So.. what happened to h0p3?

I used to see them every once and awhile on Reddit and would periodically visit their blog at https://philosopher.life which was a pretty long ongoing tiddlywiki, but now, seems like the account’s gone and their website as well.

Does anyone here know what happened by chance? :thinking:

Edit: albeit a late edit, but due to good news, h0p3 is still around :grinning_face_with_smiling_eyes:
His hypertext works can be read up on https://h0p3.neocities.org

4 Likes

That’s a great shame, I was also a fan of https://philosophers.life/. It made an excellent demonstration of TiddlyWiki. Here’s a snapshot of a version of the site from February 2019.

https://web.archive.org/web/20190201070050/https://philosopher.life/#Root:Root%20[[Legal%20Notice]]

The archive includes captures of the site up until December 28th 2022, but the later snapshots take a long time to load properly, and crash with a RSOE.

5 Likes

Also a huge fan. He followed the lead of https://wiki.waifu.haus/
Thanks for the archive @jeremyruston!

1 Like

I didn’t realize that website was also gone, I actually found it through h0p3, I always thought the way they used ascii art was amazing. reminded me of some of the cool works you can find pictures of from BBS’s

Visiting the sites via the wayback machine and using the download btn allows you to acquire a local copy of it that still works like original, I think I will hold onto a copy of both to read and learn from for archiving-sake

3 Likes

I’ll put the site in the next newsletter as a showcase. Maybe they’ll see this thread or see the newsletter and reply.

2 Likes

this was quite a deep dive having not seen either for those sites before, thank you for sharing! and if either of these people do find this post, thank you for your inspiring wikis!

2 Likes

!! @Justin_H

Does anyone here know what happened by chance? :thinking:

Boop. I’m still alive. /wave.

In a heart-breaking experience, I had to take an escape hatch due to a serious and perhaps on-going threat against my household (I came to realize an expedited restraining order was not enough). I am proud to say I didn’t go offline (though I did drop off the web for a few days during the crisis), but only someone who carefully used the information on my [[Connect]] page would have been able to follow me (to my knowledge, none made it that far). For now, I [[aim]] to limit my attack surface while remaining on the web for [[other]]s’ convenience. I try to maintain a low profile even though I’m obligated to speak in public.

Among other gateways (some more long-standing and available than others), you may find me here: https://h0p3.neocities.org/.

I didn’t realize that website was also gone,

It seems many of my old [[Root Users]] have become hard to find online. I miss hearing from them often.

I always thought the way they used ascii art was amazing. reminded me of some of the cool works you can find pictures of from BBS’s

Thank you. I would like to recommend my [[friend]] [[sideria]]'s TW, https://wiki.sideria.site. I’m looking forward to seeing where he takes it.


!! @jeremyruston

That’s a great shame, I was also a fan of https://philosophers.life/.

I remain indebted to you and this community for constructing the only tool of its kind. There’s still nothing in the same league as TW for my kind of work. Thank you.

It made an excellent demonstration of TiddlyWiki.

Thank you. That’s high praise. I’d like to thank you for allowing me to have a very special relationship with my offspring, [[1uxb0x]] and [[j3d1h]], who each wrote several thousand tiddlers over the years with [[my wife]] and me. TW has been among the most important teaching and communication tools I’ve ever encountered (and were I teaching at university again, its use would be required of all my students). We also spent a lot of [[Family Time]] every week reading each other’s wikis. Those memories are some of the highlights of my life as a father, and it has changed so much about who we are. Thank you, sir.

The archive includes captures of the site up until December 28th 2022, but the later snapshots take a long time to load properly, and crash with a RSOE.

Is there something I should do to solve that RSOE popup? I’ve run into a few people who knew they could download it straight from archive.org for a working copy, but most don’t know that TW is capable of such a thing.

I’m saying nothing new to you, but I think I should say it anyway. Though I wish there were a way to make it load faster (it’s now ~58MB in size after writing basically every single day for seven years), I’m [[grateful]] to have the opportunity to have web archiving tools and p2p filesharing networks archive and distribute the complete copy (well, minus the [[Verify]] problem). Longevity and completeness in digital archiving isn’t easy, and TW’s singefile is so well-suited to it. Because of it, I can live on networks that are reduced to simple file transfer services. I know several people who, like me, aren’t as skilled with computers as you are, and TW allows them to own their platform, even in the shadows.


