommended, that's why it's kept this way intensionally.

Imagine if yay/paru came with the os, or could be installed from pacman, then people would just recommend doing that to new users and then they might just install whatever and break the system a lot more.

[–] thevoidzero 2 points 3 days ago

That's what I thought, but then when arch install fcks up it seems even harder to fix. I ised it because I have been getting new computers so it was easier to run run it. It messed up the SSD in a way, and trying to run it again wouldn't work because it can't find the SSD that it did something to. It took a while to manually fix all that.

Also idk why arch install doesn't have easy way to partition home and root, the default suggestions's root is too small, changing it requires manually making each partition, just take an integer(%) allocated for home and calculate from there.

[–] thevoidzero 2 points 3 days ago (3 children)

Where are you going? I'm currently on last yr of PhD and thinking of leaving, but I don't know if I should abandon the PhD to leave or not. I'd like to finish it, at least do it remotely, but chances of finding work immediately after PhD are slim.

[–] thevoidzero 2 points 4 days ago* (last edited 4 days ago) (1 children)

Integration was this (idk if lemmy renders latex):

int_0^1{x^{-1} (1-x)^{n-1}} dx

[ \int_0^1 {x^{-1} (1-x)^{n-1}}dx ]

Text: finite integration from 0 to 1 of function x to power negative one, (1-x) to the power (n-1).

The limit at 0 goes to infinity that's why there is no solution. But deepseek kept trying different method reaching a conclusion that it can't be solved then then trying different approach.

About the closed form, the function without closed form was that function multiplied by x^y (1-x)^y .

int_0^1{x^{y-1} (1-x)^{n-y-1}} dx

The first one is a case where y=0. Unless y=0 or n, you have integration, just not a closed form. You can plot the function to see it as well. You'd have to try different values of y and n for it to actually plot something though.

[–] thevoidzero 6 points 5 days ago (3 children)

I downloaded it yesterday to test, and my wife was immediately was like ask about it. I had to comply, it said I can't talk about sensitive topic, but in the think it talked a bit.

Anyway, I tested simple sum, good job. I tested an integration that doesn't have closed form solution poor thing was thinking a lot, everytime it goes "this happened which makes the integral impossible" then it goes "maybe another approach" and tries again. Had to ctrl+C, but I'm at least impressed with trying to find solution without giving up, very asian.

[–] thevoidzero 3 points 5 days ago

HIV is the most famous retrovirus. The reason it's hard to cure is basically once it goes from bloodstream to the host cells, it adds itself to the host DNA and RNA, then it just replicates with our cells and that our cell will produce more of it.

[–] thevoidzero 4 points 6 days ago

Yup, considering they deprecated so many functions and removed them I'd imagine switching would be really hard.

Even while writing my new projects in gtk4 (tiny projects) I run into problems of many solutions no longer working because the functions are removed without any replacements.

[–] thevoidzero 31 points 1 week ago (1 children)

I support this if it applies to everyone, not just AI. We should be able to use everything that we see as well

[–] thevoidzero 9 points 1 week ago

I'm surprised it took this long. Journals are broken, authors pay to publish, readers pay to read, reviewers volunteer for free. Then why is this so expensive, when the cost is just online storage??

 

cross-posted from: https://lemmy.world/post/22266569

Looking for Mentor (for a PhD Candidate) that works with open source and Rust

TLDR: Searching for person holding professor position to officially act as a committee member on a US PhD defense

Hi all,

I'm in a non CS field. I'm doing PhD in hydrology and I'm good at Geospatial Analysis, data analysis, visualization, modeling and such. I really like programming and have been making open source programs, contributing to open source programs and such. And have been learning rust for last 2 years.

For my PhD dissertation I'm doing a project where I'll be using Rust to make a program with compiled plugin system that can do generalized river related tasks including data analysis and visualization. I have professors in GIS and hydrology to guide those aspects, but I don't have anyone on software side to ask questions, or to look at my work. I tried emailing some people I have seen with open source projects on GIS+rust, but no response.

I'm ideally looking for someone that holds a professor position for my committee who is good with either rust, GIS related algorithms development, and programming languages. However, it woud also be helpful to just have someone woth knowledge about such things. In either scenario, credit and authorship will be given.

I appreciate any response even telling where i could find someone matching the above description. :)

Edit: I can also provide my previous projects in GitHub, websites and such before you decide in messages.

