After September 1, 2023, all Gfycat content and data will be deleted
gfycat.com
gfycat.com
(The irony of posting a Reddit link about a service with rich content history becoming permanently inaccessible to many of its users is not lost on me, to be sure.)
Value production for the internet was mostly done for free. Both where it comes to end users and the hosting they needed. And yes, sometimes by "wasting investor dollars" (of course investigating new things, wasting dollars, is exactly what investment is supposed to do, that it does that is why investment is allowed).
And ironically this goes multiple steps up the value chain. Users and mods are complaining to reddit that they don't "pay" them enough. Reddit is complaining that Google/FB/even OpenAI at the moment don't "pay" them enough by trying to take in ads directly, and presumably training their own AIs soon on "their" data. Elon Musk is complaining that he's providing data for free (it's even costing him some money, even though not anywhere close to the amount he implies) and not getting the dollars AI companies are attracting with "his (users)" data.
The real value in OpenAI is that Google's position is that they are the arbiter of where users go. Users decide where to put their attention 99% ... and Google 1% (and FB 0.1%, and reddit 0.01%). But OpenAI is a serious threat to Google's 1%, because it can give better answers and therefore will have the perfect, most expensive, place to put ads if things keep going the way they're going.
For Google this is a lose-lose proposition. Either they win, get to keep their position because they get the best Chatbot. However, running this chatbot is much more expensive than Google search (and they will need to restart their revenue optimization with angry shareholders screaming at them. There is no adwordsLLM yet). So they'll need to run a much more expensive internet scrape on a regular basis, and figure out how to run ads on this new platform, just to maintain the position they currently still have. Shareholders don't care, they just want to see Google's dollar amounts go up 20% per year as they did for a decade. And if Google doesn't do this massive investment? Well, they may lose everything.
Funny how apt the old Futurama joke has become "the internet has always been about the free and open exchange of somebody else's ideas".
Futurama meant Napster. But Reddit has exactly this business model. Google has exactly this business model. Twitter ... etc. I bet the tech sector has actually made more money on this than even the most absurd claims of the RIAA say Napster has cost them.
> Shareholders don't care, they just want to see Google's dollar amounts go up 20% per year as they did for a decade. And if Google doesn't do this massive investment? Well, they may lose everything.
to “they may stagnate”. Alphabet shareholders are not going to be left with worthless shares because Alphabet still has the ability to earn tens of billions of dollars of profit per year, just shares worth less than they hoped they would be worth. I.e. they might have to be content playing in the market cap of hundreds of billions rather than trillions.
It does seem to give better answers.
But I don't think that's necessarily indicative of which kind of technology is capable of giving better search results. Google isn't what it was 20 years ago, so I don't think that what Google is now is terribly indicative of what their core technology is capable of being. They've largely allowed their search service to sink into a quagmire of content farms. They occasionally make a vague swipe at dealing with the absolute worst of it, but for the longest time they've had little incentive to put too much effort into it because these same content farms produce a sizable chunk of their ad revenue.
And they won't go down without a fight. I can't imagine it will be that easy for their crown to be wrested away by a chatbot with stratospheric operating costs that's been jerry-rigged into an ersatz search engine that nobody can afford to "reindex" more often than once every couple years. I wouldn't be surprised if a few rounds of proximal policy optimization costs more than it would cost Google to apply some proper attention to their search result quality.
For now.
source: old enough to remember Netscape vs Microsoft and many other examples.
Here is a person criticizing 37Signal who wants to optimize for profit.
The tweet was just last week ... in the current economic climate.
https://twitter.com/moritzplassnig/status/167278031492050124...
Plan A: continue as now, make a little money and everyone eats, or
Plan B: pretend to your investors you will be super duper profitable, wreck the internet, and then go out of business with your golden parachute.
Plenty of businesses make a profit - even a healthy profit - while respecting their employees and contributing to their community. Ones which optimize for profit inevitably become evil.
You can decide whether that is an evil move.
So much of what the users want from social media is not technologically innovative and thus doesn't require the profit motive to fuel that innovation.
I cannot see one reason why Facecrook, Twits, or Spreddit requires a for-profit business model. Can you?
That's an oxymoron if I ever heard one. A business is literally an activity that is undertaken for the purpose of profit. There are activities that are entered for reasons other than profit, but they're not businesses.