!! @PaulH

I’ll put the site in the next newsletter as a showcase. Maybe they’ll see this thread or see the newsletter and reply.

I’m sorry I’m late in responding to the beetlejuice (I’ve been careful in responding elsewhere, including scrubbing). I just stumbled upon your words today in a routine rabbithole. I was speaking with [[ChatGPT]] about why TW probably can’t be completely re-written in a language compiling to WASM given that the language may never have direct access to the DOM for security reasons (which isn’t to say there’s no optimization possible). Went hunting a bit and saw the newsletter.


!! @All

Given what little I understand about WASM, unless unlikely important changes toward open and increased user agency are made by the G-entity, I no longer believe I will be transplanting my work into a fully portable Linux environment for the web. I should probably work toward doubling down on TW at this point. Does that sound about right to ya?

I need guidance. Feel free to ignore anything I say please.

I haven’t been able to update my TW without breaking it (and I have no idea how to fix it; I just write in it at this point), and I also continue to use a correspondingly old copy of TW5-Bob because I’ve grown quite dependent upon the ability to access it over multiple tabs (unfortunately, it has grown unstable over our LAN - and I have little doubt it is because I have not been able to upgrade). The upgrade is beyond my skill given the time I have. I’m privileged enough to be able to pay for help now, and I was wondering if you think it’s feasible to find an expert to help me upgrade my TW. We’re pretty lost here at [[The Alien Asylum]].

I’m also reaching a point where there are almost no new users who don’t complain about the performance of my wiki, which is mostly a downloading problem. And, again, I am grateful to neocities who generously provides old school web hosting to a random [[pseudonymous|anonymous]] person like me. I am not complaining, I’m only stating the facts as I understand them: their single-threaded upload throughput may be bottlenecked, which exacerbates the wait. Loading over a tor hidden service or an i2p eepsite has grown egregious enough that one begins the process of downloading and then walks away to complete chores before anything functional loads (and, yet, I prize the fact that a reader has complete privacy and offline-usability in these cases). I don’t mind that we must be patient, and waiting to load a site isn’t egregious to a person from the dial-up era. I’m hoping there might be some improvements to make (I’d be surprised if newer versions of TW5 weren’t computationally more efficient as well), especially as I anticipate the wiki will only continue to grow in size.

Though I prize singlefile, I’m wondering if it’s possible to have something like singlefile lazyloading. [[This|https://www.reddit.com/r/TiddlyWiki5/comments/11ld4zt/can_a_singlefile_tw_read_data_from_an_external/]] is pretty close to the basic function I’m hunting for, I believe. Having something functional beyond a background gif might be useful, but I don’t know if that’s feasible. It might provide an opportunity for multi-threaded downloads too. It might just be what it is, and there’s gonna be prices to pay somewhere.

One route I have wondered about before is zstd with a custom dictionary. I’m currently testing to see if I can beat gzip by a serious margin.

I’m going to be archiving my [[Link Log]] (close to 50k links), and, for those who would seed the mutable torrent (or rip the whole archive through other means), I’d like to make it consumable from within the TW. Is there an ideal way to do this, or any gotchas worth considering?

Have you encountered anyone fine-tuning LLMs with their TW?


!! @Nicolas_Sivridis

Also a huge fan.

It’s a pleasure to meet you.

He followed the lead of https://wiki.waifu.haus/

I’m [[lucky]] to encounter many who start and continue using TW to have a long-term discussion with me. I met Madame Senpai [[chameleon]] in 2018. After coming across my work a couple times, she began speaking with me over a TW in [[2019|2019.07.16 - chameleon: I’m Sorry]] until our dialectic [[faded in 2022|@: chameleon]] (you may find the nearly complete transcript of our interactions over the years sprawling in my wiki). I treasure my copy of her work, and have diligently archived her words for us. I hope one day she will come back to write in public.

GitHub - whacked/tiddlywiki-org: TiddlyWiki 5 + emacs + org-mode = awesome is dope, btw, and I think she would like it as well.

