This would be a good incentive for paying $8 for Twitter Blue: you are allowed to use third party apps. It's the first mildly convincing reason to pay for Twitter that I've seen yet.
Even better: access to use third party apps is a second tier that costs $10, and give $1 to the app maker. All the incentives are nicely lined up: the user won't choose to pay more if it's not worth it so the alternative app is guaranteed to offer some added value, the app maker gets paid for their efforts and is inclined to closely follow Twitter's increased rate of feature development so at least they don't fall behind and ideally innovate, Twitter gets value in not just a bit of extra money but also platform interaction experimentation which they can mine for good ideas to integrate back into the official client, they're inclined to make the API good, etc. I don't know how most people would take to this idea, but I like solutions where the incentives are all aligned. Am I missing something?
Isn't this one of the oldest discussions on Hacker News? Generally, I thought the reason you can't pay to use an arbitrary third party client on a locked-down service is that the most lucrative clients will disproportionately take advantage of that offer, which makes it hard to evolve the platform, and harder in the immediacy to monetize.
I've only learned in the last few days how many ads Twitter has. I've never seen any of them, because I've never used the Twitter website, just Tweetbot.
Also: after abruptly pulling the plug without explanation, I'm also pretty unlikely to shell out a dollar for it in the future. Certainly: Tapbots has no reason to invest in the platform anymore.
I think I have about 200 tweets on my Twitter timeline per one Mastodon toot. And the usability of Mastodon is so confusing that it's hard to see people moving. And brands have even less of a chance of moving, so I'm probably not going to get BBC Breaking News or the official Steam Deck account there anytime soon.
I think that 200 tweets per toot metric may be accurate tho in my case it's closer to 20 (tools like Movetodon can help get you started).
The difference for me is that those 200 tweets are mostly about the same drama and the replies to every one of them are always full of trolls and genuinely shitty opinions havers (e.g. sports fan bunch of numbers insisting that the police should just shoot protestors) and somehow this stays the same even if I spend ten minutes every now and then just blocking those reply guys.
You can shout "echo chamber" all you want but the lower overall population and stricter crackdowns on trolls and defederation of dodgy instances certainly leads to less noise and more respectful interactions. I'll take one 1000 character reply I disagree with over 200 low-effort trolls and parroted culture war slogans.
Turns out the "public town square" is a much more pleasant space when it's not full of a billion people and their sock puppets shouting slurs at each other all day.
> [...]And the usability of Mastodon is so confusing that it's hard to see people moving
... multiple former-Twitter-app-developers are working on this problem right now. If the apps are user-friendly, it won't matter how esoteric the protocol or web interfaces are.
This might be one of those occasions that we'll look back and wonder why Elon chose to intentionally stepped on this particular rake. Filed under "unintentional first-order consequences". I have a feeling these individually small rebellions will snowball into something fierce.
>If the apps are user-friendly, it won't matter how esoteric the protocol or web interfaces are.
Considering how bad Mastodon itself is to use, I don't have a ton of belief. And I personally have a hard time figuring out how you can make Mastodon user-friendly. Like, what server is Ivory going to default new users to?
shrug Not particularly, no. It's tech-forward, but it's not _fast_. I'd be surprised if they're not at least looking at it internally, but things take time.
You can have millions across the fediverse. You don't need to see all those millions on any given instance. Fediblocks are good actually and while some complain about "discoverability", the filter bubble[0] actually disincentivizes the kind of attention-seeking disruption that "engagement driven" platforms like Twitter have ended up optimizing for.
[0]: I'm using the term somewhat sarcastically as the filter is just about whom you follow, not about which views you agree with. If you follow a diverse enough crowd you'll still be exposed to views you may disagree with simply because you follow persons, not topics.
It really wouldn't -- the Twitter Blue thing is BS to begin with: the checkmark is meaningless, and you shouldn't need to pay to have your tweets appear more prominently / higher in timelines.
Twitter already nerfed the API in 2018, and devs of third-party apps have to pay for the current API, so it's not like Twitter isn't making money from that... they're likely making more for charging much-more-popular third-party apps for API access, than they're getting for ads on their own (mediocre) app -- never mind the fact that Twitter for iOS and macOS were just rebranded versions of Tweetie post-acquisition (circa 2010)...
The new "pay to be verified" scheme is absolute nonsense, and makes the new checkmarks absolutely meaningless... and paying to have your tweets be more prominent makes two classes of user and tweet, effectively fracturing your user base.
In November, there were an estimated ~100,000 Twitter Blue subscribers[1] -- meanwhile, more than a million Twitter users[2] were driven to Mastodon, and a number have shuttered or mothballed their Twitter accounts -- especially in the tech scene.
