Don't contribute anything relevant in web forums
karl-voit.at
karl-voit.at
None of this is true. All these services have pretty fully-featured APIs and several independent clients.
Even otherwise, the fact that some of them may not exist at some point in the future is a terrible reason to not use them at all. I still find tons of relevant content in hosted forums from the 90s and 00s, and more recent ones like Stack Overflow and Github.
I'm pretty confident that most of the services he is complaining about will last longer than his own blog post.
Also, people may not realize this is a problem until it actually becomes a problem. In the mean time they will be happily rendering stuff on every page load, after fetching (static!) data from mysql on every single request.
Ironically, I can’t actually read this blog post because it has become unavailable. I suspect that undermines his entire point.
I know people are really enjoying dunking on this guy for his web service outage, but I really don't think this is the right takeaway.
On this topic specifically he notes:
> All major commercial services such as reddit, Facebook and so forth keep everything a secret that is not ultimately necessary to use their services. Their software is a secret, they don't offer open APIs or only very crippled ones, you don't have the possibility to get to the raw data. So no luck there. You do have a lock-in situation.
This is an issue, not of availability, but of control.
Yeah, okay, his blog might be experiencing an outage right now. But he can move it. He can shift the content to another medium. He can back it up and retain ownership over it. As he very rightly points out:
> without very good support for data export, service duplication, open standards, any content you provide in closed web-based services will be lost just as MySpace already lost twelve years of content just so, just to mention one big example.
But, hey, let's ignore that and joke about his blog being down. After all, who doesn't enjoy a little schadenfreude from time to time, right?
I suspect that undermines your entire point.
I can't say I share your confidence.
Take Google Groups as an example.
Were it not for archival work being done by various volunteers, the content on Usenet was at real threat of vanishing from the internet.
This is hardly unique. We've seen the same thing happen with Geocities and Myspace, for example. Another poster also mentioned G+, which is an excellent recent case.
Is this a reason to not use these platforms at all? I don't think so.
But it is a reason to retain your own copies of the things you care about, hence why, for my own blog, I follow the POSSE (Publish Own Site, Syndicate Elsewhere) pattern:
And speaking of that, because of purported spam problems a number of Usenet groups (and especially their archives) currently aren't available at all via Google Groups.
Maybe it's impossible, but hopefully some day we can get to a point where social media silos are not necessary, with easy to set up and use personal sites/apps that use Indieweb principles. The Indieweb movement is excellent for this but needs a lot more developer contributions to make this dent a reality.
And yet years of fantastic RPG community contributions on g+ are absolutely fucking gone. At their best they got taken out into barely-accessible formats, so they technically "exist", even if you can't find any of their content anymore.
As platforms mature, their APIs become more and more restrictive. E.g. Twitter's current API allows only a fraction of what it used to allow in its early days.
For me the point is that if you use an external platform then you no longer have control over your content; I think it still stands.
Stack Overflow is a nice counter example. Then again, all works fine as long as the goals of content authors and the platform are aligned.
the reason is that the longer I use them the more I pump data into it and the more I am getting suckered for when they shut them down. this is exactly the justification for why standards should be open.
Same goes for any product from Google, since they are notorious for service termination. In doing so, they abandoned their core promise of indexing the world's information. G+ is gone, and the Groups archive is ever-thinning: Usenet posts I made a couple of decades back were once accessible thus, since Google bought the Dejanews archives, but are no longer discoverable.
Fortunately all my most interesting snippets and contributions these days start life in my own private repositories, so I have my own record. Everyone should do the same.
The point is to communicate with other people here and now, to help each other on short notice, to spread knowledge which the readers can apply immediately.
It's like mistaking talking for books. Talking is fleeting, but also indispensable for doing anything of consequence (including writing a book).
I think the curation piece that separates chatter now vs. published organized information is where forums lie in the middle of
I was thinking about this yesterday actually. I’ve been having a lot of questions about this new linux distro I’ve been using, and piling them on in the IRC channel, but if no one answers it right away it gets buried by the next message after me. So I kept re-asking it for multiple days before realizing, you know what if this is something that only even a few people might know how to help me with, I should probably leave it to “steep” on a proper Discourse topic, or reddit thread, or stackoverflow question or whatever. I hadn’t done that because I was just super impatient and knew it might be a good half a day or 2 or 3 days before I get a reply, but it’s much more likely to get one at all than in the IRC, where someone butting in with a meme or something right after me means my question might as well be nonexistent in the conversation stream.