 

TLDR: Searching for person holding professor position to officially act as a committee member on a US PhD defense

Hi all,

I'm in a non CS field. I'm doing PhD in hydrology and I'm good at Geospatial Analysis, data analysis, visualization, modeling and such. I really like programming and have been making open source programs, contributing to open source programs and such. And have been learning rust for last 2 years.

For my PhD dissertation I'm doing a project where I'll be using Rust to make a program with compiled plugin system that can do generalized river related tasks including data analysis and visualization. I have professors in GIS and hydrology to guide those aspects, but I don't have anyone on software side to ask questions, or to look at my work. I tried emailing some people I have seen with open source projects on GIS+rust, but no response.

I'm ideally looking for someone that holds a professor position for my committee who is good with either rust, GIS related algorithms development, and programming languages. However, it woud also be helpful to just have someone woth knowledge about such things. In either scenario, credit and authorship will be given.

I appreciate any response even telling where i could find someone matching the above description. :)

Edit: I can also provide my previous projects in GitHub, websites and such before you decide in messages.

 

Hi all.

I want to develop a plugin system within my program, and I have a trait that functions defined by plugins should implement.

Currently, my code gets all the functions in a HashMap and then calls them by their name. Problem is, I have to create that hashmap myself by inserting every function myself.

I would really appreciate it if there was a way to say, suppose, all pub members of mod functions:: that implement this trait PluginFunction call register(hashmap) function. So as I add more functions as mod in functions it'll be automatically added on compile.

Pseudocode:

Files:

src/
├── attrs.rs
├── functions
│   ├── attrs.rs
│   ├── export.rs
│   └── render.rs
├── functions.rs
├── lib.rs

Basically, in mod functions I want:

impl AllFunctions{
    pub fn new() -> Self {
       let mut functions_map = HashMap::new();[[
       register_all!(crate::functions::* implementing PluginFunction, &mut functions_map);
       Self { function_map }
  }
}

Right now I'm doing:

impl AllFunctions{
    pub fn new() -> Self {
       let mut functions_map = HashMap::new();[[
       crate::functions::attrs::PrintAttr{}.register(&mut functions_map);
       crate::functions::export::ExportCSV{}.register(&mut functions_map);
       crate::functions::render::RenderText{}.register(&mut functions_map);
       // More as I add more functions
       Self { function_map }
  }
}
 

TLDR: I recently found out there is "deprecated" XFA format that acrobat still uses in their programs, and government forms have those for dynamic contents in the form that we cannot fill using other softwares. Looking for solutions.


This has been a problem since a long time. Back in 2020 I had dual boot because I needed acrobat to fill PDF forms, but after finding xournal++ program I nuked windows partition. Windows update messing up grub was one of the reason I decided to nuke windows and looking at the posts recently it's still a huge issue.

So the problem I recently encountered is that even the government issued PDF forms need acrobat reader (which is free software for PDF, but only available in windows and mac). Which I didn't think would be an issue and just filled the form in Firefox.

Turns out that was problematic as the PDF forms has fields that are automatically filled, calculated from other fields, only made available when certain checkboxes are checked, etc. and Firefox doesn't support that. Even trying to install the acrobat reader snap (which uses wine) in a VM and opening the PDF on it didn't work. The UI makes me think it's a really old version of the reader.

So without searching for other devices (and filling a PDF with my sensitive information) what solution is there? Installing windows is a hassle even in a VM, and it will use up precious SSD memory. But that's the only solution I can think of.

I also found masterpdf or something like that which the Arch wiki says has support for that, but it didn't work. It says XFA forms are converted to acro forms, and the dynamic part doesn't work. There are websites that promise to work for such forms, but I'm not going to be putting sensitive info on web apps.

 

Hi everyone,

I'm hoping there are people here who work on FOSS and have applied for grants to support their software financially. I am applying for a grant opportunity that is asking for a software from US gov agency.

My requirements:

  • I want to publish it under Open Source Licenses like GPL (not MIT) so other corps can't take this to use on their product,
  • The grant agency will get the source code, they can do whatever as long as the license is held,
  • I will develop the features they want, and request during the duration of grant,
  • I will want to continue development independently after the grant, or apply for more grants from other organizations,
  • To clarify the previous point, I do not want to give them the final product so they own it, and I can no longer do anything on the program.

So, if anyone has done similar things, please give me advice on this. Their requirement says "a web repository" should be provided at the end, so I think I can apply with the intention of giving them the software code while keeping the rights. But I don't want to make a mistake in application/contract and lost the rights to the program, I want to develop a lot further than just the features they want for their use case.