That's not good for business, and it certainly doesn't "incentivise" staying on Twitter, much less paying to do so.
> and you shouldn't need to pay to have your tweets appear more prominently / higher in timelines.
And more to the point, you shouldn’t be allowed to, because as a mechanic for social networks it just doesn’t work. No-one wants their timeline to be full of stuff from people who are paying for attention; these people are generally not good enough to get attention without paying for it.
Imagine TweetBot and Twitterrific just silently became dual Twitter/Mastodon clients, automatically following Mastodon accounts that are followed on Twitter, and when Twitter API is unavailable, customers don't notice a thing.
If ads were really the issue, all he had to do is add the ads in the API.
Instead, he caused free advertising for Tapbot's upcoming Mastodon client and pissed off power users who produce the free content the site relies on to sell ads.
Thirdparty apps wouldn't exist if the official app was good enough.
Third party apps decide how content is displayed, regardless of what comes out of the Twitter API. That's the problem for Twitter, from a business point of view.
Is that much of a problem? Twitter's ad units aren't arbitrary. They sell things that look very much like tweets. It seems like it would be pretty easy for Twitter to insist that things get rendered the way they want or else. They could even provide library code for popular platforms to do that, as they have that already written.
I've seen ads on Twitter that render in a more sophisticated way than tweets. For example, they include buttons.
Third party apps could implement this kind of thing, for sure. However, on an ongoing basis, Twitter would need to ensure that all third party apps had the latest ad presentation logic, and that they worked correctly, and none of the third party apps used sneaky tricks to hide the ads. After all, there's no benefit to the developers of the third-party app in hosting ads that benefit Twitter.
It would quite costly and fiddly for Twitter to administrate this, and IMO, the benefits of a third-party app ecosystem don't really stack up from Twitter's PoV.
I suspect that the first thing would be a simple legal letter saying "Cease and desist". The second would be a lawsuit. And via side channels Twitter could probably prod the US government to go after the apps for breaking some hacking law, and would almost certainly just ping Apple to take them down too.
It’s kind of insane that the third-party client situation lasted the way it did for so long. Twitter subsidized someone else’s paid app, that didn’t show any of Twitter’s ads, for free, for years.
I exclusively used Tweetbot, and checking out the first-party clients this weekend confirms to me that they’re still pretty bad. But I knew that eventually this day was coming. Twitter never made properly supporting third party clients a priority, by either finding a way to charge for access or finding a way to ensure ads were in the feeds. It couldn’t keep going the way it was, and obviously Elon has no interest in devoting any time/attention to it.
It was a mutualistic, not parasitic, relationship. Twitter's product development stagnated the day they nerfed the third party dev api.
3rd party Twitter clients were once considered a design playground where the best elements not only influenced Twitter's future design language, but also that of iOS as a whole. Tweetie's pull-to-refresh is emblematic of that. I'm pretty sure Tweetbot invented the (partial) left/right swipe to invoke different actions on tweets that have since become a mainstay of email apps.
Twitter didn’t have a first party client for years, the clients banned Friday are those that came first.
Core features and the name of those features Twitter has today came from third party apps.
Twitter forces third party clients to pay them money to use the new API which is needed for them to function as full clients but the new API is worse than the old one which was free to use.
Twitter chooses not to show ads for third party clients.
How? Can you even sign up in third-party apps? Those third-party app users were already Twitter users before. Maybe you meant they wouldn't remain Twitter users because they can't tolerate the default app experience?
Before twitter apps (note that twitter didn’t have a first party one for a long time; the early stuff was all third party), I barely used twitter; after I used it all the time.
I’d already kicked the habit (or rather replaced one addiction with another and moved to mastodon) after the purge of the journalists, but if I hadn’t this would be the end of twitter for me; the first party stuff is borderline unusable.
I suspect Saint Car has misunderstood how dependent power users are on the third party apps.
Is it impossibly surprising that someone might only bother to use the normal app for the purpose of signing up for the service so they can use another app they heard of? Think of recent versions of Internet Explorer: "what is my purpose?" "you install Chrome." (But, yeah: I'd say that it isn't even quite about maintaining the user but the extent to which they continue to engage; like, the user might still be a user, but are they someone who spends the entire day using the app? A number of these alternative clients were what were being used by teams of people to be able to manage the really complex/important accounts.)
The maths from Twitter's probably is that those users mostly are power users producing a lot of content other users want to see. If they leave the platform Twitter could risk losing more.
I am not sure how pointing out that Twitter has not, over the course of a decade, managed to make a client as good as third-party ones made with 1/1000th the headcount, nor figured out a way to monetize them, is “in support” of Twitter. Should I just edit in a few “Space Karens” to make you feel better?