I’ve [[recent]]ly encountered someone who has millions of words written, and I’m hoping [[they|Feel Good Lost]]'ll continue to reveal themselves through their [[TW|Public Self-Model]].


!! @Scribs

this was quite a deep dive having not seen either for those sites before, thank you for sharing! and if either of these people do find this post, thank you for your inspiring wikis!

It’s my honor. If you ever happen to [[share]] yours, I’d be glad to read it.

5 Likes

Hi h0p3 – And welcome !

I did download the wiki and it’s close to 60MB because you did include a JPG image lately, which “costs” about 500kByte in TW. The JPG itself is about 239kB, but if embedded in the wiki it needs double the space.

With a bit of compression optimization we can get it down to 63kB, which is still about 120k in the wiki.
If possible images should be “lazy loaded” with external images - _cannonical_uri.

There is the animated background GIF which seems to need an other 800k or so. If not animated it needs 46kB.
If it should stay animated I’d also use it as an external asset.

There is an animated favicon, which needs 60kB for a 16x16 icon. It should need ~5kB for 128x128 or so not animated.

TW may contain a small static favicon. If there is internet connection it may be possible to load it from the server. – Not sure about that, but I think TW requests one at startup.

With those optimizations I got it down to ~58MB again. There are 17k++ tiddlers which is impressive.

I did not find any more JPGs, PNGs or GIFs … If you know some, there may be more potential to get the size down.

But 60MB / 17500 tiddler is about 3400 Byte per tiddler, which seems to be reasonable.


The wiki core seems to be TW v5.2.7, which is OK. I would wait until v5.3.1 is available anyway. I did not test your wiki with v5.3.1-pre. There may be some problems with plugins. I did not have a look at your plugins. So not sure here

just some thoughts

have fun!
Mario

3 Likes

Thank you, sir. I’m [[grateful]] for your assistance. I also want to thank you again for your excellent Firefox add-on. It is the first thing I recommend to others, and I still use it for my password/private-data manager today. Thank you.

I did not know that image embeddings take double the space. You’re a good hunter. IIRC, there are only four non-text-art images embedded (the remaining one is as optimized as I can get it, and its embedding exists as a gift to someone dear to me). I bet [[Wiki.gif]], the background image, could best be slimmed down into an animated svg. You’re right: there is low-hanging fruit to get it down a bit further.

I transmit my ℍ𝕪𝕡𝕖𝕣𝔱𝔢𝔵𝔱 as an unarchived singlefile often enough that I’d rather not begin using external assets unless I were able to lazyload the vast majority of the tiddlers in the background. I’m not even convinced this is feasible in TW as it is. Ideally, only ~3-4MB of data would need to be downloaded (and this could be split into at least 3 different files, at least two being images on the default front page) to at least have a partially navigable site while the remaining bulk of the tiddlers load in the background. Presumably, at least some macrocalls, essentials like the searchbar, and who knows what else likely require every tiddler to be loaded before the wiki would be fully navigable. If this were feasible, then decomposing the wiki from singlefile into external assets would be more worthwhile.

But 60MB / 17500 tiddler is about 3400 Byte per tiddler, which seems to be reasonable.

I agree. In fact, Tiddlywiki’s ecosystem might even be unreasonably effective at what it accomplishes (I don’t even have a word for what it is, despite searching for one for years). I am still surprised by just how much is packed into that ~60MB, and I endeavor not to waste much space.

I would wait until v5.3.1 is available anyway. I did not test your wiki with v5.3.1-pre. There may be some problems with plugins.

Yes, sir. Thank you.

That’s great news, I’m glad you found the thread, and your still kicking @hop3 :grinning_face_with_smiling_eyes:

Though, I’m sorry to hear about your situation, I can only hope it gets resolved and you can return to your own normalcy.

If you don’t mind me asking, since the last time we spoke, I’ve thought of one or two questions I’d like to hear your input on!

