« first day (284 days earlier)      last day (445 days later) » 

00:42
@mathcat mathcat is an old man confirmed
Also you have your pronouns as "hey/they" in your CGCC bio :p
Hey!
 
1 hour later…
02:09
just had the wonderful experience of rebooting my Pi and having Cinnamon decide to reset all my UI settings for absolutely no reason
I love Linux I love Linux I love Linux
02:23
Burn Linux and Windows to the ground. In their ashes, like a phoenix whose mother was exposed to Chernobyl's bathwater, a new operating system shall rise, which we shall slay. Its blood, impure and anemic, shall be spilt upon the sacred texts of Unix and x86, and on each of the following days we shall speak and bring forth the OSI layers one by one.
But yeah in all seriousness there is no recovering from our current condition
We need a new OS, and it needs a strong, well funded, and benevolent backer who can persuade all the major hardware and software vendors to support it, and sell it to enterprise customers
Plan9!
Windows is doomed to continue sucking as it has little incentive to be good. MacOS isn't really a competitor due to its closed-garden-ness and hardware-specific-ness. Linux is, despite being better than Windows, doomed to eternal bugginess and irrelevance
The fact that Windows now has WSL might make it even less likely that people'll use a full Linux distro for their desktops
@user YES
Distros like Ubuntu do "just work" (most of the time)
02:29
Yup. Linux is gonna end up more like Adobe Flash For Nerds than a real OS
Eh, I wouldn't be that pessimistic
It'll probably live on for a long time, just won't be as popular as Windows or MacOS
Aside from the enterprise/server market, where it's more successful for obvious reasons
@user Oh yeah no, I don't mean that it'll die like Flash did
Flash isn't dead
It has been born again as Ruffle
I mean it'll end up being this add-on sorta environment thing present on every OS that's kinda weird but works
I wonder if we'll get a new (popular) OS within our lifetimes
02:31
Like Windows has WSL, MacOS probably has an equivalent, Linux is Linux, ChromeOS is Linux and exposes a Linux container
It feels like computer-related advances have kinda slowed down :(
@user From the ground up? My guess is no
idk if I'm just imagining it though
Unless I happen to become a tech billionaire
In which case I'm 100% doing it for the good of humanity
@RydwolfPrograms Depending on how low the ground is
Every OS out there probably has tons of cruft accumulated over the years
02:32
The most probable thing I can think of is some big company sponsoring a Linux flavor, getting a ton of sway on changes to the kernel itself after cornering a bunch of market share starting in the enterprise environment, and it slowly improving from there
@RydwolfPrograms Having money isn't enough to ensure that you make a good OS and that people adopt it
Yeah it is
It's a good start, but there's probably a lot of luck involved
You could spend a whole lot of money and hire a lot of smart people and still end up with something crappy
And people won't just switch to another OS like that, there needs to be a very good reason for them to make such a huge change
Having money means you can afford smart people. Look at Elon Musk. And having money means you can bully hardware and software vendors into supporting it, take the owners of big companies out to expensive dinners and convince them to switch their whole company's tech stack to it, and pay for ads showing off just how convenient this newfangled OS thingy is gonna make the ordinary Joe's life
Tesla worked out great, but Musk's somehow driving Twitter into the ground
02:34
@user How about "it's entirely free to switch, you get to keep all your files, you can go back at literally any time, and it's not only 50% faster, but way more customizable and has exclusive features all your friends are using"
If you don't know how to use that money properly, you're not going to do much
@RydwolfPrograms That's a lot of features
Throwing money at a problem isn't going to get you all of that in a reasonable amount of time
I'm fairly opinionated about OS's already, and smart enough to listen to people smarter than I am
If you try to aim for a goal that high, your project might just get stuck in development hell
I'm not gonna do a Musktwitter
tbf it's hard to screw up that badly
I would've screwed Twitter up if I'd taken it over too, but not so efficiently
I think I'm going to uninstall the typing userscript now, it scares me to see someone typing for a long time
Ahhhh it's going to be a 2000-word essay that totally destroys my argument and me and leaves me a crying wreck isn't it
02:39
1. Entirely free to switch: the enterprise customers will be where the money comes from
2. You get to keep all your files: Not too hard, actually. I bet a half dozen experienced coders could figure out all the edge cases in a month or two
3. You can go back any time: Also not too hard, just either store information needed to reconstruct their original filesystem on disk, or in the cloud
4. Faster: I'm sure Windows spends a lot of time doing things it really doesn't need to be doing. 50% is probably unreasonable but removing three decades of backward compatibility and tech debt can't hurt
Plus you could get so much additional security and privacy stuff done right if you started from the beginning
Is this hypothetical OS going to be rewritten completely from scratch?
Because Windows may have tech debt, but there's also changes they've made after learning lessons over the years
And it'll be designed with near-costless forward compatibility in mind. It needs to be designed around bulletproof interfaces that will last ages, so that individual parts can be easily torn out and reworked over time.
A completely redesigned OS will probably be buggy af for a long time
Where are you going to get the money to fund this? It's going to be in development for a long time
02:43
Nah, not if it's written in Rust, and you have one of those things Microsoft has, a room with thousands of different computers crossing different platforms and stuff to run week-long automated test suites before every release
It'd be cheaper to keep Linux servers going than switching to this new, relatively untested thing
And if some new tech advancement happens right then, you might have to deal with that
I don't think a company would bother investing that much money into a project that would take so many resources, with so much risk
@user Maybe, until my team of a couple dozen cybersecurity researchers starts finding bugs in major web server software, some proprietary drivers, maybe the Linux kernel itself. Good marketing material; look how risky it is to run Linux! If that had been the bad guys who found it, you woulda just lost millions in a lawsuit!
Feels more likely that some niche thing from academia slowly becomes more popular among nerds and then among regular people
@RydwolfPrograms I mean, you don't need your own team of researchers for that, there's already bugs in the kernel and other software that people find occasionally
But if you make something new, it's even more likely to have bugs that no one's thought of before
Sure, but pretty slowly, and they don't tend to be too newsworthy
