If you think that, this one sucks compared to my other one as far as free speech goes.

A live chat for every page on the internet

ok, it's a little pointless seeing how there aren't enough people on it to populate a live chat for every page on the internet. But the idea is there.

Is it scary that I kind of see some of him in me?

I'm even trying to build my own defunct OS but in javascript-ish instead of c-ish.

Please stop this meme. There are three things that run the world, business, law, and code.

Do not put these people at the center of code. And because they won't be able to really code they will be doing the soft jobs of coding, aka, being in charge.

You don't know how easily it could be modified to do that. WebRTC was basically built for that.

I had sending each other screen captures to be one of the embed types as a goal. The way you do that, turning that into chaturbate is basically one line different.

Also it supports video. You just have to click your username to the left of the compose line. It only supports video targeting the side panel at the moment.

Well I've definitely made some improvements. There is now an #import statement. The key is to not mind using more than one user.

I just need to set it up so you can browse other user's code. If you check back in a bit there will be a links panel to the left that will help you find things to import from and faq and other things. Right now the getscript user is good to pull from. It allows you to pull any script off the web and include it in your project with a callback once it's loaded.

#import getscript
function start () {
 //Import pdf.js
 getscript('https://cdnjs.cloudflare.com/ajax/libs/pdf.js/2.0.943/pdf.min.js',()=>{
  //some code
 });
}

Also getscripts takes an array of urls and is inside the getscript user.

The monaco user lets you insert a highly capable text editor into the browser.

You can see it at work by putting notebook in your build and running startNotebook() in the init.

Also monaco is the text editor you see when you edit code.

The obvious thing to edit is the appearance. Go ahead and give it a try.

Just rewrite init_chat_ui and then include your username in the build and run it.

Bonus points if you split apart the css and html in your version.

I need to make it easier for users to read each other's code though, because that's kind of the point.

Here is the chat client's code, for reference if you want to change any function in your globals:
https://pastebin.com/Q5gynyUT
Here is the chat server's code (run's front end):
https://pastebin.com/W06L3J6y

It could be built far more serverless (no front end server peers) but this is what I wrote in two hours as a demo.

If it doesn't run the server probably needs to be run.
user:chat-server
pass:chat-server
and run it. I have the code backed up in case someone fucks with it.

The other thing I like about it is format. I like knowing the percent change because I want to understand trends.

But it also says

The ratings are based on the number of skilled engineers world-wide, courses and third party vendors

Of the four or so indexes I've found they are the only one to claim any connection to number of professionals using it. All the rest also factor in search engines except the github only indexes which I trust even less.

It would be nice to be able to scale the factors of the index.

Edit: Well fuck. That really is an unsubtanciated claim on there. https://www.tiobe.com/tiobe-index/programming-languages-definition/
I feel soiled.

There is one that is search engine and stack overflow (which is which language sucks so bad or has such shitty programmers they spend half their time in a search engine).

Tiobe claims that it is based on the number of people employed doing it. So they probably just give people questionairs asking what languages they use at work.

I liked C when I used to code in it exclusively.

I've writen a ranking based on size-growth fronteer. C still wins over Java because it is growing so much faster and I fucking hate Java.

1. java, c
2. c++
3. python, sql (sql is apperently turing complete)
4. c#, visual basic .net
5. php
6. javascript
7. ruby, visual basic (not sure what makes one .net)
8. delphi/object pascal, R
9. assembly, PL/SQL (also not sure what that is specifically)
10. swift, matlab
11. perl, go
12. objective-c

So one thing I would like to do is make C development closer to javascript (I do node). I'd like to see a cpm equivalent to npm. Conan.io is nowhere close. And I'd like to see more popular http libraries that either take advantage of theading or hyper-threads/async.

I'm wondering if someone were to just rip npm and give it a new default public repository and host npmjs.org at cpmjs.org if the end result would be at all useful in the context of c.

Nope. I figured it out. It was mostly a glitch on why it wasn't working easily. You just add a <track> element with a link to the subtitles.

Here's the deal with the glitch. The subtitles need to be on the same domain. It doesn't matter if you have permissive cross domain set. For this it has to be on the same domain. The real glitch is that Firefox doesn't tell you that's a problem while chrome does tell you what's wrong.

But on firefox you can just use a dataurl instead of a regular url.

Compared to your solution it saves you from constantly reading the time from the player and associating it with a position in a subs file which you would have to parse yourself.

Way easier to just to link to it if you control the site or inject the <track> with jquery on a site you don't control.

Thanks for giving your thoughts though.

THE COMPLETED RESULT:

Paste this into a browser on cytu.be and you get subs for Boondocks:
https://venus.i.ipfs.io/ipfs/QmPw7u4LkByUMPkcLbA9LLHiA8qGNannLfL1n4qfY5tCLT

Here is a working example inside Firefox:
https://www.iandevlin.com/html5test/webvtt/html5-video-webvtt-sample.html

<track label="English" kind="subtitles" srclang="en" src="captions/vtt/sintel-en.vtt" default>

<track label="English" kind="subtitles" srclang="en" src="https://venus.i.ipfs.io/ipfs/QmTWJQxaPn6TMKfZugH4DqpbMp23otMcvgEXzZTUFeoFLh/boondock.vtt" default>

Folder for webVTT subs: https://venus.i.ipfs.io/ipfs/QmTWJQxaPn6TMKfZugH4DqpbMp23otMcvgEXzZTUFeoFLh

Currently uploaded Boondock Saints (Folder): https://venus.i.ipfs.io/ipfs/QmeQFtHC8K9DiM8ZybxFJgfZVk3skfS261Azk7uou8VLBi

Our Working Files:

Convert SRT files to webVTT
http://www.webvtt.org/

The downside of this link is that it assumes custom controls (as in our own). Well video.js already does that.

I know from doing this earlier this earlier that video.js does show the CC menu. You can select the item which didn't seem to toggle them on even in the javascript, but you can toggle the track on in the javascript. This changes the state of the track but doesn't appear.

^ That is where the extra work comes in. Unfortunately this link doesn't seem to help me close the gap between the video.js controls and the javascript.

BTW, why not look up how to do it in video.js? Because video.js has multiple versions of out dated documentation floating around and I said fuck it. It's still an html5 player and by polymorphism any JS that puts subs onto html5 video should work on that player.

This is the best source: https://developer.mozilla.org/en-US/Apps/Fundamentals/Audio_and_video_delivery/Adding_captions_and_subtitles_to_HTML5_video

You add a <track> to the video and you have to do some other modifications to the video player besides.

I've gone down this route before and it didn't seem to work specifically because of the extra things that have to be done. The point being that video.js doesn't make <track> work out of the box.

This one get's it kind of close. Does it completeling avoiding any html5-video property.

http://v2v.cc/~j/jquery.srt/

Parse srt files into an array: https://www.snip2code.com/Snippet/83321/Simple-SRT-parser-in-JavaScript