honestly we should have collectively realized way earlier that putting all the useful, readable, un-touched-by-SEO help content for basically every niche hobby fandom and ideology in the hands of one for-profit entity was not very wisdom-pilled of us
we can still easily fall into this trap if there isn’t a good way to migrate communities between instances. And even if we could just take /c/technology@beehaw.org and move the whole thing to /c/technology@feddit.de or something, that would still break all the indexers’ links
What we really need is some sort of torrent-like system for this content with something equivalent to magnet links.
Sounds like you’re describing ipfs :D
I love the idea of IPFS, but every time I’ve tried to use it, it has always been very slow.
amusingly another chicken egg problem. More chickens, faster the eggs. Wait that metaphor works!
Yes. When everyone enters info on corporate sites, sooner or later they’ll decide to monetize it.
Reddit going evil on charges and showing their colours in the AMA has been a wake up.
That is the main reason why I’ve been blogging on my own website since 2004 https://paradies.jeena.net/weblog/2004/apr/ersteintrag (and switched to English in 2010 https://jeena.net/posts )
Yep. I blog infrequently but I’ve said a few times in my posts, I am writing this article because I need to remember the steps to do this weird niche thing in case something breaks in the future. If it happens to help someone else out, great.
I agree, but I also have serious concerns about this being the replacement strategy. It could be because of my ignorance of how this all works though. Like many of you, I am new and here because of the reddexodus.
These servers are going to cost money, and for many of them the money will run out. Is there a function to preserve the collective content of an entire server once it goes dark? I know that you can migrate your own account to another server, but what happens to everything Google has indexed at Lemmy.world if the worst happens? Is it all just dead links? What if many of the users do not migrate? Is it just gone?
I am concerned that in the current state we are setting up to burn everything that loses a couple admins or becomes too old to economically host.
In practice the content is distributed to all the other servers, so people who have been reading it before will still be able to on their own instance, but you’re right the indexed domain is gone and so are the results in Google.
But there is one difference, one instance of lemmy only stores a very small fraction of the content. And it’s much easier to fuck up one reddit compared to fuck up thousands of lemmy instances simultaneously. So if one instance goes down, the rest of the fediverse is still up and running.
I was on a mastodon server and the owner decided it was not worth his money to keep running. He did not inform anyone on the server or allow any account backups and all was lost.
With federated services, I feel like it’s somewhat important to get to know the admins of the server you use. You don’t have to be best friends, but at least know their name, motivation for running the server, and how it’s funded.
There should be a service or extension or something like that that performs regular backups for you. Of course, lemmy needs to implement a backup/transfer feature first.
Before reddit removed them most of this compiled knowledge was in the subreddit wikis. I honestly believe a return to communities with wikis is the long term replacement.
I think it’s a fair concern. We’ve seen other parts of the fediverse successfully implement crowd sourced funding via patron and similar to keep mastodon servers running and I suspect if Lemmy remains “the place to be” admins will have reasonable success with a similar model. Lemmy is super efficient and can support 100s of users on a single box so I think if 1% of users paid like $5 a month you could probably still support 99% of users “for free”.
I’m sorry, but clearly you have not looked for niche information on Google for a while now. Lots of links end in dead ones, particularly when I am looking for vehicle information on older models.
I’m not sure what you are trying to say, we shouldn’t be concerned because this problem already happened?
A lot niche older vehicle information, if it wasn’t hosted on Reddit, was often on forums funded by enthusiasts, which eventually ran out of money and no longer exist. This is exactly the problem that I’m concerned about. Particularly so if a certain community balloons in popularity and an admin nukes it to keep the server costs under control for the other members.
Completely what I’m saying, but to add on it is not just forums. With the new web, I’ve hit a deadend on many OEM websites as well, and part websites, and others. I’m sure cell phone and computer information is similar, in fact after trying to research a power supply for my old prebuilt I know it’s a fight.
