Subject: Oh dear
Author:
Posted on: 2021-06-27 16:41:27 UTC
I, uh, don't understand coding and whatnot too well, but I do remember most of my writings I've published over the internet being in the form of Google Docs links, so... Welp.
https://slashdot.org/story/386970 (original source: https://workspaceupdates.googleblog.com/2021/06/drive-file-link-updates.html?m=1)
TLDR, google is changing how Drive sharing links work. This may break some or all links to documents (they're being vague), causing them to only become usable by people who have used them in the past.
We have a lot of missions stored as google docs. I suspect this won't hit missions that were published to HTML from docs and shared that way (as that would make no sense, although it is possible), but if the raw docs link got passed around, it's very possible the mission is about to drop off the web, and would need to be reinstated by the author. Which is an issue if the author is no longer present.
So... uh... a lot of things need to get archived. Before the 9/13 deadline. Or we'll lose stuff.
Google is now telling me which files will be affected, and it's a pretty small collection. On my main Drive, there's only 16 in total (and I have a lot on there). It looks like they're mostly images, plus a few PDFs and a couple of archived HTML files. No Docs, Sheets, or anything else standard.
My PPC account is even better: there's only four files in danger, all of them images. I don't know what Google's selection criteria are, but I think we're pretty safe.
hS
I don't seem to have anything affected, but then I don't have a lot shared from Drive, and what there is is likely published.
I'm still concerned about docs with URLs that end in /edit and variations on that, and there are an awful lot of those from Iximaz, SkarmorySilver, Zingenmir, and others. I'd love to hear from you or anyone else with un-published docs about what Google has to say about your stuff.
Also, a request to all Docs users: Please credit yourself as the author by name somewhere in the doc. If it's published, there is otherwise absolutely no way to tell who it belongs to within the doc. I'm begging you, please please please do this.
~Neshomeh
I was intending on just popping in, seeing how the Board was doing (pretty dead, sadly, but it seems all forums outside Reddit are dead nowadays) and popping out. However, it seems I can actually be of some miniscule use. My missions are, admittedly, rather cringe, but for the sake of the archive I believe they should be preserved. Unfortunately, I don't know what y'all need me to do. What's the archival process for G-Docs pages? Do you need them in a particular format? I saw the spreadsheet, but I didn't see any archival procedure written there.
Also, for old times sake, a Discord link would be appreciated. I don't really use Discord much more than I ever use any social media, but seeing some old faces and new recruits would be nice.
Good to see you! Nostalgia seems to be in the air this year. I hope it continues. ^_^ All is well with you, I hope?
Re. Gdocs, fortunately, I believe this is an easy fix. For each doc, simply select File --> Publish to the web, and that should generate an HTML page with a stable link ending in /pub.
We think these will be stable, anyway. To be extra-sure, you can always save pages at Archive.today, too. This is a little tedious for Gdocs (Archive.today preserves as much of a page's function as possible and there's quite a lot to preserve in a Gdoc, so it takes a while), so anyone and everyone pitching in would be most welcome. Just check to be sure someone else hasn't saved a particular page first. {= )
Oh, and while I'm at it: Gdocs users, for the love of Kanun and Spelin, please attribute yourself and any co-writers as the author(s) by name somewhere in your docs! There is otherwise no way to figure this out from a published doc alone, and it's giving me anxiety. I mean, I know who wrote which agents for the most part, and I can usually figure it out when I don't, but not everyone is me. {= P So, please, for posterity.
~Neshomeh
Oh, I've been fine. A little older, a little wiser, and full to the brim with existential dread. Haven't read any fanfic in years, but I saw the link to the Board saved in my favorites and decided to hop in a sec. Can't say I'm going to stick around, but it was nice to see that some old friends are still active.
Right, archival. So, I just published the three pieces I wrote from G-Docs. I think it's best to let the host RC #901\1Y page die; nothing important is there, and the wiki page serves the same purpose. I'll see about going through the archive.today process soon, perhaps with the published webpage to save some time. I'll edit the wiki pages soon with the updated links, but it should auto-redirect in theory, yeah?
Here we go, the actually proper archived links.