  1. what advice, tips, or insight would you offer someone less experienced with using a tiddlywiki for long term use? like naming conventions, structuring, that sorta thing

  2. If you were to give a tour of sorts of your wiki, where would you recommend someone to start? I usually click through at random and just sort of read as I go, I’m afraid I haven’t deciphered your tabs yet haha

  3. What got you started with your wiki?

As far as the size of your wiki… have you coinsidered segmenting or archiving parts of it? segmenting it into different sites and navigating between them like a traditional website might cut down on load times further if you do go the route of mario’s recommendations.

Though, keyword might- I know from my own experiences with TOR nodes, no matter how optimized the website, there’s always going to be a bit of a delay on loading due to the design of it. A worthwhile tradeoff if you ask me :man_shrugging:

1 Like

since the last time we spoke

I didn’t mention it, but I had two guesses (I think you’re the only one to have talked with me about https://talk.tiddlywiki.org/ until recently).

what advice, tips, or insight would you offer someone less experienced with using a tiddlywiki for long term use? like naming conventions, structuring, that sorta thing

I’m going to write muh long-winded gibberish here, and please feel free to ignore it. I’ll do my best to answer your questions, but I doubt I have any cleanly satisfying answers for you.

I have been asked this many times over the years, and I’ve thought about it plenty throughout the wiki. I still feel unqualified to answer it well. First, I’m not programmatically skilled with TW or web development, and I think my skiddie opinion about your question deserves some doubt for that reason; it may be a far more useful tool in the hands of many others I’ve encountered over the years. I can also say that many people don’t have the same goals or logistical requirements that I have in mind, and that may significantly influence the fitting answers to your question. I’m a zealot, including about what I think consists in making [[good]] use of this tool, so my opinion should be taken with an extra huge grain of salt. Text is king to my eyes, so I probably don’t value what many might about TW, nor do I stick with it for reasons that most would. I think TW can be exceptional for readers, ownership of the means of production (including distribution), those with strong spatial reasoning and memory, and non-technical writers (including myself). TW has just enough batteries-loaded boxes to help you be creative and perhaps more likely to default to doing close to the right thing to efficiently find the right way out of those boxes through use, but I think its flexibility is part of what makes answering your question generally or definitively pretty difficult.

For some people, if they aren’t publishing with others in mind, I think mostly plaintext files with a proper text editor and strong command line tooling is better than Tiddlywiki (I have little hope for browsers, the next gen cross-platform terminals, at this point, though I am an incompetent fan of [[Nyxt]]), especially if they’ve understood what they like about Tiddlywiki. For my offspring (who have each written ~10MBish public TWs before the age of 14), I would prefer they pour their time into their [[OS|NixOS]] and something like neovim, sublime text, or emacs, with custom tooling from the ground up for their private text work. For those willing to put in the effort (work which is minimized by LLMs at this point), TW is often worse for an author, archivist, or data transformer than other more agnostic tooling options. At some level of investment, I think the non-author is the primary beneficiary of TW.

For naming, I tend to stick to what I call dynamic and static tiddlers:

  • Dynamic examples:
    ** Primary Tag
    ** Primary Tag: Title

  • Static examples:
    ** Date - Primary Tag
    ** Date - Primary Tag: Title

Dynamic tiddlers are like the clumping garden I grow, and static tiddlers tend to form the streams. That’s probably an oversimplification, as I see streams of chronological thought given sets of my dynamic tiddlers, too.

If I understand correctly, technically, you can do all your [[tagging|Tagging Theory]] and fieldish work in the title (though, I have some very long titles, and I’ve found it sometimes problematic). I prefer tiddler titles that encourage me to crystallize and recompress the content, provide footholds while searching lists, and increase the odds that I’ll remember.

I think there’s something to be said for minimizing your orphans and working toward connecting all the dots, but I’ve sort of a unificationist approach. One commenter has said that I’m trying to stuff an entire universe into this document.

Of course, I’m more likely to ask you why you are doing anything at all. What’s the point in using TW or your computer? The how matters far less than the why (or in virtue of it). It is there that I would suggest a person write every day. Don’t play a toolporn game. Don’t worry about making it beautiful or even fun. It’s not so much about productivity or collecting objects, and far more concerned with engaging in an explorative and transformative process itself. Work on what matters in your shoes, even when you don’t want to. Your TW is a communication device with yourself and others. What’s worth thinking about, holding yourself accountable to, and [[reconstructing]] between yourself and others? Find and do that. That is my primary advice, [[WINTCIS]].

