As an example, orphan articles (ones not linked from any other pages) [1] are categorized and tagged on Wikipedia. There are many editors that spend their time trying to find places to link to orphaned articles, which improves discoverability of that information.
The best programming language-specific Wiki I've found is cppreference.com. They have a very nice stylistic approach to dealing with the different C++ standards.
Aside from Mediawiki.org, Ward Cunningham, inventor of the wiki, has written extensively about wiki patterns. He’s also created a reference implementation for a federated wiki that attempts to solve the fork & revision control use cases centralized wikis struggle with. See https://github.com/fedwiki
My initial guess is ads, but is there more to it?
I agree with you in general though, the GitHub wikis really don't feel the same.
So if you are setting up your own wiki then you need to spend a lot of effort to create and maintain the ToC/landing page and build those interconnections up, and just hope Google indexes everything so you can search etc (... and if you have an internal wiki at work then you are basically screwed because Google won't index that). Otherwise you write your articles and they "disappear" and are not readily found.
At least that was my experience of working with media wiki and dokuwiki - the sites always felt "empty" even if you had teams of people contributing articles, because it was hard to see what was there in a meaningful way. And so they always felt moribund and abandoned even if they actually weren't.
Patiently waiting for someone to create an overhauled UX. Potentially one that does some sort of clever LLM-powered hierarchy so people can browse and peruse page hierarchies sliced-and-diced by some sort of topic/area-of-interest. So say I enter "engines" or something as my area of I treat, the LLM auto-categorises the pages into a sensible hierarchy that I can view, then I change my term to e.g. "steam engines" and it recalculates the hierarchy and shows something sensible for that etc.
I think that would be http://fed.wiki/view/welcome-visitors (the author of wikiwikiweb's new thing)
That conflicts with things like known governments and corporations: authoritarian, if not also bureaucratic. The attempts to control wikis make people much less likely to contribute.
If you don't know what Fell Running is, you probably would have little use for it and it wouldn't ever be on your radar -- as would be most things outside the pale of pop culture or your own version of the Roman Empire.
Any tips on docs that integrate with support tickets and maybe some AI would be great.
Both systems are able to link to each other. I worked for a company where Jira was configured to take a support ticket, and then link it to a knowledge base article in Confluence based on namespace and topic I think.
My understanding is that Wikipedia has been in decline (in terms of user activity) since 2007 (much as Stack Overflow has since 2014).
The statement hits home with me because over the past 20 years I have actually gone back and forth between having a wiki as a personal website and now finally back at blog again.
I find that markdown + tags is the best way to organize my personal knowledge base that I call a blog. My attempts at using Wikis always felt overkill.
I still don't really really agree with that either, though. I tried swapping out my simple static blog for a MediaWiki instance and quickly realized why you don't see many people doing that anymore. Maintaining a "complex abyss of ever-evolving thoughts" and actually writing stuff are often mutually exclusive
I use Vimwiki: easy to version control, no database to maintain, familiar (to me) interface. Back up by git push/pull to/from many places.
The implicit invitation to collaborate is what gives the wiki longevity and the possibility of resurrection.
Other things I do that one doesn’t these days but you’d be eager to do in the past is that I’m public about my life. Funnily enough, it was someone else’s comment about Wikipedia deleting their article (which I did manage to recover) that pointed me to a Japanese mathematician. His website filled me with such nostalgia. There were all these stories of his life and things like that.
We used to put things like that on the Internet. The one thing I did miss back then was the ability to make small updates to people’s websites to fix typos and so on. So my website is a wiki (it’s just Mediawiki).
It’s been vandalized before by bots but I make nightly backups to R2 so I just dump and restart if things get ugly. Otherwise, it’s been fine.
One thing that might be fun is if someone one day happens upon my site and feels that sensation of looking at someone’s lived life.
Some of us still do. But I think it was always very much a minority even on platforms that encouraged it. See Similarworlds.com, a successor to the Experience Project.
Any ideas?
Consider writing plain HTML and calling it a digital garden, so you aren't locked into the chronological feed blog mindset.
Maybe Obsidian Publish? https://obsidian.md/publish#:~:text=Explore%20Publish%20site... Although it's a paid service and it seems a little clunky.
* A blog post is a snapshot: what did the author think at the time they wrote the post? If they change their mind or learn more, they write a new post and link forward and backwards. I know how to write for this environment (write what I think now, try to write things I'll feel glad to have written later) and how to work with things other people have written (consider the date, it's just one person's view).
* A wiki page is unclear. When should it be updated? How much should I trust that it was up-to-date as of the last-updated date vs that just being when someone fixed a typo? A few wikis and sites with wiki-like approaches (Wikipedia, gwern.net) manage to handle this well, but I think it's generally much more difficult and rot-prone.
This style is more popular with some media sites — e.g. journal.tinkoff.ru (in Russian) does this with their “instructional” posts — but I think it should work for personal blogs just as well.