Prologue: Of Daisies, Bronies, Ponies, and Plotholes: https://archive.ph/nJl8P
Death and Ressurection: https://archive.ph/rkjvZ
The Lost Land: https://archive.ph/20BIA
Not entirely sure how viable this would be, but having a single website archive to store everything as a dedicated "PPC Stuff" site might work? There's some wordpress and google sites which have been defunct for literal years, but are still up. Having things downloadable as a PDF somehow might also help a bit? Not sure. This seems like a tricky problem!
Publish to AO3, or a website, or something.
If you must, use Google Docs' HTML publish feature (which might actually be affected by this, but really shouldn't be). And keep backups.
Also, I don't know what the "Google Docs' HTML publish feature" is, or how to use it.
Way back, on two separate occasions, the DOGA website was taken down by its hosts: once because a badfic author strenuously objected to their fic being missioned, and once for undisclosed reasons which the host thought might be the use of profanity. (The first incident led to my only sequel mission; the second to the Department of WhatThe always being named that.) This is why my stories are on GDocs: they're less likely to go down.
FFn famously had the same issue, though I've never had trouble with my Driftwood missions. AO3 purportedly - and currently - doesn't, but... for how long? Websites change.
As, apparently, do Google. Is there any stable hosting solution? Other than getting stuff into the Wayback Machine as quick as possible?
hS
It was founded by archivist types. They don't take things down without cause. Actually, they frequently don't take things down with cause.
archive.org and the wayback machine are also good options.
Unfortunately, if you're not willing to pay for hosting, you can't guarantee stability in your website: free hosts are rarely the most economically stable. Neocities is currently the best option there, but for how long I don't know. The good thing about a website though, in contrast to google, is that if you make sure to use a relative path structure when you write it you can just drag and drop it onto any webhost and it will work. Google docs is more locked to Google's ecosystem. Exporting is possible, as is, through the publish to web system, getting basically a website (which you can then download and modify until it can be distributed via any webhost...). I think the google publish to web links will actually survive this (browse to one in incognito mode after all this goes down and see...), backing up those is insurance. But passing around raw docs links is fragile.
Generally speaking, Google has a habit of discontinuing services, though. I don't trust them. Although I'm pretty sure Docs makes too much money for them to get rid of it right now, I wouldn't be shocked if publish to web eventually goes away (although, who knows, maybe it won't).
In short, no, there is literally no such thing as a stable host. Servers go down, websites crash, companies go out of business, multimillion dollar conglomerates decide they can't be bothered to support your usecase.
Now I know what you're going to say, and linguistic shifts are an issue, but we know that mathematics is essentially a universal language, so all we need to do is figure out how to say "Agent + Suvian = dead Suvian" and we're golden.
^_~
h--
no actually, actual content: AO3. How is it supported and why is it so stable? There were dozens of fanfic sites of varying degrees of specificity; why has this one remained stable when most of the others - even the ones by devoted fans and writers - have faded away?
hS
Maybe I could try doing that if you want, but my handwriting is not the neatest (it's legible, but wiggly) and I don't have unlimited time. Manuscripts would take up less space and be easier to make than rock carvings, though.
Now I'm half-tempted to ask him to ink the first page of "Rambling Band" for us, just for the fun of it. And then one of us might have to try our hand at illumination for the full effect. Who doesn't want an excuse to play with gold leaf, right? {= D
(I doubt this will actually happen, but it would be pretty funny to see something so silly treated so seriously!)
~Neshomeh
A frontispiece of the Fellowship poster done in medieval style, with J&A as bow-wielding figures flanking it. A full-page incipit like this one of "It's happened again". Elaborate illuminated capitals for scene changes, showing things such as Laurel playing Led Zep on the harp, or J&A playing cards. Speech in red ink to stand out from the black. The author's notes as jotted marginalia. And a full endpiece of the Death of Laurel as a reversed Pietà, with Legolas just looking up in confusion while J&A look sardonically on.
I reckon calligraphy runs to maybe 100 words a page, so it'll only be... well, with the illustrations... 40 pages? A nice quick little project. >:D
hS
You are giving me ideas. I do modern art/collage/drawing for fun, so now I'm thinking about doing a modern art illumination of something. Yes, I know that sounds weird.
