• Any/All (Genderfluid)

🍁Furry🍁

Genderfluid || LGBTTQQIAAP+ Ally || Bi/Pan || Poly || Feminist || PLURAL AF || Actually Creetur Shaped ΘΔ& || Table-top and gaming nerd || 3D Enviro/Asset Modeler and Surfacing Artist || Frequent Writer and Lover of Prose Poetry || May be Skunk Brained || BEWARE my content can be NSFW. 18+ 🔞 || Twitter Migrant

RIP Dogbomb
1963 - 2019


Natandthespectrum on Dreamwidth
natandthespectrum.dreamwidth.org/
NatAndSpectrum on Mastodon (Chitter.xyz)
chitter.xyz/@NatAndSpectrum
NatAndSpectrum on Mastadon (Floofy.tech)
floofy.tech/@NatAndSpectrum

staff
@staff

happy friday july 26th! here’s the financial update from june.

i’m sure you all have some questions.

Q. Why did this take so long?
A. honestly? thought I already did it. that’s on me.

financial update follows.


RoxannaRachnid
@RoxannaRachnid

Oh, both of those new features kick ass :O


You must log in to comment.

in reply to @staff's post:

at launch, these synonyms will be managed by us with suggestions made by users

No offense but this just seems...unsustainable. Even if it would be made more manageable in the future with an improved system, given the current turnover rate for handling bugs and improvements that could be a while, to put it kindly. Granted I'm sure you have an idea of what you're doing, but I can very easily see "we'll manually manage the designations of every tag synonym" being something that takes much more time and effort than initially expected, and both of those things are already at a premium for a organization as small as cohost.

i mean, they clearly don't intend to manage the tag synonyms themselves forever.

i think the logic for the rollout is this

  • tag synonyms and user-managed synonyms together would be One Big Feature, but they can break it into two smaller features
  • by breaking it into two smaller features, they can get it at least partially out sooner, allowing them to receive bug reports and fix any issues that they didn't catch in testing
  • as a bonus, while they won't be able to hit every tag synonym themselves obviously, people can at least start to use the tag synonyms and that's an immediate benefit to everyone
  • tag synonyms should only need to be defined once each so it's not like it'll be an ongoing maintenance task. they can add new synonyms as they find them
  • and then eventually they'll have the user managed part in and they can go basically hands off

it makes sense to me at least. it's also not like adding all the synonyms is mandatory for the site to be functional and usable. it's an added convenience that will only get more convenient over time. which also means they don't have to go and add all the synonyms they find themselves. it's something that can easily be put to the side for higher priority tasks while the user managed part gets implemented.

i mean, they clearly don't intend to manage the tag synonyms themselves forever.

Well yes, but the question is; how long would it be between when synonyms are rolled out and when the user-managed system is rolled out, given the current backlog of bugfixes and features. The post does say specifically that they plan to add the user-managed system "over time", which makes it sound like it won't be a thing right from the start.

And yeah, it isn't mandatory, but if they are going to specifically take user suggestions it would be courteous to not put them aside for ages, especially if they're announcing it in an infrequent update post.

That's probably why it's temporary.

Even if it ends up stagnating it's still a strict improvement over not having them. Possible downsides (incorrect synonyms that need to be un-synonymed) are likely rare and able to be handled quickly.

as the other comment hinted at, 10% of the work would be 90% of the coverage here, so it's okay if we're not perfect at first. this is a space that will mostly fall under my responsibility so updates should be quicker than the standard turnover time for development. we're building out an infrastructure to allow that.

but you are right, we can't do it forever! it'll be on the side for me, incrementally improving until we have a better system that can really go yard.

Doing this also means you end up pushing a lot of unintended meanings onto users posts. For example, if I post in #Riven, from a code standpoint you have no idea if I'm posting about the sequel to Myst or the League of Legends character.

yeah it sounds to me like the wrong thing to start doing now. a lot of work for something that can be mitigated in other ways (a solid on-boarding lesson for new users on how to use tags, tutorial type things on the site to guide new users - tumblr these days asks "are you sure you want to post without tags?" etc) until there's the actual time and resources to work on the less manual system for it....

Can the absence of any discussion of the site's financial sustainability (…other than the 58% increase in monthly losses) be interpreted as the site currently not being in any short/medium-term financial danger?