@user Not really...most major bugs in that sorta stuff could be prevented with Rust, modern testing, etc.
You could even have the important parts use that correctness proof thing
Writing bulletproof code is way easier than it was in what, the 60s
Formal verification would be cool, yeah
@RydwolfPrograms Can Rust stop stuff like Spectre?
If so, that's really cool
02:48
No, that's hardware-level
That's not a bug in OSs or software, that's a bug in the CPUs themselves
Ah, ok
@RydwolfPrograms Still a little doubtful about this but don't know enough to comment more
(Which I think is gonna be more and more common with time)
Well, maybe not more common, but not any less
@user That's why I said, when I'm a billionaire :p
The expensive part isn't the development tho
That's half a dozen nerds and a whiteboard
It's all the paying off companies to support it, the marketing, the stuff that actually gets it adopted
Anyway, I got homework to do, have fun with y'all's lame operating systems
@RydwolfPrograms only real problem is a lack of available applications
If you make an OS from scratch, there won't be any applications available that you haven't already written
And there won't be any language interpreters or compilers available except for the one you use to build the OS with
Meaning you'd need to make something that runs on Linux to even have a chance of getting anywhere early on
Because if you ain't go no things I can use on my computer, then ain't no point using it
> And having money means you can bully hardware and software vendors into supporting it
But they have to write their applications from scratch
03:03
Trust me, I'm aware hardware/software support is the most important factor
@lyxal Well, they need to port them. Expensive, but that's why the budget would need to be so large
For the app owner it's a win win; their devs get more experience, their product works on more platforms, and it's entirely free for them
@RydwolfPrograms You'd want to guarantee a decent marketshare then
Because sure, you can bully them with money, but if there's no audience there, they'll quickly say "not really worth it"
There'll be an audience, I'll guarantee that
How? Your new OS doesn't have any major applications yet. At best, you're creating something that might get a bit of hype, a few reviews and then fall into obscurity because "oh what's this? important app I use isn't supported yet. Guess I'll have to wait a while until it becomes more mature"
Well the OS wouldn't be made public until after the apps are ready
And I'd get OEMs to sell machines with the OS preinstalled instead of Windows
And there'd almost certainly be some pretty advanced WSL type stuff for both Windows and Linux
As a fallback for less used apps people still need that haven't been ported
@RydwolfPrograms how do you know which apps to port? And what if the teams that originally made the apps don't want to work on the port because they either a) want to move on to something else or b) they don't work there any more or c) there really isn't the best documentation
03:09
I buy them out
It'll happen a few times, call it an investment, take some VCs out to a nice dinner if it'd be too pricy
If I'm a billionaire I'm committing to it :p
@RydwolfPrograms looks like you'll need to rewrite a lot of windows dlls then
> Also, there are licensing and practicality issues with including Windows DLLs.
Oh yeah no way am I doing it like that
So how do you plan to get .exes to work
As well as whatever the hell file format Mac uses
(.pkg I think it is)
It'd be similar to WINE probably, or more like a whole VM. It'd require some complex legal stuff probably, but for anyone with Windows installed prior to installing the new OS, it'd just use that
I don't think Apple would be too happy with you allowing their executables to be run on your own OS
They'd probably launch legal action too
03:14
Who said anything about Apple
Mac doesn't have many platform specific apps at all
And too small and dedicated of a market share
"hey what's this new OS oh wait the apple only apps I rely upon aren't supported guess I'm not switching"
Linux is trivial and Windows would be a VM that's been vivisected, filled with API endpoints, given the old install's license key, and tucked neatly away
@lyxal Mac is not a competitor
People who use Mac already have everything they need. It's a perfectly fine OS.
And one with plenty of rabidly loyal users, and not many users overall
@RydwolfPrograms this works fine until Microsoft buys those OEMs back
I think the play is to play a longer game
Once the enterprise market has been taken care of, and a few percent of normals are locked in by being early adopters or buying OEM machines, it's a matter of time
If you have the choice between Windows and the new thing, that also costs less (OEM Windows licenses still cost a ton), and is the cool new thing that works so much better, it's an obvious choice for at least a fraction of buyers, over and over again, until Windows erodes enough to make it a fair competition, where the new thing will beat it handily
I think a better strat would be to develop an OS for a market that is relatively uncontested (or even a new market that didn't exist before) and expand from there. Let all the apps be developed for a different market than the existing companies already control and then scale up with interoperability between platforms
03:22
Enterprise, there ya go
Build up the foundations and then strike
Most servers still use Linux
Oh dang that's right - legacy systems
Yeah good luck getting people to replace those
It could also be designed with IoT and embedded stuff in mind
Legacy systems are already in short demand for people who can support them
03:24
@lyxal Nah, you don't get them to upgrade them, you wait for them to finally get too old, then present them a much better solution than the competitor
Bold of you to assume that's how companies with legacy systems work
Everything gets too legacy at some point
It doesn't need to be everyone at once, just a steady trickle
Maybe, but i still see profressions using Windows xp for things
And Windows XP is getting a little old. When a super cheap super effective new thing comes along, that might be just the spark
End user development is a pain for adoption of a new system. Because people who make their own stuff on their own systems won't want to move their stuff around to a new system if they can just patch this or have a workaround to something
I think the best thing to do for a new OS would be Linux clone that is gradually and eventually rewritten into something proprietary.
Also an idea I've had for a while is to make automatic translation of Android apps into usable desktop applications
So have a research team figure out how to transpile Java apps into whatever language your os uses
At least, that's what i've always thought the plan would be
Something gradual where the competition won't know you've taken over until it's too late
03:53
Just become dictator of the world, make everyone use your new OS, then allow yourself to be violently executed in a coup
(third part is optional but recommended, because dictators are bad)
Coups are usually bad too, especially the ones that violently execute people
 
