/tech/ - Technology★

>>1019323 (OP)

You can thank Pajeet for this shit.

>good God the 100+GB games,

This one kills me the most, I don't really care about ram, the biggest framerates but huge waste of disk space is a real kick in the nuts

>>1019323 (OP)

>200MB for a firefox browser install.

Of all the things wrong with browsers, you complain about this? 200MB for a complex tool is absolutely acceptable, what's not acceptable is individual apps all shipping with their own copy of chromium.

>good God the 100+GB games

That's a good complaint, lazy devs not compressing audio and textures at all should be shot.

> Not even feature filled fleshed out browser, just a proof of concept, minimal browser that can handle a modern web page without the drawbacks mentioned formerly.

The standards are incredibly complicated, and the situation is not helped by google breaking them intentionally to sabotage non-chrome browsers.

On top of that, the standards are very hard to implement progressively due to the cascading nature of CSS and a bunch of other quirks.

>>1019443

>what's not acceptable is individual apps all shipping with their own copy of chromium.

And why not? How much more bloated is chromium compared to windows, gtk, or qt? HTML+CSS is so vastly superior to every other GUI framework ever devised it's not even a question. Why would any new app be made in anything else?

>>1019445

>And why not?

Because web browsers are complex tools, not small libraries.

Including an entire browser is like shipping your program as a bootable disk with a complete copy of Ubuntu that has your software installed in it.

>How much more bloated is chromium compared to windows, gtk, or qt?

A lot more, because it's much more general purpose than "UI framework".

It's like asking how much more bloated a full Linux install is compared to notepad.exe.

>HTML+CSS is so vastly superior to every other GUI framework

Doubt.tiff

Any sane GUI framework does not have cascading style rules as the default.

Also: no major differences between implementations, no external forces changing the standard daily, and decent backwards and forward compatibility.

Even if you decided to use web tech for the UI, you could always open a web page and let the user interact there, it's a pretty common thing already with routers and local servers.

>>1019445

Found pajeet.

its not dead, its in the "do it as fast and dirty you can"

>>1019441

Game assets take a lot of space dude. It's not just code or libraries.

>>1019486

You know you can compress stuff right? Right?

>>1019427

You are welcome.

>>1019323 (OP)

>Are the standards so fucked no one can write a functional modern browser?

Yes. Recall that Bank of America, Boeing, MasterCard, PayPal, and Wells Fargo are all members of w3c. Thanks to "coder camps" and "front end development" establishing abuse of Javashit web APIs and Google/Amazon's CDNs as common practice (and what's more, "computer science", the modern web is being used for things that it was never initially intended for: think web crypto, web VR, webasm, webGL, web audio, canvas. Most sane (and humane) people would think these would be much more suited to a standalone application where you don't have to suffer from the disadvantages of having to provide a retard-friendly API for a massive variety of use cases. The end goal, funnily enough, seems to be replacing the entire concept of the computer with the browser, except it's programmed in Javashit and it runs as interpreted Javashit code on top of C/C++ code.

The bloat is within the standard. Even most modern browsers, apart from (((Chrome))), aren't compliant. I'd like to see you write a <200MB browser implementing everything in it. The only way to escape it is to be out of compliance with the standard. Links, for example, doesn't have to deal with pictures, video, fancy animations, or video games and audio synthesizers running inside the browser, so it can be a 4MB executable.

>>1019443

>>1019488

>lazy devs not compressing audio and textures at all

Have you heard of space-time tradeoff? You do understand that decompressing audio and textures (which are not small) in realtime also incurs performance penalties, right? Right?

>>1019323 (OP)

nice hat ;)

>>1019443

"lazy devs not compressing audio and textures at all should be shot."

wow, you're homocidally ignorant..

I haven't bothered with any game that ships larger than 10GB, but still know that uncompressed audio and video can be utilized without needing to DECOMPRESS it (which uses CPU & takes time). I guess those devs assumed that in the age of 4TB personal HDDs you wouldn't mind trading some disk space for better speed (less compression = less lag)

>>1019488

you know compression has limits right? Right?

and graphics and sound do not compress very well, plus to then decompress them takes CPU as I said above..