(TOS being the original series, Jay and Acacia's original series of PPC missions that started it all)
You should check it out. It's... kinda a big deal around here, seeing as we built a whole community around it and all. :-P
Firstly, OTW (the Organization for Transformative Works, the not-for-profit that runs AO3, among other things) has attention, and thus donors. OTW also gets attention and support from other nonprofits. Because OTW is nonprofit, AO3 is not obligated to make money, and because they're run by donation they are not beholden to advertisers. AO3 was designed with the express intent of being an archive, and the content rules are lax. There's not a lot that can't be posted to AO3, and very little that is grounds for removal. The founders have spoken on this and they basically said this was a deliberate choice. AO3 is intended to preserve fan content regardless of its type or merit. And of course AO3 is run by volunteers, and everyone involved is pretty passionate about keeping it as a refuge and archive for fanwork.
So basically so long as the OTW doesn't suffer a hostile takeover and they have enough money to keep the lights on, there's not a lot that can go wrong. So, y'know. Donate if you can.
So, not sure how helpful this’ll be, but we were discussing this on the Discord and I figured I’d make a spreadsheet for posting original versions plus Wayback Machine links. Currently, anyone with the link can edit, though I may change it so anyone can comment but only certain people can edit.
(Also, if anyone else is more qualified, I can switch ownership of the spreadsheet. Just figured I’d make it to give us some sort of head start.)
I can pull a list of original URLs from the wiki. If other people help find archival links and/or updated ones post-change, that will be dandy. I'll need it when it comes time to replace them with AutoWikiBrowser.
~Neshomeh
I, uh, don't understand coding and whatnot too well, but I do remember most of my writings I've published over the internet being in the form of Google Docs links, so... Welp.
According to Slashdot, "Users who have shared a file that is affected by this change will get an email from Google informing them of this change and how to opt out of needing those files from being updated. These emails will be sent out to users starting on July 26th. ... Users have until September 13th to decide if they want the update applied, so if you have no files that are publicly accessible, then you won't need to do anything."
So, keep an eye out starting July 26, opt out any files you want to remain public, and you should be fine.
In the meantime, actually using Publish to Web for files you wish to remain accessible seems like a good idea. It makes them load faster, too, which is nice on the reader's end. Also if, say, someone of an archival bent wants to scan through it multiple times for mentions of canon characters so the mission can be listed on the characters' wiki pages. Hypothetically. ^_~
~Neshomeh
Why do they keep changing things? -_-
It looks like my auto-archive trawl a year or so back correctly captured both published and raw-doc links, so they're technically archived, but for example the DAS-JIVE 1 front page looks like this; it's rescue-level archiving, not a published one. And I haven't updated it in the last year, either.
hS
I don't even know how many hundreds of raw Gdocs links are currently in use, but it's more than a few. And it'll be annoying for me to work out which ones are broken, since I've probably opened them all at one time or another.
At least I don't have many of my own to worry about; I never much liked Gdocs as a hosting solution. I might have a few things that didn't fit well elsewhere, but shouldn't be many.
~Neshomeh really didn't need another project right now.
...For mission listings that are on the wiki, anyways.
We'd need to crawl each page and find pages that contain links that go to docs.google.com (optionally filtering to a URL format for specifically raw docs rather than web published ones), and then print the relevant pages and links. As updating would involve fetching the doc, converting it, getting it on a webpage and relinking it, that'd be harder to automate, although it's sort of technically possible.
Fortunately for us, Fandom is... actually just MediaWiki. The same wiki software that runs Wikipedia (they were actually both founded by the same guy...). And running the most popular wiki software on earth means API documentation is ample. The hard part is getting a list every page on the wiki. Once that's done, you can request the HTML, and with that parsed you can use XPath or your regional equivalent to fish out google docs links (or regex, if you don't mind considerably more fragility).
So while this is nontrivial, it is actually doable. It might be a little while before I have the time and spoons to go do it though. Sorry.
See the link in my above post, if you can—I'm not sure if that's an admin-only feature or not.
And I have AutoWikiBrowser to generate a pagelist and find & replace links semi-automatically.
The tricky part is getting a list of unique links, since the Fandom tool lists every time a link is used and the page(s) it's used on, and then making sure I have an appropriate replacement when the time comes.
Or, since some of the currently broken links are broken on purpose because their authors deleted the docs, actually removing them, which can't be fully automated because it'll involve rewriting bits of articles. On the plus side, I suppose it's high time this was done anyway, much as I don't care for the task. {= (
~Neshomeh
If you can give me a list of links associated with the pages they go to (ie, text, some format with page followed by link or link followed by page), I absolutely can consolidate it to a list of unique links with the list of pages each link appears on.