They could be a C corp (could also be an S corp but that comes with a 100 share holder limit), but then they have a duty to maximize profit to the shareholders. In a private corporation, the shareholders may be less inclined to sue, but the option is still there and as such the CEO must work to ensure the company is profitable. If they decide to share all profits with the employees, and the employees are majority shareholders, then this likely falls within the sphere of maximize profits.
There are things like a publicly-traded LLP which allows employees to own a portion of the company while offloading tax liabilities to the shareholder, but that is usually limited to companies which depend on depleting natural resources (think oil, gas, coal, lithium, etc)
You can have provisions about tax liabilities, but that is a specific risk that must be mitigated in an LLC that a corporation does not have to deal with, thus it’s an added risk
A "How dare you be technically correct!" variation is both an expression of mob-correctness trumping technical correctness _and_ an acknowledgement of being technically correct.
A sort of, you're right, I support you, but the crowd might not.
Read shit how you like tho
In a business where profit is used to pay shareholders/owners you don't count the regular salaries as 'profit paid out' either. This is no different; any profit we have is reinvested (while financially stable, there has been no occasion where the profits exceeded the amount prudent to reinvest in growth or other factors).
If so, then the word salary here is just another word for owner dividends, at least to some extent.
<https://www.wikiaccounting.com/retained-earnings-for-non-pro...>
<https://insidecharity.org/2018/05/18/earned-income-for-nonpr...>
Note that this may occur on a short-term basis given normal market variability, or over the long term The latter case is more interesting.
In the short term, such business profits simply accrue to accounts, and are typically held as cash or cash-equivalents.
Over the medium term, funds may be moved to various forms of generally-liquid assets (government bonds, equities, etc.), as part of the organisation's money-management strategy.
In the long term ... the organisation gets to decide how to invest those resources. There are several options:
- Increase wages, as you suggest. This is of course only one option.
- Increase benefits. Healthcare, daycare, commute / housing / educational supports, etc.
- Expand workforce. Hire additional employees.
- Expand business. This might entail expanding or improving a location, opening additional location, going into multiple lines of business, or aquiring other organisations.
- Expand endowment. This is particularly the case for many educational institutions, and those endowments may range into the billions of dollars.
Generally you'll find these options discussed in a long term capital plan or similar strategy document.
For specific forms of business organisation, such as a co-operative business (which may or may not be a not-for-profit organisation), there may also be dividend payouts to members. My understanding is that these are not equivalent to stock dividends, though I'm hazy on details.
What you don't see for non-profits, however, is a mandated and frequently automatic form of return to shareholders, through either stock dividends or inflating the value of a joint stock issue. That cuts two ways: not-for-profits don't have the financial drag of spilling out money to investors, but they also typically have less access to financial capital markets.
Sounds like a great place to work, I wonder how does it look like in practice.
Also plenty of organizations look like businesses in every way except their tax status. Thrift stores, for example. Goodwill isn't a nonprofit anymore, but even when it was it also employed all the same abusive labor practices and suchlike that you'd expect from a for-profit business.
Web hosting ain't free, especially at their scales.
First, they want a rainy-day fund to cover fixed expenses when income fluctuates. Second, and more importantly, they might want to create a massive endowment that provides them with stable income so that they're not beholden (or accountable) to donors or their usual business partners.
The only difference, at the end of the day, is whether they ever want to IPO.
The point of non-profits is that their goal is to accomplish something other than profiting. Sometimes the goal can cost a lot of money to accomplish, often much more than just the cost of paying its employees.
For example a non-profit that raises money for cancer research might have very low costs and raise tons of money without raising eyebrows. Because they put the majority toward their goal, and everybody understands the goal, and "profit" is not that goal.
Sad they didn't manage to be sustainable.
For instance, smash melee used gfycat heavily. There is in depth analysis of the game using gfycat as visuals. There will be knowledge loss in that community in Sept 1.
Honestly, I expected most of the information to be locally hosted on smashboards or their wiki. (And if not now, they will be soon).
I'm worried about other communities though.
"technologically adept" and "uses gfycat instead of wiki or forum" is mutually exclusive
If gfycat provided them with an ideal - for what they and their users wanted - and free (I assume) way to host their clips that the community liked then apart from the eggs in one basket issue can you really say they made a bad decision? If (and sure it's a big 'if') they made sure everything was backed up and ready to easily move should something like this happen, then it could well have been an idea decision that worked well for years?