It's just that the internet allows them to be async conversations. A forum is fairly equivalent to email, just usually more public than email and the underlying infrastructure isn't federated.
Email is a persistent conversation which gets archived indefinitely, not an ephemeral one like talking with our mouths.
Corporate email (and IM, etc) gets purged pretty quickly, within 2-3 years, unless you save a particular message / thread explicitly.
Forums are -- or used to be, anyway -- a hybrid form of communication: it's both immediate but also useful for achival purposes. Many forums will tell you to read a specific thread when you want info about something in particular. In some forums, users use specific threads for archival/progress of a project.
Chat is ephemeral like you describe, but forums aren't necessarily.
>If you're, for example, contributing to a reddit thread about something which is irrelevant or anything with only a short-term relevance, this article does not apply to you right now.
Forums often end up with some posts that have a more long term benefit, and those should be hosted somewhere else.
I have parsed many old forums after a search engine sent me there - this has been my primary education tool for different things. Niche issues encountered by several strangers across the world can be resolved by someone who made a post one time so it can occupy some hardware storage, perhaps many times archived in several scattered drives depending on its contribution value, from where it can be served to anyone else who needs it in the future.
Web forum archive shuttering described by the author would limit information available for the search engine to parse. But the author's solution is to reduce contribution effort because forums may disappear one day? [1]
I agree with you - within a week of something posted is when the "work" gets done (and idea transmitted out of internet into physical group of people who can relay the memory). Seeing a snapshot of a helpful interaction between people skilled in different domains sharing observations/stories/explanations is often inspiring to me and so I would want the creators of this content to put their best foot forward instead of treating it as disposable communication space.
Spot on. We should be encouraging users to redecentralize and use distributed open systems. Not walled gardens.
Users are not enough, we need concerted engineering efforts.
it takes engineering effort, time, money, and reason to keep any system running for a prolonged period of time, centralized or other.
My posts on every platform over my whole lifetime can't be more than what can fit on a flashdrive. It shouldn't be an insurmountable goal to store them, it's irritating that it's so difficult to get data out of all these fragmented systems.
I have seen so many large and important web sites from early 2000s to have been gone by now, to say nothing of personal blogs.
https://en.wikipedia.org/wiki/Coral_Content_Distribution_Net...
Facebook, Reddit, Imgur, Youtube, HN, SO, and others are where all the people go because it's easy and, well, all the people are there.
Those are your potential readers and customers. So you go there, as well.
I remember having to visit multiple forums and buying stuff directly on small websites. I needed to have credentials for all of them, but that was the hardest part, really. I liked the experience more.
Reddit et all are the same, they're technically a collection of smaller forums, with the important difference being that they're controlled by a single, big company.
I think it's a dangerous thing for the future of the Internet and information in general. As Google and Facebook demonstrate, one ban and you lose potentially everything digital that you "own" (don't get me started on that).
Anyone would be insane to take that risk, but most people don't even realize it exists.
The "contribution" is retrospective, it is only a thing that exist when you look back.
The centralisation works because brings people together. Even with torrents and bitcoin, the two decentralised networks that actually work, are valuable only when there's The Pirate Bay and Binance. Without the centralisation tools, these are simply cool technologies in the lab.
So like, maybe you sound like the type trying to build a brand or something, and would benefit from more eyes on your content in a given time? Maybe some people are apocalypse preppers and like the idea of if their posts had a greater chance of surviving among the limited number of digital archives left standing in the ruins, lol
edit: fwiw the author has web archive links at the bottom of all their posts. Since the page isn't loading, I couldn't see it.
A post complaining about how you shouldn't rely on major sites because their content doesn't get backed up... is failing to load... and being saved by the same centralized, private third-party backup service that also backs up major sites.
You can't make this stuff up. ;)
I post on Reddit and HN rather than a private blog because that is where my thoughts will be seen and responded to.
