[ home / rules / faq ] [ overboard / sfw / alt ] [ leftypol / siberia / edu / hobby / tech / games / anime / music / draw / AKM ] [ meta / roulette ] [ cytube / wiki / git ] [ GET / ref / marx / booru / zine ]

/tech/ - Technology

"Technology reveals the active relation of man to nature" - Karl Marx
Name
Options
Subject
Comment
Flag
File
Embed
Password (For file deletion.)

Join our Matrix Chat <=> IRC: #leftypol on Rizon


File: 1665038028083.png (83.19 KB, 920x993, ClipboardImage.png)

 No.17153

This thread is for discussing the future of imageboard software, with particular relevance to large imageboards such as leftypol.

Vichan, one of the most popular imageboard softwares and the parent of (the parent of) leftypol's, has recently had its founder Hotwheels say "STOP USING VICHAN YOU FUCKING NERDS", after a major exploit in the embed function dating back to 2011 was found.
( https://nitter.net/fr_brennan/status/1570456345244991496#m )
He insists "Please STRONGLY CONSIDER deprecating vichan on your instances as I have repeatedly begged users to do."

However, despite its struggle to find developers, it has the most developers and most consistsntly active developers out of the imageboard softwares I am aware of (that said, I don't know much about the futaba forks)

The other non-fork competitors seem to be lynxchan (I've seen a board using it call it an abortion) and jschan. Pretty sure both are run/maintained by approximately 1 person at any one time. Furthermore, despite some cool nifty features including websockets instead of polling, they're in many places well behind vichan in terms of expected functionality. I personally believe that a migration within the next few years to one of these would be tolerable but horrible, and that if their approximately 1 maintainer vanishes, the project will stagnate like the others.

So what is the future, on the software side, for leftypol? Is the best path forward to edge back closer and closer to vichan by pushing our differences upstream, and ride it out until it has no more maintainers?

 No.17154

> a major exploit … was found.
Tell me how it can be exploited. The "dangerous" part was the third capture group in that regex. It is not used in the replacement string, only the second one is used. So how the fuck are you going to exploit something that is immediately discarded?

Has it ever occurred to you that maybe there's a small possibility that retards posting on Twitter do not know what the fuck they are talking about?

 No.17155

>>17154
>Has it ever occurred to you that maybe there's a small possibility that [the project maintainers] do not know what the fuck they are talking about?
I assumed they knew better than me and took their word for it without looking. Fair crit.

 No.17156

>>17154
youtube comment calling the kettle black

 No.17157

imageboards are essentially "if it ain't broke, don't fix it"
no point in creating something entirely new IMO
if the vichan central branch fails, the forks will live on on their own, you can already see this happening now
lynxchan and jschan are written in javascript which is an instant deal-breaker, probably even worse than php

 No.17162

>>17157
Dumb question, but if it's not running client-side, why is it even worse? Runtime issues?

 No.17163

I don't understand why people seem to have such a hard time developing new imageboard software. an imageboard is a dead simple program to make. if you don't care about trying to do something innovative and literally just want to do "imageboard but in x language without the cruft of vichan or lynxchan" you can easily do it.

I have to assume that the issue is that the potential developer base is split between /g/-tier pseuds who can't into developing a working web application, or people like me who have no interest in just developing yet another imageboard engine. I've for awhile now been considering making something that would attempt to innovate on some of the traditional aspects of an imageboard, but it would just be for my own amusement experimenting with some software ideas and not something I feel much of a need to get done and release into the world.

>>17154
that "retard posting on twitter" is the developer of vichan you fucking moron

 No.17164

Every imageboard that tries to do something "different" (keyword for appeasing social media users) is dead. The only successful one was 8gag.

 No.17165

>>17163
Making a basic imageboard is super easy. Of course, yes.
But making one with the features or conveniences that typical users and moderation might expect takes more time and effort.