>>1019441

>priorities

Pretty much every CPU being used by almost every human is backdoor'd by israelis. That is a real kick in the nuts.Do something about it besides playing more games, cuck. take what you learn in those games and BE A REAL FUCKING HERO, SAVE THE WORLD, DEFEAT THE ZOG or take the blue pill and keep being a cuck.

>>1019493

stop trying to scam us. I don't have a wirus and don't need technical support, same with my grandma & her cat. and no we're also not interested in a free cruise, or extended car warranty.

>>1019497

>Have you heard of space-time tradeoff?

Of course, enough to know they can't be extended indefinitely: if you use larger files, I/O time goes up and eventually overtakes any gain in processing times.

You also have disk access times further slowing things down, especially when loading thousands of loose files.

For a practical example, see .bsa vs loose files in any Bethesda game, the archive is much faster to load as decompression is orders of magnitude faster than disk reads.

>>1019499

>and graphics and sound do not compress very well

kill yourself pajeet

I made a small addon myself that basically lets me toggle javascript on and off via hotkey in my browser. I noticed that websites that really rely on javascript to even display anything are usually garbage anyways and might as well not be visited. You can find lots of the good content still on simple .html pages with no scripting going on at all.

If somebody would write a browser that rendered well made pages correctly and doesn't even try to tackle javascript, without being some autistic console/curses abomination, (or being old and abandoned) I'd start using it immediately.

>>1019502

>kill yourself pajeet

what a shining example of quality disc- oh wait, no.. actually, it's just cancerous shit again.. ya that's why I don't usually chan

suicide? never

indian DNA? nope

try harder at having a purpose, oxygen pirate.

>>1019505

as much quality discussion as the claim that images and sound don't compress well

>>1019505

>oxygen pirate

Fuck off.

>>1019502

>they can't be extended indefinitely

Sure, but compressing the shit out of something using an expensive algorithm so you can halve the size of your game asset folder isn't necessarily optimal.

>>1019528

Of course, but the recent trend is more towards "lmao let's have 20 GB out of 30 as uncompressed audio files" a la Titanfall rather than hyper-compression.

Hell, even the transparent compression in windows 10 can literally halve the size of many games without a noticeable overhead, see

https://github.com/ImminentFate/CompactGUI/wiki

On top of that, having expensive compression but cheap decompression is usually acceptable as there is usually no need to compress the files on the user systems instead of giving them a compressed copy.

>>1019505

>oxygen pirate

Didn't Ephraim Kishon write some satirical short story about Jews trying to tax breathing at one point?

my god, you really are a bunch of autists. Not even the fun can-guess-the-correct-amount-of-rice-in-a-jar kind, more the clawing-at-own-arms-pissing-themselves-screeching-in-a-corner-kind, but instead of just screeching random gibberish, it's screeching about how jews ruined everything. Jesus.

>>1019513

>oxygen pirate confirmed

lol, triggered?

people who fail at logical responses and resort to the asolute trash of arguments, explitives and "kys" suggestions, are themselves the BLOATWARE OF HUMANITY, to bring this full circle and hopefully turn the sliding back into on topic discussion.

>>1019534

it won't be satire if you allow your environment to be poisoned to the point that your children are forced into one of their dome cities in order to survive.

>>1019537

they didn't ruin everything, but they are far from innocent. actually read their shit, like the talmud, its disgusting. they condone slavery, and rape, even of children, as long as its done to a non jew. they are the ultimate racial supremacists - "GOD'S CHOSEN PEOPLE" think about it, that's not based on behavior, it's bloodline. you call on Jesus so I guess you're christian? Jews (pharisees whom he called out for their corrupt bullshit) had Jesus killed. relevant since we're in /tech/ look into Talpiot and unit 8200, they DID backdoor everything, and had an easier time doing so after terrorism was an excuse for shit, after they had a part in the 9/11 attacks (5 israeli mossad agents were arrested for celebrating the attack, in a van with explosive residue; no arabs arrested for dancing that day).

there's alot more to it but hopefully that opens your eyes to the fact that while it's not PC to say so, the terrorist nation of israel is a serious problem that needs to be dealt with. I am tempted not to get in the way of the luciferian plan because they promise to eliminate zionism and islam in the same struggle. there's more going on in the world than what graphics card you're using to run their psychological conditioning tools called 'games'

>>1019546

>>1019537

You two, get a room and fuck off.