My Hacker News comments will probably live for at least a decade or two. And I had no issues archiving this thread right now. https://archive.is/JrJEe
All the big platforms suffer from not giving you much choice to export your own writing once you've sent it. I think it is as much for reasons of attempted lock-in as it is due to incompetence and scaling pains.
Facebook, Twitter, Reddit, and other platforms will all present a challenge when you try to "find that thing I wrote two weeks ago".
The only solution I can see is to build services which allow content export in usable format and to continuously audit these backups yourself.
I think it is a solution which only seems difficult at first, but is possible already with tooling available today, and can be made accessible to a non-programmer with some effort.
For my own project, I consciously chose .txt as the base format for both text and metadata, and have tried to make even the metadata as readable as possible.
I regularly do a clone test of my blog and message board, meaning I download the data export, unpack it on another (disconnected) instance, and ensure all content has made it across. (Thanks to using PKI for accounts, full continuity and sync can be achieved across multiple instances.)
I think the key to maintaining a trustworthy community is keeping the size small, to where all the activity can be monitored and reviewed by a human. Once you're facing a firehose, you're at the mercy of all the forces out there.
I think y'all are missing part of the point. If I host my own content, I get to decide on its availability. If there's a problem or an outage, I can fix it. If I don't want it up anymore, I can take it down. If I want to keep it up forever, I can do that.
If I post to someone else's web forum, I have little control over what happens to that content. I'm at the whim of the site admins as to its availability.
Someone else also mentioned private blogs that are exposed both as NNTP and as HTTP(S)/HTML, and this would also be good; if you are doing a blog, you could reject articles without a References header (which must reference a blog post) unless you wrote them by yourself (the articles without a References header are the blog posts, while those with a References header are the comments).
A subset of Markdown (without HTML, and maybe a few other things also removed) could be used if wanted, as long as it has a MIME type defined (so that you can identify NNTP articles which use it vs those which don't), and is perfectly readable without a Markdown parser.
I thought of making up a URI scheme for referencing Usenet/Unusenet articles. They mandatorily include the message ID, and optionally include server name, timestamp, newsgroup name(s), and/or hash. (This is better than using Google URLs.)
I would also suggest perhaps even with the Hacker News, to add a NNTP interface (or an extension to a NNTP client; at least for my "bystand" NNTP client software, protocol extensions are implemented as programs that communicate NNTP on stdin/stdout, and possibly those writing other NNTP clients might want to implement it too).
I spent 16 years maintaining those things in the decentralized style, and I think it's fair to say that effort was of literally no benefit to anyone, other than the hosting company I paid hundreds of pounds to for the ultimately doomed VDS. It's a relief to have finally handed them over to people whose job is to maintain them.
Edit: Also when my blog had a comments section I got like 10-20 legit responses to posts and hundreds of spam messages that were a huge pain to manage.
Reddit, Twitter, and other fast moving websites are best for ephemeral discussions anyway. In most cases, that's a feature rather than a bug. Writing up a thoughtful blog post that's meant to be a lasting reference for many years takes a lot of effort, planning, preparation, editing, proofreading, and other work. The upper limit for this type of material is relatively low due to all of the effort.
On web forums Reddit and Twitter, people don't feel obligated to write perfect posts, so the conversation flows more freely.
There's a place for both long-lived and ephemeral content on the internet. I don't think it's productive to give blanket advice for people to avoid certain forums or deliberately withhold information to make a point.
edit Upon reading the content, yes it does. Please ignore this thread.
It's unquestionably useful today, it's somewhat distributed by mirrors and the database export, it has an API, but there's no question that if SO folded tomorrow that a lot of utility and a lot of effort would be lost. However, I am confident that if SO did not exist and those contributors did not become addicted to the dopamine boost from "+10 Karma!" notifications that they would not have instead created permanent, Google-able, individual blogs with an equivalent amount of utility as what SO provides.
I can have a super interesting, deep philosophical conversation with a friend, or a 100 friends - and once everyone leaves the room, its like the conversation never took place.
What's the point then? Well I benefited... Maybe I got some karma points so I feel better too.