These are certainly possibilities! It’s happened elsewhere in the Fediverse… but already we can export most of our data and migrate to a different instance. Getting these base features right is important before enhancing their functionality. Planning for the future is important too. So far I’ve been impressed by Lemmy, though it’s not nearly as portable as Mastodon or Calckey or Pleroma etc. Part of that is that in Lemmy/kbin we don’t follow other users… we subscribe to groups (subs/communities/magazines).
Still, with the nature of ActivityPub, it’s inevitable that migration tools for Reddit-like federated apps will get built quick-like
It would be nice to have some sort of IPFS + Lemmy (or other federated network) witchcraft going.
Makes a lot of sense, especially due to the drama earlier on with Imgur and its image policy
we should have collectively realized way earlier
some people have, but whenever you’d mention it, you’d be met with “lol take the tinfoil hat off”, “but we’re already using [for-profit platform] why would we move when everyone’s here” and “but it’s haaaaaaard”.
Source: https://xkcd.com/743/
The fact that the alt-text directly mentions Diaspora is more than amusing in this context
Had to zoom in to find out why it is suddenly year 200. There is a tiny 1 in there.
Hey! I’m not probably autistic! I’m definitely autistic, there’s a difference!
I’ve said it numerous times over the years, the Internet has been centralizing rapidly and it benefits none of us.
In 2005 you’d wander around, going from peoples’ personal pages to forums to whatever else people linked. In 2015 half of those websites were dead because everyone got their content on reddit anyway.
I just can’t agree more with you. Like wow this reddit blackout has truthfully opened my eyes to the massive, giant and incredibly amount of useful information that is currently resting on reddit servers.
Need some bots to start porting all those posts over to Lemmy lol.
One thing the FOSS world really needs to get on right now is some form of search engine accessible distributed content archival. We need a way to store useful content from the past in a way that no one individual or group of individuals is capable of deleting it.
I believe archive.org fits the definition
is there a tool that makes searching archive.org reddit (etc) posts easier?
GOOD!
I’m considering switching to Kagi because of this. Its results are impressive.
I’ve been tinkering with it a bit. It’s okay so far. For work stuff it’s been somewhat helpful (though the problems I’m solving appear to have nothing to do with the code I was debugging). Considering getting a personal subscription to kick the tires for a month or so.
I was amped for Kagi when I first heard about it. But they bumped the price up after the LMM boom. Still might have to bite the bullet as part of desire to use paid ad-free services.
Depending on how much you use it, it might not be that much worse though… The old price was 10$/mo for unlimited searches. Now they offer different tiers starting at 5$/mo for 300 searches.
Personally, I use about 300-500 searches per month, so my monthly bill is actually less than it used to be (5-7$).
The cost is why I’m probably not going to plug it into my Searx install.
This is the first I’ve heard of Kagi, how does it compare to duckduckgo?
Seems nice based on my trial but they are really pushing the envelope on my price tolerance.
5/mo is too much I think.
I’m not sure what to think about the price. I can’t really imagine life without a search engine, even though I was alive for a couple of decades before search engines existed. I pay $400/month for my car, but my search engine arguably gives me more value (I am lucky not to need to drive a lot). I wouldn’t pay $400/month for a search engine. But $5-10 to have a degree of freedom from the tracking and results that aren’t just trying to get my money? I am intrigued.
Idk, maybe it should be usage based. I feel like 60/yr is too much. I’d be fine with 19.99/yr but idk what they’re costs are. Otherwise, I do like the idea. Confession, I haven’t used it yet but I plan to signup and try the first 100 searches free.
I switched and I’m happy. I rarely use the
!gshebang to see if google has anything more useful and it rarely has.
It is - but you can still access via archive.org and similar resources.
Doesn’t help for searches though
You can copy the address of the search result into the way back machine or Google cache
You’re absolutely right, true, but that will work for you and me, but not for your typical user, even the more advanced ones will be stumped at that point
Google should just redirect to the archived page if the link to Reddit is dead.