If you were to give a tour of sorts of your wiki, where would you recommend someone to start?

That’s a really hard problem. I’m probably not going to be terribly useful to you here, like I’m stacking gibberish on top of more gibberish. /smh.

The [[Readme]] is my first suggestion. I wish I knew how to give an ideal tour of it even to any given individual, let alone larger sets of people. I’ve seen my fair share of people who say my work cannot feasibly be summed up or categorized (I’m not claiming that, but I can appreciate why it’s hard to describe and navigate). There are also places meant for experts or individuals, and I’ve had to make perhaps absurd assumptions about what my reader understands or can parse. I’m on the hunt for what matters most as best I can, and I [[hope]] my readers and interlocutors do the same in their exploration.

I usually click through at random and just sort of read as I go, I’m afraid I haven’t deciphered your tabs yet haha

Consistently reading [[New]] or [[Recent]] will make it so you don’t fall behind, but catching up is no simple task (e.g. even peeled apart from its context in the rest of the wiki, digesting just the [[Link Log]] alone is a monumental task in itself). I think of the work as having a clear linear timeline (paranoically so). It has a very strong journal component to it, and I don’t think the work can be well understood without that mindset and interpretation. The reasons to wander it non-linearly are likely highly particularized to one’s context.

I can’t say all my tabs have retained their usefulness over time, and only some of them are used every day. I’d very much like to see how other people organize their sidebars. It’s hard to do well, imho.

In the end, I think artificially intelligent beings will be the only agents to cover all the ground thoroughly multiple times (including the snapshots). I think they will provide far better bridges into, analysis, impersonations, and translations of my work than even I could. I’m looking forward to the evolution of [[Aispondence]]; I’ve been writing with machine learning in mind for a long time now. It’s happening much sooner than I anticipated though (I continue to be wrong or at least inaccurate again and again).

What got you started with your wiki?

The first tiddler was [[2016.10.17 - Letter to Mom and Dad]]. [[⧖]] picks out the larger chronologically written story of the entire wiki (most every tiddler is hardcoded into it [upon each manual update], and it still takes ~3 seconds to load, compared to the macro that takes my machine ~40 seconds to generate), and why it continued and evolved as it did (as do the daily snapshots). At this point, I see my ℍ𝕪𝕡𝕖𝕣𝔱𝔢𝔵𝔱’s reason for existence as being causally overdetermined; there have been and remain multiple sufficient reasons for me to continue my work. How I got to the place where I began writing my ℍ𝕪𝕡𝕖𝕣𝔱𝔢𝔵𝔱 (originally h0p3’s Wiki) is a long story I’ve attempted to flesh out over the past 7 years. Unfortunately, it’s quite expensive to read (besides myself, there are probably only 3 people who have read even half of it).

As far as the size of your wiki… have you coinsidered segmenting or archiving parts of it? segmenting it into different sites and navigating between them like a traditional website might cut down on load times further if you do go the route of mario’s recommendations.

I have considered it. I think that would defeat much of the purpose of it. I’d probably be more inclined to export particular sections as their own child projects. I work hard to limit how my words can be taken out of context in a variety of settings. I’ll agree, however, that for almost any given problem, if there’s anything useful in what I’ve written at all, most of the wiki is either seemingly unrelated ephemeral appearing cruft or indirectly serving as background language use for definitions and context.

Though, keyword might- I know from my own experiences with TOR nodes, no matter how optimized the website, there’s always going to be a bit of a delay on loading due to the design of it. A worthwhile tradeoff if you ask me :man_shrugging:

Agreed. The latency trade-off is necessary. Though JS is not recommended for the Tor browser (and, some folk deny it should ever be used, XD), Whonix does a fine job. Once it is loaded (curl over socks5 is solid), TW performance is a clientside matter, and, for longer sessions, TW demolishes most of the other options for serving over Tor. That is one of the reasons I have chosen to use this tool: it doesn’t punish people seeking [[anonymity]], [[privacy]], or offline control. I’m a long-time contributor to the Tor network, and I’ve found very few sites last; TW makes it easy to archive and own the means of production, serving the author and the user well, imho.