I haven't accepted it yet, but I think people who host their own personal websites need to accept that they're hosting a personal website, and it's going to change over time as they change. People already know and accept and see that as a feature of a personal website. It's necessary as the tech changes too, a personal website with the latest and greatest tech from 20 years ago renders like garbage in a modern browser.
I really don't mean to sound mean, and I do sincerely empathise and sympathise with the author, because every year or 2, I have the same revelation that my website hasn't been updated in a while, and it's not my fault, it my platform just isn't technically correct and it's too restrictive and _that's_ what's stopping me being consistent on my personal website. But let's be honest, that's a me problem for not updating it or adding to it.
Every year or 2 for the past 20 years, I'm sure many of us could write the same "<my current website structure> rots. <The one I had an epiphany about a few days ago> wait".
The irony being that one without the "latest and greatest" tech would still render fine - tbh, even in 2005, some of this "latest" tech was recognised as not the "greatest".
Examples? I don't think daringfireball.net has changed that much since 2002, neither blog.codinghorror.com since 2004.
Many of my nerdy friends at the time also had fully flash based, or just a flash animation on the front of their websites.
We also all made liberal use of blink and marquee tags which might still work in some places, but are officially deprecated and unsupported in others. At the time, these sorts of things were considered the latest and greatest. Hell, I remember one maniac playing around with Microsoft Silverlight on their personal site as we were in college. We knew some was good, some was bad etc. None of it survives as intended now though, unless it was updated.
There aren't many personal like websites I remember that have stayed personal sites for over 10 years. But some have, kind of. Take gwern.net or stevepavlina (not a regular reader anymore): both of them - if you squint - look pretty similar minimal 2000 - 2010 style as when they first started, but the implementation has had to evolve as even things like CSS have had a bunch of breaking changes over the major versions.
codinghorror uses Discourse for comments, and that didn't come around until 2014, and the only thing that looks like it might be from 10+ years ago on daringfireball is the content: the tech that renders it and the design all hinges on 2010+ technologies and companies.
I love minimal styles and ideals. But if you want to create a personal website that survives time and looks how you expect in a modern browser, even a minimal one, you have 2 options:
1. it has to either look exactly like some variation of motherfuckingwebsite.com, and that will eventually vary between browsers 2. it takes some effort to maintain over a period of years
Sorry, I wrote the lengthy response above and I think just realised me and the other person who responded to you have misunderstood your comment: I think me and the other person initially assumed you're making only 1 point, that latest and greatest tech wouldn't render, but your examples clearly use modern tech from well after 2004, which makes no sense. On rereading, you're making 2 points, right? 1 point is that tech from 2004 can still work (but you didn't provide example) and other point is that those 2 sites you did provide are example that "website structures" (e.g. linear blog) can survive time but didn't actually state that point. Is that right?
Assuming so, you're completely right on your second point, and people who've done that (maintained a blog for years) are the real legends who've conquered person website anxiety. They're better than many, including me and OP, who have the issue of feeling like they need to completely rearchitect their website every 2 years to a wiki, or digital garden, or knowledge base, or whatever the latest PKM tech buzzword is. This is why I was saying that there's no need to crap all over e.g. blogs for rotting, as your examples prove blogs can survives decades. OP is just another response to that feeling of "uuurgggh I can't quite wrangle my thoughts into a neat, atomic, chronological list of blog posts and people who read blogs will judge my personal website, so I'll tell them it's not a blog".
https://xxiivv.com is a wiki. it’s also very personal.
It's timely that Oxford 'academics' have no idea what 'brain rot' means but Urban Dictionary (A wiki) gets it right.
https://corp.oup.com/word-of-the-year/
https://www.urbandictionary.com/define.php?term=Brainrot
Most Wikis don't work, it'd be interesting to work out what it takes. Starting list - https://en.wikipedia.org/wiki/List_of_wikis
Encyclopedia Dramatica (NSFW) is better than Conservapedia for instance.
I have never seen Brain Rot used consistently with any of the UD definitions, and always consistently with the OUP's definition.
Rotting produces compost, fertiliser. And at the same time makes way for new things. Great, isn't it?
The reasons as always quite complex: from the general decline of the public internet due to centralization / enshittification (and now wholesale appropriation), to poor technology choices and missing value propositions that could induce the next wave of adoption and development.
Yet there is still no tangibly better alternative vision for open source knowledge management, especially if of the collaborative kind.
One interesting direction - yet after more than a decade still largely in embryonic phase as far as broad adoption - is wikibase [2]. It runs as an extension of mediawiki and makes it relatively painless to integrate structured data in a semantic web style (e.g. [3] for an example of integrating veris [4] data).
Its not clear if the wiki era is permanently dead or it just waits for some rain to blossom again.