It’s also a super clunky way to search. If I’m skimming posts for technical issues that I need a quick turnaround for, I’m probably not going through that hassle unless I’m desperate.
Tacking “Reddit” onto search queries almost became a prerequisite. Never imagined I’d have to replace that with “-Reddit”.
It’s made researching a media centre setup very difficult this week…
Give it some time, people will get comfortable here, the revolution dust will settle an we will be adding ‘-Reddit “Lemmy”’ to search queries (fingers crossed!)
But how would this work with broader federation? Searching other instances like beehaw or kbin? We’ll needan new search optimization to search the fediverse more efficiently.
I guess google will just have to suck less if they want us to keep using it.
Had this happen today. Was searching for some programming related stuff and top pages are all inaccessible Reddit posts.
Same, but it’s just growing pains.
We should start rewriting posts in lemmy with the correct information.
Sad thing is most search engines suck/haven’t really indexed mostly anything in the fediverse. Wonder why
About 4 people at work Monday discovered the blackouts and learned the reason from following Google results. I’d say that shows the effectiveness of the protest. That’s 4 individuals that I work with personally who wouldn’t have known otherwise about the api problem that now do. I can only imagine how many people are in that same boat.
Same. I found it funny though. Showed that if we tried we can cause some chaos
Same. Had some things I needed to look up for my 3D printer and much of the results were inaccessible.
Was a pain.
Ditto, actually. The 3D printing communities I’ve seen here are just so much smaller.
Same with Pathfinder 2 questions.
Hopefully it will help people realise that a profit motive being attached to everything is actually counterproductive societally.
People rely far too heavily on reddit for public resources. Here’s hoping that changes now.
This also highlights the problem with a lot of communities moving to Discord, which inevitably ends up as repositories for critical information, but can’t be indexed by Google. Reddit is still valuable as a problem solving resource, and I hope they fix this API fiasco.
I’m willing to bet the lack of api access going forward will make all reddit posts disappear from crawler results anyways. I’m no expert, but I imagine the crawler is picking up on all of the interconnected references to reddit that are all due to free api access. As soon as those connections disappear, so dies the value to the entire community. It will be just like the garbage results we get from every single source now. This is the path of neo digital feudalism.
API calls are almost always private between the caller and the endpoint (think telegram bots or mobile apps). There isn’t really a technically feasible way for a crawler to somehow “infer” any kind of knowledge of how api calls are being used unless the result has some kind of publically visible side effect (E. G. The program using the api is generating a web page and uploading it somewhere crawlable). Google et Al go by how many links from other pages to the page of interest exist (inbound links) and multiply by a smattering of other things like quality of keywords, length of content etc.
That said, if you’re implying that the api changes mean that:
- people are less likely to use reddit because they can’t access it via RIF/Apollo
- less useful content is added to the site to be indexed,
- fewer inbound links will be generated that point to existing posts
- pages stagnate and drop in ranking
That is a plausible concern.
fewer inbound links will be generated that point to existing posts
pages stagnate and drop in ranking
This is what I mean, the external references people had in the periphery will dry up. Like if I’m not using Infinity to generate better refined search results, now I don’t post the link to Stack Exchange, and this reference fails to cascade across various copy paste blog resources. Now the original reddit post is a dead end source with no external weighted reference value. It’s all of these advanced features implemented in the periphery using the free API that create the usefulness in the first place.
Searching reddit will be just like YouTube searches now. No matter what technical wording you use, you’ll never find technical references again. I can type the title of a video on YT verbatim and still won’t get the correct results, but I can log into an old account and find the content in my hundreds of playlists I kept as references. It is still there, it is still public.
Yeah that makes sense! I totally agree! Search is becoming pretty difficult these days!