3 hours later…
07:06
@Ginger welcome to the club!
07:36
I was out shopping today and saw Popcorners on a shelf
And I couldn't help but internally laugh because of all the okbc memes about Popcorners after the super bowl ad
i still haven't fully processed the fact that that ad actually happened
We're going to eat a lot of snacks together
@lyxal Which ad was that?
(also, TIL Australians watch the Super Bowel)
Why are people so obsessed over this "superb owl"? It's just a bird.
4
> I am the one who snacks
08:14
@user I don't
@user the Popcorners ad
It was: a) all over okbc before the super ball even happened, b) in my recommendations because I've watched so many BB/BCS clips that YouTube said "hey this is content you'll like", c) it popped up in meme clips and d) the official breaking bad YouTube channel linked to it
Actually, d never happened
Thought it did
 
3 hours later…
11:17
@RydwolfPrograms ಠ^ಠ
12:07
@RydwolfPrograms Did you mean: Canonical?
 
1 hour later…
13:09
I forgot this existed
I'm glad I remembered
13:24
@lyxal problem: android apps are compiled for the dalvik vm, desktop for the jvm
13:39
@Ginger
See the above image
what
It's a hat
yes
Wanna know what the style is called?
big shot
13:40
Kromer
Unironically
me when [Number 1 Rated Headwear1997]
TIL that .museum is a valid TLD
 
1 hour later…
14:56
Trying to decide what classes to take next year
So far I have:
- AP English IV
- AP Statistics
- AP Government
- AP Macroeconomics
- AP Psychology
- Computer Science Practicum
- Engineering Science
Which leaves me with one empty space
I could just do an off period I guess
I know I should take AP Physics 2, so I'll be more prepared for the Physics C exam I'm taking either way, but physics class sucks and I'm tired of taking classes I don't enjoy
CSP is basically gonna be two off periods tho
I don't know if I need a third lol
That might be too chill
Maybe I'll get overcommitment withdrawal and get so productive I get a heart attack
I'm at the point this year where a backlog of 15 overdue assignments is progress lmao
15:23
mm yes thanks edge
2 hours ago, by Ginger
me when [Number 1 Rated Headwear1997]
That is ^^^^^^^^^^^^^^^^^^^^^
From the position of your screenshot message ofc
16:25
Edge doing some big brain AI stuff.
Must've liked the kromer.
 
7 hours later…
23:27
@RydwolfPrograms yeah i think you want to do another off period
Yeah but I do have 8 college level classes this year
Not counting CS
Whereas next year it'll be a much lighter workload regardless
depending on how you're approaching college applications that could also be a factor for the first semester
I wish I hadn't taken a bunch of required non-AP classes in 9th grade to get them out of the way
Since they lowered my GPA, and if I'd done that next year instead of at the start, my class rank would be higher at the time I'm applying
Although I did go from 30 to 18 just in the fall semester of this year (every single one of my classes this year is 6.0-weighted), so extrapolating linearly maybe I'll be top 10 by summer
Wish we had more granular information than semesterly
Being 1st or 2nd would be pog but I probably won't be able to since I don't literally kill myself trying to do every extra credit and makeup thing possible

« first day (284 days earlier)      last day (445 days later) »