Why make yet another solo project when there are already hundreds? Even big widely-used IBs struggle to find more than one active dev. Like, if you get fun and enjoyment then that's fine, that's one thing, but practically I've seen too much wasted effort on throwaway softwares that don't matter the next year.

>>17164
Masterchan's 8gag's make-ur-board novelty still lives on other sites, like 8chan.moe, 9chan, etc. so I don't know if that's fair to say. But yeah, a novelty is no substitute for a community.

 No.17166

>>17165
I think it depends on what your goals are. personally, I think trying to go for mass adoption is generally a flawed way to go about making software since you'll just end up reproducing the conditions that have already been set for what people want, which will be stacked against you as a developer since people have largely been conditioned to want software design patterns that were designed by large corporations with a huge amount of resources for development and marketing. at best you will end up doing something like what the fediverse (Mastodon, Pleroma, etc.) has done where you make something that just clones fundamentally bad design principles (social media dopamine feedback loop).

in the case of imageboards that's not exactly the case, but imageboards have their own problems in terms of the culture that has built up around them. I think most people on leftypol would agree that trying to attract the average imageboard user (a chinlet incel fascist) is not desirable. so I don't think there's anything wrong with just pursuing some idea knowing that it won't catch on and become the replacement for vichan or lynxchan.

having few people contributing to the software is only a problem if you choose a bad programming language where you can never "finish" a project because you constantly have to maintain the software to keep up with changes in the language and ecosystem. if you use a stable language that empowers you as an individual to get shit done, you're pretty much free to experiment with whatever you want, to hell with being popular.

I agree though that trying to make a "fully featured" imageboard engine could be more of a challenge than an MVP. I still don't think it'd be that terribly hard though.

 No.17168

>>17163
I know who cripplepaste is, that's why I did not give him the benefit of doubt and just went straight to the insults.

 No.17169

>>17168
you know what, fair enough

 No.17171

>>17162
yes
also it's bad to write in

 No.17564

If you want a "base" from which to extend and customize features, Wakaba is the best option out there
Clean, efficient codebase, written in Perl is which is eminently hack-able and has a massive library of modules on CPAN

 No.17602

File: 1668389344252.gif (1.42 MB, 450x258, Coding.gif)

>>17153
This topic has been beaten to death several times on /tech/ alone, let alone other parts of the internet/chaniverse. I'll leave this copypasta here which basically explains the whole thing.

——————–

Vichan: the software that just won’t die

Why vichan hasn’t been replaced and probably never will

Vichan and Vichan derivatives such as lainchan have long been the standard of imageboards in the “alt chan” universe. Briefly, Lynxchan looked poised to take over as the imageboard standard, but it’s reliance on mongodb left it vulnerable to data corruption and unreliability issues.

Any semi-competant coder can create an imageboard with 80%+ feature parity with vichan within a few weeks or even days. So why, after all this time, has no imageboard managed to take over from vichan?

Wordpress is blogging software written in PHP, from the dark ages when PHP was a terribly designed language. Like vichan, wordpress is a structureless, poorly written, outdated, spaghettified oldphp-based piece of shit. Yet, it persists. There are countless plugins, themes, and tutorials created for it, which enable countless nontechnical small business owners to create good looking websites with prebuilt themes. The very ease of use corresponds with it being hell to maintain and write plugins for. Ironically this self-induced difficulty has created a whole ecosystem of plugin and theme developers, selling overbloated plugins that perform functionality that could really be done in a few lines of code, mindlessly installed by nontechnical bloggers. Even conservatively, around 30%+ of all websites use wordpress, including major corporations like airlines, banks, and media conglomerates. The fact is, there is a self fufilling prophecy, where a software’s popularity can meme itself into existence. Wordpress is the Kim Kardashian of cms software: popular for being popular, despite an utter lack of any other redeeming qualities. And vichan is essentially the wordpress of imageboard software.