Or at least dual license to protect the Open Source Side while giving the grant organization rights to take the code for their other programs because of the money they spent.

 

cross-posted from: https://lemmy.world/post/18129059

This feels like it should already be a feature in a terminal. But I didn't find anything that let me do this efficiently.

I had a rust library for converting list like 1-4,8-10 into vectors, but thought I'd expand it into a command line command as well, as it is really useful when I want to run batch commands in parallel using templates.

I wanted to share it since it might be a useful simple command for many people.

 

This feels like it should already be a feature in a terminal. But I didn't find anything that let me do this efficiently.

I had a rust library for converting list like 1-4,8-10 into vectors, but thought I'd expand it into a command line command as well, as it is really useful when I want to run batch commands in parallel using templates.

I wanted to share it since it might be a useful simple command for many people.

 

cross-posted from: https://lemmy.world/post/17984566

Hi all,

mpv communities seem to be tiny in lemmy, so I'm sharing it here.

This is a program I made for music control from local network.

You can run it in a computer with some local media files, or youtube links or any other links yt-dlp supports. And then with the server, you can control the media player and the playlist from any devices in your local network. So that you can just show a QR code or something to house guests for parties, or have it bookmarked within family to control the music.

I wanted to make something similar to how youtube app let's you play in TV and such, but my skills were not enough to do that. So I tried a simple alternative that works with computers. In an ideal world, I could make "Play with local mpv server" option come while on other android apps, but I have zero experience in android app development and it looks complicated.

I know some other programs also give option to control media, but I wanted to give it a go with a simple implementation. Making the web-server was a tricky part. Only tutorial from the rust book was useful here as every other web server developement in rust seems to be async ones using libraries so I would have to make a complicated system to communicate with the mpv. Using the simple Tcp connection let me make a thread with mpv instance in the scope. I do need to support https and file uploads and other things, but I haven't had any luck finding a solution that works with simple Tcp connection like in the tutorial. Let me know if you know anything.

Github: https://github.com/Atreyagaurav/local-mpv

 

Hi all,

mpv communities seem to be tiny in lemmy, so I'm sharing it here.

This is a program I made for music control from local network.

You can run it in a computer with some local media files, or youtube links or any other links yt-dlp supports. And then with the server, you can control the media player and the playlist from any devices in your local network. So that you can just show a QR code or something to house guests for parties, or have it bookmarked within family to control the music.

I wanted to make something similar to how youtube app let's you play in TV and such, but my skills were not enough to do that. So I tried a simple alternative that works with computers. In an ideal world, I could make "Play with local mpv server" option come while on other android apps, but I have zero experience in android app development and it looks complicated.

I know some other programs also give option to control media, but I wanted to give it a go with a simple implementation. Making the web-server was a tricky part. Only tutorial from the rust book was useful here as every other web server developement in rust seems to be async ones using libraries so I would have to make a complicated system to communicate with the mpv. Using the simple Tcp connection let me make a thread with mpv instance in the scope. I do need to support https and file uploads and other things, but I haven't had any luck finding a solution that works with simple Tcp connection like in the tutorial. Let me know if you know anything.

Github: https://github.com/Atreyagaurav/local-mpv

 

cross-posted from: https://lemmy.world/post/11271385

Basically, you can choose some slides from an opened .tex file to copy. It also has the function to see which graphics files are included in the selected files, so you know which ones to copy.

Here is the Github link: https://github.com/Atreyagaurav/beamer-quickie

The PDF pages are shown using the SyncTeX (if available) so that you can visually choose the slides as long as there is a single .tex source file, (might still work without synctex for simple cases).

I've made it on Linux, so it hasn't been tested in windows. You probably will need to compile gtk on Windows if you want to make it work. So if someone is really interested let me know, I can give instructions. Even in linux you'll need to install dependencies.

 

Basically, you can choose some slides from an opened .tex file to copy. It also has the function to see which graphics files are included in the selected files, so you know which ones to copy.

Here is the Github link: https://github.com/Atreyagaurav/beamer-quickie

The PDF pages are shown using the SyncTeX (if available) so that you can visually choose the slides as long as there is a single .tex source file, (might still work without synctex for simple cases).

I've made it on Linux, so it hasn't been tested in windows. You probably will need to compile gtk on Windows if you want to make it work. So if someone is really interested let me know, I can give instructions. Even in linux you'll need to install dependencies.

view more: next ›