>>1019488

Most image file format have some kind of compression, same for audio file formats.

>>1019513

>>1019537

>>1019559

(((jews)))

Bloatware is another problem caused by C and UNIX and the minimalist philosophy. Mainframes in the 60s and 70s placed a lot of emphasis on compatibility between programs written in different languages, but that requires using your brain and thinking about the common types used by programming languages. PL/I was designed to be compatible with Fortran, Cobol, Algol, and assembly programs and it includes a superset of all the types supported by those languages. Multics and VMS have data descriptors and common data structures to share data between programs written in any language, based mainly on PL/I. On these systems, the choice of programming language is about you, the programmer, but on UNIX, programming languages are about other people. There is no way to combine programs written in different languages other than shitty C "FFI" bullshit that doesn't even cover basic data types like arrays and strings. This leads to huge duplication in code and extreme slowdowns. You can't use Perl libraries from Python or JavaScript libraries from Java, and it gets worse. My computer has at least 10 garbage collectors and none of them are compatible. Some of them have their own schedulers and audio libraries, like operating systems on top of your OS. A Lisp machine has only one garbage collector, so there is no duplication in memory management. A Lisp machine has only one metaobject system, so objects can be combined even if they have totally different metaclasses. Every language gets bignums for free (only slower if there's an overflow) and a single address space that is shared between all programs, so you can just share an actual data structure instead of pipes and XML and all that bullshit.

The other flaw of minimalism is even worse. You can no longer depend on any library or command being present on the computer, which leads to every program having to reinvent wheels. Even simple string functions are "reinvented" by C programs when they should only have to exist one time in a library or the compiler's built-in functions. C compilers can't even do basic error checking that BASIC compilers and interpreters did. This "philosophy" is at the root of what makes C and UNIX suck. It leads to autoconf and all that "conditional" bullshit browsers have to do to avoid incompatibility. These web pages need recent browsers to even view them at all and at the same time need all this bloated library code. Electron putting multiple copies of Chromium on a computer that already has one or more web browsers is another example of the UNIX philosophy. A maximalist design would not only not need any of this bullshit, it would also integrate "web" technology and interfaces with native GUI design using the object system. The web as we think of it with "apps" and all that crap probably wouldn't even exist on Lisp machines because the native GUI is more powerful and easier to use (JavaScript needs bloated frameworks to do what Lisp machines could do without them).

There are many reasons why GNU Emacs is as big as it is
while its original ITS counterpart was much smaller:

- C is a horrible language in which to implement such things
as a Lisp interpreter and an interactive program.  In
particular any program that wants to be careful not to crash
(and dump core) in the presence of errors has to become
bloated because it has to check everywhere.  A reasonable
condition system would reduce the size of the code.

- Unix is a horrible operating system for which to write an
Emacs-like editor because it does not provide adequate
support for anything except trivial "Hello world" programs.
In particular, there is no standard good way (or even any in
many variants) to control your virtual memory sharing
properties.

- Unix presents such a poor interaction environment to users
(the various shells are pitiful) that GNU Emacs has had to
import a lot of the functionality that a minimally adequate
"shell" would provide.  Many programmers at TLA never
directly interact with the shell, GNU Emacs IS their shell,
because it is the only adequate choice, and isolates them
from the various Unix (and even OS) variants.

Don't complain about TLA programs vs. Unix.  The typical
workstation Unix requires 3 - 6 Mb just for the kernel, and
provides less functionality (at the OS level) than the OSs
of yesteryear.  It is not surprising that programs that ran
on adequate amounts of memory under those OSs have to
reimplement some of the functionality that Unix has never
provided.

What is Unix doing with all that memory?  No, don't answer,
I know, it is all those pre-allocated fixed-sized tables and
buffers in the kernel that I'm hardly ever using on my
workstation but must have allocated at ALL times for the
rare times when I actually need them.  Any non-brain-damaged
OS would have a powerful internal memory manager, but who
ever said that Unix was an OS?

>>1019789

Hi multics fag. All you say about Unix and it's philosophy seems to be a big deal. Should I read more about operating systems to know if you and Unix Haters Handbook are right. Where can I read more about multics? Is Unix all bad? Writing small programs, that cooperate with each other seems to be preety sane...