Also, hell yeah about both of the major new features you're working on, seriously!

this "we'll do the tag wrangling ourselves" part deeply worry me, as a person who has tag wrangled. 100 tags could easily take 4 hours and beyond to wrangle; checking the info, double-checking to make sure everything is getting wrangled correctly, clicking all the boxes, syncing the things... like all that takes a lot of time and effort, and it's very fiddly work.

it's extremely time-consuming, and i imagine there's a lot of tags at this point. (it can also be extremely frustrating if you're not the kind of person who thinks it's fun spending five hours linking relationship tags to the correct character tags.)

i worry that staff underestimates how much time they'll have to spend on this, and how much energy it might take them. like, it's a lot of work, and staff is working on other, equally important stuff--it'd be very easy to get consumed by one or the other. and 4 people to tag wrangle some hundred thousand tags isn't... like, i just think it's a system where you need to have a system of more tag wranglers very quickly after you get it going, especially when there already are so many tags in existence on the site.

we're well aware of just how time-consuming and difficult tag wrangling can be. we're not saying "we'll do it outselves at first" in some sort of smug easy "obviously we can do this" way.

We're not going to have a ton of coverage at first, and it'll be slow and incremental. We think that's okay. Tags don't need to be perfect tomorrow, or even next week. Out of those hundred thousand tags, how many actually get used with an extremely regular basis? Does every tag need to be wrangled as soon as humanly possible? It's something we're going to keep chopping at with a focus on tackling the highest impact stuff first. 5% of the labor can get 70% of our coverage.

It's something we're going to keep chopping at with a focus on tackling the highest impact stuff first. 5% of the labor can get 70% of our coverage.

Yeah, I'm pretty sure merging art/creator-related tags alone (ex: "ArtistsOnCohost vs 'ArtistOnCohost' vs 'Artists on cohost'" or "'my OC' vs 'my OCs' vs 'Original Character' vs 'Original Characters') will cover the bulk of posts that regularly need synonym tags, while the rest will be almost entirely redundant fandom tags (acronyms!) and niche topics/rarely used categorizing that will take like 200x the amount of time to get through.

this is true. i think merging tags for lots of media and stuff gets really out of hand for a manual job and is a bad thing to focus on now, but those big more over-arching tags for art, "the global feed", etc seems like a reasonable and achievable thing

the very short gist of it is that a few days ago cohost's dns went down for as-of-yet unexplained reasons, and the date happens to line up with when the domain was renewed according to whois reports. (the domain also has some flags set on it that suggest it temporarily expired, but this part is pure speculation on my part)

(and one problem with not posting about what happened is that it leaves people to speculate...)

over time, we plan to add an ao3-like “tag wrangler” system to help manage this load.

what does this mean, specifically? previously, staff repeatedly insisted that taking volunteer labor was Not An Option, even in an extreme financial emergency, as it would be both "unethical" (your judgement, not mine) and potentially impossible due to your funding agreement; ao3's tag wrangler system relies on lots and lots and lots of volunteers. is the plan to hire someone else to be in charge of this, to split it amongst existing staff, or has your position on the ethics of accepting volunteer labor changed?

yeah, staff please elaborate. Right now there seems like a lot of manual work to be done by overworked devs, including artist alley posts. That doesn't seem sustainable. I'd love more details on what you'll do and how it won't cause a huge workload

I think the "identical" synonyms are going to be things like #ghost and #ghosts, whereas searching for #spirits might suggest #ghosts and #whiskey as similar tags, but wouldn't put them in your search results or affect your muted tag filters.

To me, the point of synonyms/identical tags is mainly to streamline stuff like title variance (#X-files and #the x-files) spacing (#ArtistsOnCohost and #Artists on Cohost) or category tags like #art/#artwork, etc.

While the suggested tags are for discoverability and finding things that are similar to what you searched for without having to know exactly what to search for. So, getting #Tiger suggested when searching for #Lion

That's how I'd do it, anyway, and that's the impression I got from their Identical and Similar distinctions.

No, the identical and synonym thing is referring to how the tag system views differing tags for tag search. It will show tags that don’t string-match your search but are the same thing (search “video games” also get posts tagged “videogames”).