The other thing is that Discord search is god awful. There’s absolutely no way to modify your search for better results, whether that’s to require something to appear exactly as typed, or to exclude certain results, it’s just you put in the words and hope you get the right thing. Sometimes that works out, but sometimes it will make the dumbest connections and render your search useless unless you want to trawl through pages of crap you don’t want. Like I’ve found out that Discord considers the words universal, universe, and university to be the same…
Definitely saw this coming… can’t imagine what will happen if Stack Overflow pulls something similar. All WebDev/DevOps work will halt overnight.
I’ve been trying to put my issues/solutions in a personal blog or wiki, but there’s so much old info out there in sites like Reddit/SO/medium/etc, it’d be a huge loss when it goes away.
Maybe it really is time to get open sourced AI and bots to archive useful information so they don’t get monopolized.
deleted by creator
One thing that is designed to be future-proof is MarkDown, I’ve been taking profesional and personal notes and exporting important information from web pages to markdown and hosting it on my own PC for a while.
MarkDown it’s text based so you can have a huge amount of data with just a tiny bit of space. And it is easily translated/rendered as HTML. Apps like Logseq, Obsidian or Markor are good starts for managing huge vaults of information.
I’m thinking that whe should create a MarkDown community here.
edit: There’s one already! https://lemmy.ml/c/pkb
deleted by creator
Theres plenty of software designed to store personal wikis and info in markdown, of you are interested checkout Obsidian, Logseq and Jopling (in that order for me)
We’re going to have to actually read official documentation instead of relying on some greybeard’s wisdom on SO 🥲
Well, at least stack overflow database dump is available.
At least with SO, they have historically put up dumps of all user data on archive.org (that stopped recently but it’s allegedly coming back). If something were to happen, at least the information would still be decently accessible, just not indexed as well.
I mean if you do hit this, like I have. You can just use google’s webcached view. or sometimes the internet archive.
I found this covers most of my needs: https://cachedview.com/
I just add “forum” to the back of my search
If you’ve ever owned an older car, you know that this is the absolute best approach.
Good luck getting exceptionally niche advice for things like that on reddit. Forums get so much more specific, you get an entire forum dedicated to one car model that was only built for 5 years and a bunch of people there know literally everything about it, like the fact that you’re better off getting an aftermarket PCV valve because those are built a bit better and don’t fail early, or the fact that the shifter cable has the tendency to get water in it so you better be careful shifting out of park on a really cold morning, you might just snap the cable if it’s old.
Yeah the problem there is VerticalScope has a hidden hand in all those pies and draconian monetization policies.
Oh…why didn’t I think this…
100% has this happen today. Wanted and answer, the only answer was on Reddit, and the Google link was busted.
So… How is Lemmy set for SEO?
SEO as a concept needs to die. dont get me wrong i want Lemmy to show up in google results but doing that by spamming keywords and unrelated “related” posits is not the move
Then what is the correct move? How do I locate content related to keywords?
wish i knew, maybe something with AI?
the reality is that current-day SEO isn’t even Optimization, it’s Lying. try googling a recipe or an alternative to a popular software and see how the first four pages are all ad-ridden, useless spam-bot articles designed to retain users rather than give them the information they are looking for
exactly my point–neither of these (excellent) sites will show up in your standard google search because they have the integrity not to abuse SEO
It’s not, lol
This has been deeply frustrating, but since that’s the whole point, I support this collective inconvenience.
All in all it’s also a testament of how bad internet is now. All the information is concentrated in few sites that, if gone, gets lost.
Also, I find that basically every search result that isn’t reddit is sponsored content.
Search something real specific like “Best aftermarket injector coils for a 2009 Toyota Corolla” and you’re going to get 100% advertisements and listicles for search results, likely written by somebody who doesn’t know shit about cars.
Append “reddit” to that search, and you’ll be led to a post from a car mechanic giving their opinion on the matter. And, well, I do trust a random stranger on the internet more than I do an advertisement.
You can prepend a link with “cache:” to view Google’s cached version of the site. This works automatically with the url bar in at least Firefox and Chrome (likely other browsers as well). If your browser doesn’t support that you can enter it in the google search bar and the result will be the cached version of the site (if available)