I agree with with concept of "keeping every bit of data forever" and having it be freely accessible to others. But I disagree that some forum's data retention policy should dictate when and where I share information.
As I rule I no longer discuss cultural , political, or socio issues on the internet. In my social circles , with some friends, I might talk about this stuff.
On the internet it's very easy to forget the human and start acting like a jerk.
I've also accepted I don't need everyone to agree with me. For example I don't use social media or online dating. These things weren't good for me mentally. I don't need to write 80 Reddit posts about how right I am.
That said, I have found Reddit to be much friendlier than Stack overflow for tech questions. On SO you get banned if you write too many subjective questions.
'What IDE should I use ' gets flagged and 2 or 3 other questions like that and your banned.
As long as you aren't straight up belligerent , you won't get banned on Reddit for that.
It's a careful balance. Once Corona ends I very much plan on deleting Reddit. Right now it's an interesting distraction , but like other social media it can't replace real human interaction
Imagine all the information that's been lost that was posted to private forums and they were killed off, databases crashed, etc.
At least with USENET your stuff was all over the world. And archived by Deja, eventually.
USETNET, which Karl's post mentions, was a fantastic place to go, because of the "pull" principle: you go there when you want to engage, not as distraction when you would like to focus on something else (email, Twitter). (We also had local groups in our university labs for social and technical interactions. This avoids the spam issues.)
Then there is the social element. On Slashdot, HN or Reddit, I don't recognize names (PG and RMS aside), but in newsgroups like comp.ai.nat-lang, comp.std.c or comp.compilers I felt I knew a small but committed crowd of regulars (and their views) after few months.
There's more to it, but I stopped caring and went to reddit and made some contributions there purely out of spite.
If you want to own your own data, post on your own website first then syndicate anywhere else.
There's no way to find people to trust. There are other uses of Reddit, like content discovery, and completely destroying your life with obsession.
Same with Usenet -- far more useful than Reddit, Google Groups or places like HN, but fell victim to the pressures of spam, moderation and lack of standards evolution. I probably have a deep backup with News posts from the 90s but it is essentially lost to the world. It was never intended as a historical artefact, so in 100 years it will seem like a dark age, no hardcopy records and a tiny digital archive. In contrast, FB will still have everything but you won't be able to access it.
And just like it was all gone. There isn't a metric to quantify the novelty of a service, but I genuinely felt like we burnt one the last resources on product/engineering/building/hacking when Google decided to pull the plug on that.
This article highlights that fact too.
Perhaps all passing conversation is "Irrelevant" in the author's perspective? I don't know, but honestly... if this were deleted after 3 days, I wouldn't care.
I will suffer through the "interface dictatorship" to reap the benefits of a huge audience and active moderation, both of which ensure that questions get answered quickly and correctly.
There is one overlooked aspect of public web forums like Reddit. Although it is temporary, you can get the results you want now by getting the word out to as many people as possible.
BTW: Tapatalk combines their own hosted forums and those they don’t host but implement APIs for, so it both centralized and decentralized.
With that level of expectation, I'd expect you to encourage people to donate to the Wayback Machine / Archive
In a google search I often append "reddit" to the end of my search string and often do a custom recent time-range for the result, instead of the most SEO optimized answer which is usually outdated
I think I've contributed one ‘significant’ fact on Reddit and one on HN.
When I did it though, I sent an email to their support team asking for a copy of all my data and then for them to delete it all, which they did (AFAIA).
TLDR Open abstractions on top of open data trivially and rapidly distributed.
[1] https://hacks.mozilla.org/2018/07/introducing-the-d-web/
[2] https://blog.archive.org/2018/07/21/decentralized-web-faq/
Maybe a separate project of the Wikimedia foundation, sure.
Maybe not. You're definitely not the only one who wants something like that.
Years ago some folks broke off from Slashdot and started posting to comp.misc. They're still going today
I had a quick look around, and the tools aren't trivial to set up. If someone who has experience with NNTP servers could bundle things as nice Debian packages with easy setup then that may help spread it again.
Spam can be reduced by using moderation, I wonder if there are new ideas around for countering spam in non-moderated groups?