VERY excited to see tag synonyms are becoming a real thing, though i'm kinda worried that having them only managed by yourselves seems like it'd be a lot of work? there's only a small handful of you guys as it is, and it seems like you're already stretched thin between moderation, site maintenance, feature development, business stuff, etc.

and yeah i realize that tag wrangling was mentioned, and is supposed to come later, but idk it just seems odd to not have that implemented from the start to ease your guys' workload.

thanks for making cohost. i really hope for the financial updates to be more consistent in the future, as this is one of the things i was expressly expecting to happen based on prior statements. i worry about the sustainability of the site a lot and i think consistency is, like, the one thing that would actually help to reassure me. cohost is basically my favorite website on the internet but i'm just becoming more and more afraid of continually hedging my bets on it. regardless i am glad to see this here now and i am still hopeful to see cohost move in a more sustainable direction. nothing would make me happier

hi staff thank you

to the rest of my fellow chosties, begging you to touch grass and have patience. there is very good valid feedback here, but some folks are being very disrespectful and needlessly cruel to staff. they have a lot of shit going on, you are not the center of the universe. They Know what’s important, let em cook please.

it can be incredibly demoralizing for a small team to have a handful of people screaming at them thinking They Know How It Works. a good way to not have good webbed sights anymore and only evil ones is to burn out the nice people who are making your nice webbed sight at great personal struggle.

nice to see everyone is excited about tag wrangling (i am personally neutral about it), but i am just excited to be able to add an image after a text and a read-more. niiice:eggbug-classic:

You forgot? what the hell is management /doing/. I can't imagine any other website being better run by a Business Managment grad fresh out of college, but somehow you've done it! congrats on being unaccountable to anyone and losing unknowable amounts of capital, if your sloppy financial posting is to be even believed.

april 2024 mini update

first off: we are pleased to say that we have secured funding for an additional six+ months of operation. this comes from someone other than our original funder [...]

a major condition of this funding is that we are consistent in posting public financial updates going forward, which is really something that’s better for everyone, so we’re going to actually be on top of that.

Hellos, I am confused about how exactly to send feedback to staff at this point (due to multiple channels and never really being sure if the "Cohost Support" page is an active community space). So, I am posting here in the financial update to say:

This is your yearly prod with my complaint that I believe section 6 of the Cohost TOS, the IP grant provisions, are a long-term danger to users because (like most social media sites) the TOS grants ASSC an effectively unbounded grant to commercially use user-posted IP but (unlike most social media sites) it does not have a provision for a user to intentionally revoke that permission (for example by deleting the content). I believe this is fixable without changing anything about site operations or code. I don't necessarily expect this to be high on your priority list but I believe it should be on there somewhere. I wrote a post today saying this same thing but in more words.

great updates! related to the tag changes, have you thought about disallowing or automatically truncating a "#" as the first character in a tag? because i'm noticing a lot of people who come from other sites incorrectly think they need to type "#touhou" in the tag box instead of just "touhou", and this creates the very dumb looking tags like "##touhou"

of course people should just use the tag input correctly and if people WANT to have their tag be for example "#relatable" then the current system allows that while automatic truncating would remove that possibility... but also the easiest solution to people making mistakes (and accidentally tagging their post to a tag nobody is looking at; nobody follows "##touhou" instead of "#touhou") is just not allowing them to make a mistake in the first place

Thank you for making cohost!

mild negativity/concern below:

I am a big fan of the tag synonym idea in the abstract, but I am concerned about the fact that it is going to be starting off as staff-run when it increasingly feels like cohost staff is struggling to handle the current workload and the site remains unprofitable. I don’t recall where you’re at with the tipping and the pseudo-patreon system, did the payment processing issues result in both of those options being canceled? If there are still viable income-producing projects in the pipeline, I think it would be better to focus on those rather than expending effort on adding new quality of life features which, when completed, will increase the ongoing workload for staff.

How much time, effort, and focus will be expended on 1: Coding in tag synonymization 2: Staff writing in synonyms themselves 3: Transitioning to user-run synonymization?

I know that I don’t understand the dynamics of the team or the process this would actually take. I know I’m just some random person with no real authority to speak on the matter. I know that you’re doing your best and I deeply appreciate everything that’s been done so far.

But I do still have concerns, and seeing that your already-existing obligations result in sometimes just outright forgetting to release these makes me wary of adding even more obligations to the pile.