At a minimum, the smash community members who built that data have an expert-level understanding of Gamecube graphics programming, Gamecube assembly language, and Super Smash Bros's internal scripting language to create those posts.
---------
Yes. This is a very technically adept community. I'm not very much a part of them, but I can see the expert-level reverse engineering work needed to get this working.
And the knowledge won't be forever lost if it is lost. The importance of hitbox / hurtbox reverse-engineering is common in all fighting games (Blazblue, Guilty Gear, Street Fighter, Marvel vs Capcom, Super Smash Bros). The various communities always work on reverse-engineering all the frame-data and hitbox/hurtbox information ASAP whenever new fighting games come out, so that expert-level fighting game players know what their training routines should be.
Actually this was possible before Dolphin existed. They did it on raw hardware. The devs left an extensive debug mode in with various displays like hitboxes. So a little Action Replay cheat was enough.
I can assure you melee uses Youtube much more heavily. Anything not currently hosted on YouTube can be reupload fairly easily.
I agree rehosting isn't hard. But there will be information loss by the deadline without serious effort.
EDIT: This is reminding me to back up some niche twitter melee analysis given that whole situation.
EverythingGPT, generate a 140-minute movie containing this clip, Oscar for best supporting actor, clever twist in the end, stilted dialogue in parts, 4k.
If only a single copy was obliterated I would be just as dismissive because any of the destroyed material can still be accessed and archived by using another copy.
There is something funny about Snap buying a company and then setting a timer to delete all the videos. Shame about the data loss.
Service Notice
The Gfycat service is being discontinued. Please save or delete your Gfycat content by visiting https://www.gfycat.com and logging in to your account. After September 1, 2023, all Gfycat content and data will be deleted from gfycat.com
Does anyone have a view into why? Did they run out of outside funding? Another interest rate kill?
Getting bought up has not generally been a positive thing for anyone but the founders cashing out in a long time.
All independent image hosts suffer the same fate. There's simply no money in it. The moment you try to monetize people will move to a new free service, and the cycle starts over.
Maybe a service like this should start with ads and paid plans (that unlock restrictions), keep their running costs low, and never stop being an image hosting site?
Too many start without ads, get users and some investment and then spend it all on expensive AWS hosting/CDN bills and employees that are hired to create a social media platform... and then they stop being an image hosting website, fill the site with ads, focus too much on growth, and so on. I don't blame them for taking a big pay check, but I don't think the entrepreneur type works well for image hosting.
The problem is that it's just not that profitable. What is an advertiser going to do, with the knowledge that internet people like to look at cats?
I was stating we have enough of it as is. No need to be snarky when you couldn’t be bothered to read. Of course it exists, what kind of troglodyte do you take me as, sir?!
Why did they "have" to raise money and expand? Maybe this was sarcasm and I missed it in which case ignore.
They were profitable but then they haaaaad to go and raise money.
:-)
Even if they were to insist on making their money by pushing ads, doing it in a way that doesn't degrade the service would go a long way toward keeping users around.
When I think about the massive amount of data image hosts collect, I have to imagine there should be ways to make money from that information without any impact on the user's experience uploading or viewing photos.
As far as I can tell the biggest problem image hosts have is dealing with our insane copyright laws. Users just want to share information, while companies want to act as gatekeepers and toll collectors which often leaves image hosts caught in the middle. They hold massive data sets of our collective culture and interests, but have a lot of limits on what can be done with it. Most of their content was uploaded to them in violation of copyright laws to start with.
Then there are all the short links on the web, what happens when t.co breaks when someone refuses to pay the bill?
https://arstechnica.com/gadgets/2023/04/hosting-site-imgur-w...
Also why did snapchat shareholders allow the compaby to spend money on a service and then just kill it. Was gfycat a snapchat competitor? How dothr anti competition offices allow it. Why do the shareholders allow it.
At Snap's scale, I doubt there's tech they're interested in.
I went to check the top charts in the iOS App Store to see where it sat, and I didn’t even get that far because it was at the top of the Apps tab as the 4th thing listed by Apple under “iPhone essentials”
(I haven't gone further than the login to try and actually purchase one).