M. Haddad has also discussed a current problem. Users have current subscriptions to TweetBot and face repeat subscription charges at their app store. Migrating those users with time left on their subscriptions (and who want it) to Ivory, the in-development app from Tapbots for the FediVerse, requires an update to TweetBot. However, app stores will not accept an updated app when that app is rendered non-functional by Twitter.
Twitter has left 3rd-party vendors with no sunset period, or ability to deal with their customers gracefully, directly impacting not only their apps but their revenue. It is as yet unknown how amenable to 3rd-party Twitter app vendors the app store owners for iOS and Android will be.
As a user I wouldn’t have liked one big app for everything. It’s an interesting idea to combine the two worlds, but I’m happy to use dedicated apps that can focus on each network’s unique features. Should also be easier to build and maintain that way.
And it doesn’t seem like there’s a lot of hope for the future of 3rd-party Twitter apps anyway.
When you make a business that cannot stand on its own feet and depends on another company as the basis of your company, then these are the kinds of catastrophes you have to admit to yourself is possible. When the company you depend on changes ownership, the "break glass in case of emergency" options should be very well understood and ready to go at a moments notice.
Yes, it sucks something you created is struggling due to things out of your control. However, you built your something with a foundation out of your control. I'm just struggling to be sympathetic with the business model.
Can you name a single business that doesn’t depend on dozens of other companies to stay in operation? Like, if AWS decided to ban someone, would you say “I’m just struggling to be sympathetic with the business model”?
The difference is that my business that depends on AWS pays for the service, and big businesses that depend on AWS have contracts about support and payment.
There is an implied social contract that comes along with even free services that a company won't make a change like this effective immediately and won't make a change like this with zero communication. Part of Musk's approach to leading Twitter is simply ignoring these type of social and in some cases even legal obligations to others.
If this incident truly represents Twitter's change in approach to third-party clients, that change in policy is being executed in the most petty and malicious way it could be done.
> executed in the most petty and malicious way it could be done.
This is not to excuse Twitter's actual behavior, which is clearly shitty. But: More petty and malicious would be to shut off access, then contact devs and promise that access would be restored if they just make these tiny changes to their app, and also sign this NDA forbidding discussion of the process. Then string along for a week or a month demanding just one more tiny change after change.
And you think that AWS won't just rip the rug out from under you someday when you're no longer profitable for them or "shifting to a different focus" happens?
Every single one of those contracts has fine print that says that Amazon/Google/Microsoft can revoke service and end their relationship with you whenever they want.
The value for Twitter depends on who uses/used the client, what value they were providing to Twitter, and whether they'll stick around to continue providing that value.
Can you be more honest with the discussion and realize that you're being deliberately obtuse to the conversation? Building a company that absolutely has no alternatives to provide THE thing that makes your business, then the phrase "all your eggs in one basket" applies here. That has not been good business strategy 101 for as long as I've paid attention (and probably a long long time before that too).
Okay? The fact is that plenty of businesses are in that position for one reason or another, and I think it’s pretty callous to withhold sympathy on that basis.
Okay, what was the back up plan? Nothing? none? really? nope, no sympathy.
Oh, you had a plan, but you couldn't really get it up in time, well, that's some learning pain. I'll cry a tear in some beer with you.
Tried to make a plan, realized there was no alternatives, but carried on anyways? That's very brave (ahem, read not smart) of you, but if you're a sole proprietor with no employees depending on you, then we can also share a beer of how to be smarter in the next venture. Got employees that are now hat in hand because of your "bravery"? Nope, again, no sympathy for you, but I'll be at the bar with your laid off employees burning you in effigy.
I'll quote an old literature teacher of mine, "You're wrong" when asked what I thought of the assigned reading. What I thought can't be wrong, it's what I thought. If you asked me what the author actually meant, then I could be wrong if I gave you the same response. As are you in this case, if you think that was what I meant.
Did you accuse me of saying "look at what they were wearing", because that's not at all what I was implying and you've read into that of your own volition. I simply said if you're going to make your business so entwined with another company's business in order for your business to thrive, then you better have back up plans for that other business no longer being available. There's no victim blaming going on here, except in your mind. As someone who's been part of a business that suffered "all eggs, one basket" syndrome, it is something I'm keen to not suffer again. If you can't handle business advice that questions your business plan, then maybe you shouldn't be running a business.
You say there’s no victim blaming here, but your other comments to me make it abundantly clear that you are in fact victim blaming. We’re not going to agree. Have a nice night!
> However, you built your something with a foundation out of your control.
The trouble isn't this necessarily; a lot of business models don't exist unless another company does. The App Store is a good example of this; if Apple just disappears then so many other businesses disappear with it.