Like wordpress, there is a whole ecosystem of people who have become familiar with vichan+derivatives, their configuration options, their ticks, their idiosyncrasies, and have created an ecosystem of themes, scripts, and so forth. Not to mention a userbase which is used to the vichan style UI/UX experience and revolts at the slightest change. After all, that’s really what usability is: familiarity.

It’s entirely possible to create a piece of imageboard software that has feature parity with vichan, but it is much more difficult to create a piece of imageboard software which slavishly recreates all of vichan’s idiosyncrasies, ticks, warts, and all, which is really what people want: a drop in replacement. And since the whole point of creating new software would be to get away from some of the warts, the potential writer realizes that there is really no point in trying to begin with, other than maybe as some masochistic coding exercise. This is why lynxchan was really doomed from the beginning, whatever it’s technical merits or drawbacks.

The chans are dying. Really, they have been dying a slow death for quite some time, given a bit of a boost from the anti-sjw/Trump mid 2010s era, but quickly returning to a dying status, accelerated by the death of 8chan. Now, in the minds of normies, chans represent at worst, a hellhole filled with incels, white supremacists, pedophiles, and god knows what else. At best, they represent a “boomer” style of communication that’s deeply uncool, as opposed to the dopamine inducing, shiny-colored-box experience of tiktok and fbi.govs.

The conservatism and resistance to change of the current chan userbase is inherently connected to it’s age. While there are of course some young people, the preponderance of current alt-chan users are 25-40 year old millennials who came of age during the 2000s “golden era” of 4chan. This can be seen most clearly on lainchan, which is rife with arcane references to gibsonian cyberpunk novels, 1990s anime, ‘hackers’, and disgraced free software guru/activist Richard Stallman.

Like an aging video game with a slowly eroding multiplayer playerbase, the chan era is slowly passing from history. Many of the young, edgy people who would have been attracted to chans in previous eras are now on platforms like disc.ord. The whole appeal of php-based chans is that cash strapped youth could buy cheap shared hosting and slap a chan together with minimal technical knowledge and little money. Now, fbi.gov allows them to do the same thing with absolutely no technical knowledge, no money, and no effort. The barrier to entry is far lower, not to mention that a fbi.gov “server” does not expose the owner to the same level of legal risk as ownership of a standalone website. The private nature of fbi.gov servers also keeps it away from the prying eyes of journalists and others who would likely freak out if they saw the same or similar content on a website.

It’s only an influx of new people/users that would lead to the space for a radical reimagining of anonymous internet communication to succeed. As long as the chans remain a place for aging neckraging neckbeards, driven nostalgic for a bygone era of the internet, rather than a vision of the future, the chaniverse will continue to bleed users to other platforms.

To sum up, what this means is that there is really no chance for vichan to be displaced as the software of choice for chans since chans are dying. If chans were rapidly increasing their overall userbase you would see a proliferation of new and sometimes experimental software and formats. But you see the opposite.

 No.17603

>>17602
just to follow up, I've been straight up TOLD by the people who run leftypol that even if I wrote a replacement chan today, they wouldn't switch to it.

Unless you have some alternative userbase which is willing to adopt brand new software, there's no point.

 No.17604

>>17602
The technical aspect and the sociological "userbase" aspect are completely independent of each other. Most "normal" people don't give a shit about features, all they know is that you can post and that's it.

 No.17653

>>17603
Did they say why they wouldn't change?

 No.17659

>>17653
basically they're focused on improving leftypol's VIchan variant and its functionality and rewriting it would be too much work, plus they wanted to do it in scala if it were done at all.


Unique IPs: 8

[Return][Go to top] [Catalog] | [Home][Post a Reply]
Delete Post [ ]
[ home / rules / faq ] [ overboard / sfw / alt ] [ leftypol / siberia / edu / hobby / tech / games / anime / music / draw / AKM ] [ meta / roulette ] [ cytube / wiki / git ] [ GET / ref / marx / booru / zine ]