4 Likes

You are welcome and thank you for the kind words.

I’m sure you regularly back up this data to a different physical drive, even if the File-Backups plugin creates several backups. But they are on the same drive. – Just to be sure :slight_smile:

I will send you a download link with the modified files in a private message, so you can play with them yourself. It’s the Wiki.gif and the Madmans’s Photo Slideshow.jpg .

I did use the https://squoosh.app/ which is a single page app. Similar to TW it does not send any data to the web. Everything is done locally in your browser.

I did use the MozJPG compression with Quality 66. I did set Resize on - to Custom with 1013 x 635 px. Which is 2/3 of the original size (1536x963) – The file size reduction is 74% from 239k to 63k

The animated Wiki.gif is 245kB – Removing it from the file reduces the size by 652kB, which is 21/2 times the size for the embedded image. It seems to be the animation that makes it that big.

So an animated SVG should be a viable option and should be significantly smaller . – I will send you a static Wiki.gif with the same dimensions with 46kB file size. Embedded it should only need 90kB instead of 650kB

I did not modify the $:/favicon.ico which seems to have ~50k embedded. I did just remove it to calculate the difference. So you should create a static one, with a 128x128 resolution as PNG or GIF, which should have about 5k or so.

As you wrote, those are “low hanging” fruits. … But I also think those are the only optimization options, if the rest it text only.

Your server does send the wiki in a ~20MB compressed format. The browser says uncompressed it’s ~61MB. – Since images cannot be compressed very well, optimizing the gif and JPG will also result in about 1.8MB less, that needs to be sent over the wire.

So doing some more math. 1800k reduction / 3.4k per tiddler = 529 new text based tiddlers.
So it should be worth about 1 more year with the same size as it has now. … In my books this would be a win.

have fun!
mario

1 Like

I’m sure you regularly back up this data to a different physical drive, even if the File-Backups plugin creates several backups. But they are on the same drive. – Just to be sure :slight_smile:

Yes, sir. That’s a good point. There is a serious danger of losing one’s password manager data.

rsync scheduling: NVME1->NVME2 (with long-term btrfs snapshots otw) nightly, NVME1->HDD1 weekly, HDD1->HDD2 monthly. I use Resilio Sync with short term remote archiving for my mutable torrents of key data across devices (encrypted backups can be useful) as well.

I will send you a download link with the modified files in a private message, so you can play with them yourself. It’s the Wiki.gif and the Madmans’s Photo Slideshow.jpg…In my books this would be a win.

Yes. Thank you, sir.

But I also think those are the only optimization options, if the rest it text only.

/nod. Everything else I can think of is an ugly hack by comparison.

@h0p3 Very glad you responded sir!

kyr0ss

1 Like

It’s good to see again, sir. I hope fatherhood has been excellent to you, and I’m glad to see you writing in public.

1 Like

This is a pretty nifty tool, shame they don’t have a download button for offline use, but I’ll save this to my bookmarks for sure.

Hey @h0p3, question for ya! (Well, really a couple, I suppose…)

Do you type all your responses into your wiki first, or transcribe your interactions into your wiki as they take place? And, do you do //all// your interactions, or just specific ones? I’m curious of the criteria, and how it started off. Did you go back to previous posts and add them to your wiki?

Also, do your root users act as a sort of honourary vip or more as admins to assist with the wiki?

And, does each root user create their own ANSI art circle?
Their pretty iconic when it comes to your wiki, so learning about them would be cool. I’m not that old to be quite honest, so early 2000’s era weblogs and BBS chatrooms with ANSI art is a favorite topic of mine to look into.

1 Like

I’m drafting for you now on the site.

Awesome, I’ll wait for your draft to be posted here :grinning_face_with_smiling_eyes:

Seems you are using a large wiki too. Here is tiddler count of mine A Pre-Plugin: Colorful Graph Macro Showing Tiddler Counts - #3 by linonetwo

And online version is https://wiki.onetwo.ren/ , I serve it using nodejs tiddlywiki’s lazy-all mechanism, so it is faster. And I use TidGi mobile that is lazy-all by default, so open it to write some quick not don’t need to wait it open for a long time.