[1] https://en.wikipedia.org/wiki/Wikinomics
[2] https://en.wikipedia.org/wiki/Wikibase
[3] https://www.openriskmanual.org/wiki/Special:WhatLinksHere/It...
I agree a lot with this. As a person who speaks many languages, something that gets evident very early is that isn't one "The Web", but many webs like English Web, Spanish Web, Portuguese Web and so on.
This is extra noticeable on Wikipedia, since many articles exist in many languages. The drawback is that sometimes information is split across the languages, so someone speaking English, Spanish and Swedish can sometimes build a more complete picture from just one Wikipedia article, if the data isn't in the other article languages.
Enter Wikidata+Wikibase, which makes the knowledge itself trans-language, and instead only the definition/value names need translated, but the composition itself is language-agnostic.
If this imaginary article with separate info in three languages for one article could all use Wikidata as a base, they can all share the same knowledge and make sure that people who only speak one of the languages, come out with the same understanding.
Basically, Wikidata if successful, will multiply the knowledge on the web!
> I’m happy to report that this page (like most housework) will never be finished. It is a living document that grows and matures, just like most of real life. It is not a “work in progress”, for this would imply not much intrinsic value until that magic day it is completed.
> A novel is a work of art that, once completed may continue to exist forever in that finished state. An encyclopedia must be published at regular intervals to reflect new information gathered since the day it was published. Periodicals are timely only when first printed, then fall behind the times – get the latest issue to keep up. The technology behind web documents allows us to update information as often as is necessary. In this context, publishing dates become an outdated concept.
> While it is possible to “finish” a web document, the fixed information becomes stagnant, thus abolishing any desire for a return visit. This is something I call a cob-web page.
From a practical perspective: Blogs may rot, but wikis decay. Larger projects with established community manpower may not struggle with offsetting the maintenance and complexity that traditional wikis demand. For personal writing, however, the burden of preventing decay falls entirely on the author- and it's not a trivial burden. Like others have mentioned, there seems to be an absence of great wiki software offerings that do a great job of mitigating said burden. The few I have tried introduced an inherent complexity and maintenance overhead that significantly detracted from the core activity of writing.
Regardless, I'm hoping that it's just an engineering problem that has yet to be solved instead of an unavoidable characteristic of the medium itself. I would love for the wiki to make a big comeback.
As a result, my project is effectively also a wiki:
https://github.com/gritzko/librdx/tree/master/abc
The idea is to put motivational and explanatory text into the parallel wiki, while all the API docs stay in the code the normal way. These are seriously different things.
The next step to unit tests all the code docs. Or, the other way around, to document tests to make them joy to read. That is the only way to solve doc rot.
Overall, I am trying to get as close to Literate programming as practically possible: https://en.wikipedia.org/wiki/Literate_programming
Because code is hypertext, IDE is a browser.
Just launched https://lmno.lol blogging platform (no tracking or ads here). Read blogs from anywhere. No JS required.
GitHub.io blogs feel like something that came and went, but maybe the SEO wasn't there enough for people to stick with it.
The resurgence of newsletters is also another revitalization of blogs with better syndication (e-mail over RSS).
Probably not a great time to operate a WordPress blog though.
It was a covid project of mine, and it's growing even if many people don't know that it exists.
Also, there are tons of small blogs out there.
https://rakhim.org/honestly-undefined/19/
I'm personally in the top left corner and bottom right corner at the same time, which is sort of funny.
I have used WordPress since 2004-2005, and I've also written a Python static site generator before using Flask + Frozen-Flask[1]. I've also made stops through tools like Sphinx, Hugo, Gatsby, and VitePress[2]. But my personal site continues to run WordPress[3].
I think I'd prefer something like VitePress these days for a technical documentation site. It has a lot going for it for that use case. And it feels built to last.
On true wikis that one can self-host, I recently learned that MediaWiki with a reasonable theme like Citizen[4] is a nice choice for an open source powered private wiki. Although I do find the Mediawiki markup language a little cumbersome versus simpler markup languages like reST or Markdown/MyST in the Python community (or GitHub-Flavored Markdown or Asciidoc supported elsewhere). But Mediawiki has a lot of nice features -- after all, Mediawiki powers Wikipedia. The theme makes it work properly on mobile, adds a little more structure for automatic ToC, and makes content editing a bit simpler.
It still isn't nearly as polished as commercial wiki-like software (e.g. Notion) but it's better than open source wikis used to be.
On the subject of the blog post, I think bit-rot or info-rot is the natural order of things. The kind of software you run isn't going to change those facts. And if you're curating knowledge about technical computing subjects (that isn't about durable topics like, say, C and Linux system calls), you should expect exponential decay.
I do find it kind of amusing how many tools and frameworks developers have created for making it easier to edit HTML pages, though. Truly a foundational 21st century problem that deserves a technical solution that can last for decades without itself bit-rotting.
[1]: https://frozen-flask.readthedocs.io/