I don't think software bloat is mostly Unix fault, because Windows has much more bloat, than Unix-like operating systems. I would rather say, that it's caused by corporations, which create bloated software, just because they have a lot of manpower, their employees don't care about software they make and people - often normies, that think more is better.

>>1019789

>Bloatware is caused by UNIX

Listen, I also hate UNIX and hate everyone for making it the most prevalent kind of OS, but don't be a fag. Stop blaming UNIX for everything, it makes you look like an idiot.

>inb4 UNIX created the illegitimate state of Israel.

The nu-programmers are just niggers and instead of making FUCKING ART by making program as small as possible (as they used to) they've decided to program as they like, fuck the fact that something could be done with one line of code, lets make it 10, why the fuck no

>>1019829

>And what if the GC doesn't fit your program's requirements? Can you implement your own GC or fall back to manual memory management?

No, but you are able to create your own areas for objects to be stored in. These areas give you control of the paging behavior and the GC behavior for the objects in this region. For example you can set an area to not be automatically GC'd. You can still garbage collect that area by manually running a GC on it.

>>1019853

I'm seeing the opposite, i.e. people obsessed with appearing smart, trying to play code golf with real software. They'll opt to do things in "novel" ways that require examination or comments to understand rather than in simple, visually understandable ways. They'll pull in new deps to avoid writing some three-liners or to avoid having to do verbose checks here or there.

>this discussion again

So what if code runs faster on older hardware? Modern computing isn't about making things good or great, it's about selling a product, even opensource is guilty of this.

The days where things needed to be minimal is no longer a thing.

The only thing being minimal helps these days is servers, and even then it's a second priority to security.

Additonally, when things are made good it becomes harder to make something that can outsell it.

Modern computers are all about consumers and marketing.

If you can push most of the stress onto the better hardware then not only do the companies and employers benifit for not having to pay you as much, but the hardware companies benifit aswell as they can now sell more of their shit.

There is in fact an incentive to keep bloated projects instead of pushing for clean and well made projects.

Take linux as a good example, sure it works well and is pretty lightweight compared to the compitition but it also a fucking mess filled with the scraps that companies push into it.

And that's how they market linux, it's designed to be sold as a product to the companies that fund it.

So the companies all have their own little ideas as to how things should be, and as such it's a fucking mess.

The idea of minimal died decades ago.

It's not even worth exploring the possibility for most people. Because even if something like firefox is bloated as shit, it still works, and that is all that normalfags care about.

>>1020275

>writing your code to be used by normalfags

That's where everything went wrong

the main problem is the web is bloated so web browsers are going to be bloated. people dont want to sit around for their game assets to decompress to maybe save a GB.

>>1020275

>The days where things needed to be minimal is no longer a thing.

This is not true.

Making good software is not about being minimal for minimalism's sake; it's about the programmer knowing and understanding his / her work and how it benefits a user.

Slapping stuff together without any consideration to consumption of resources makes for a bad experience on the surface, and what's important is that it makes for bad software at the core.

>>1020301

>and what's important is that it makes for bad software at the core.

so?

>>1019456

>Any sane GUI framework does not have cascading style rules as the default

so gtk and qt is out of the question then?

>>1020322

If you want to run shitty software, go ahead.

>>1020324

that's the point.

normalfags are still going to run windows or fagos because "it just werkz"

And weither you like it or not, normalfags are the largest consumers and as such marketing and product people are going to cater to them endlessly.

The people who make software are paid by the people who consume it, so it's always shit because normalfags will never not buy it.

>>1020331

That doesn't make what I said irrelevant or wrong.

My point is still correct as much as yours here is correct.

>>1020331

There should be an initiative for quality software, where only proper software is developed, and you can fund that initiative to support general development of quality software. Once enough autists fund it to secure its existence, normies will no longer have any influence over such software.

I was able to do so much with my first PC..

>P4 3.2ghz

>512mb DDR

>Ati 9800Pro

Now I'm using a coreboot'ed x200

>Core2Duo 2.4Ghz

>8192mb DDR3

>intel gpu

And refreshing the catalog on 8ch takes at least 10sec.

>>1020536

>And refreshing the catalog on 8ch takes at least 10sec.

Right war, wrong battle. The catalog makes 250+ requests. Disable images and try again.