The trouble is the fact that the relationship between these two parties is so one-sided. Unless these third party Twitter clients have some sort of contractual foundation to stand upon, then yeah, what they've enjoyed so far is basically akin to an ad-hoc, at-will support, that can be terminated at any time with no notice, for no reason.
A large percentage of all companies work that way. Tool sharpening shops next to a big steel factory. Windows cleanup/uninstall
tools. Car dealerships. Restaurant franchises.
Yes and the point here is: Burger King stopping to ship Whoppers to franchise restaurants without warning would be as unnecessarily cruel as what Twitter has done.
I've said this for years and the same mistakes are repeated all over again about the risks of building your entire startup or business solely on someone else's platform or API where you don't own the data or the source.
Whether if it is Instagram, TikTok, Snapchat, Twitter, etc is no different to them locking down API access from third-party apps and can do so at any time, official or not.
No different to this event with the closure of a third-party Discord client [0]. I guess I was right in [1] as it eventually happened to Twitter.
Your maximalist capitalist point of view has no place in civil society. To quote our greatest modern philosophers: be most excellent to one another dude!
Did the developer have any special agreement or contract with Twitter that the API would always be available? If not it sounds like an unacceptable professional risk to base your livelihood on the whims of another entity.
Musk famously does not care about contracts. Note that the only reason he ended up buying Twitter is that Deleware's corporate courts really don't fuck around and were likely going to force him to buy it pronto despite his attempts to wriggle out. But he'll happily screw over lesser people, as he's doing with various vendors, like Twitter's landlords: https://www.businessinsider.com/twitter-landlord-sues-compan...
Explain to me how a small software vendor, even with the best contract in the world, is going to force Musk to turn back on their access in a timely enough fashion that they'll still have a business left? And be able to do it for a modest enough amount of money that the ISV will still be turning a profit?
That wasn't the point of the previous post. Clearly if they did have a contract there would be recourses to take, and yes that won't stop Elon being Elon.
The point was that when you build on top of someone else's platform, [1], then you are beholden to their whims. This is a substantial business risk, which has been happening since forever.
I sympethize with Tweetbot et al, but this day has been inevitable for some time. They are not the first to suffer this fate in tech, and they won't be the last.
[1] I'll add, without paying - ie not being an actual "customer".
The thing is, I can't think of a company, with even a 1/10 the userbase that did not at least announce turning of such a service.
These changes impact at the very least the users of Twitter itself.
And given the history of Twitter, one can argue that the success was in part due to these third party clients.
It opened with "Did the developer have any special agreement or contract", so yes, I think that was the point of the post.
I agree that building on other people's platforms is dangerous, but that's true regardless of the contract you have, because contracts only matter to the extent that you can afford to enforce them.
Not sure about special agreements or contracts, but Tweetbot along with other high-profile third-party apps (like Twitterrific) had way, way more API tokens available because of their size when Twitter started restricting the number of API tokens an app could have assigned for the lifetime of the app. API tokens for new apps were way more limited, around 10,000 I believe. Tweetbot and Twitterrific, while they never disclosed how many tokens they were given as legacy apps, were still to this day accepting new users, so the number seems to be way more than 10,000.
And they had access to a less limited API (request-wise) to make their app work better without constant rate limiting nags.
Ironically, the best information about the status of TweetBot is on the FediVerse. Paul Haddad one of its developers has been posting at https://tapbots.social/@paul (directly followable as @paul@tapbots.social by FediVerse users).
In an experiment on Sunday evening, M. Haddad switched out the Twitter API key used by TweetBot, which had already been disabled since Friday 2023-01-13, for an older one that it still possessed that had posting limits. People rushed to confirm that TweetBot was back, only to discover that they could only log in and read, not post. It was approximately 3 hours from then until that API key completely stopped working too, along with all of the other API keys that Tapbots had.
The Twitter API itself is still present, but specific API keys used in several 3rd party apps have ceased functioning.
No developer has received any word from Twitter at all about what is going on.
However, unconfirmed rumour, reported on some WWW sites, had had it that internal communications within Twitter revealed that the disabling of the API keys for specific 3rd-party Twitter apps was intentional, and that Twitter staff had yet to work out how to announce this publicly. M. Haddad took the disabling of the other API keys once they were put into use as confirmation that this rumour was indeed accurate, that this was not an accident or just some bug that would be fixed, but a deliberately targetted move.
This combines with other developments. 3rd party apps could show chronological Twitter timelines, and skip advertisements. The chronological timeline has this same weekend been removed from the Twitter WWW interface and 1st-party app, and the Wall Street Journal reported that the same day that the 3rd party apps stopped working Twitter offered advertisers halved advertising rates if they paid for a blitz of up to $250,000 worth of advertising to Twitter users that must run before 2023-02-28.
Effectively, Twitter is now marketing itself as distinct from the FediVerse, with no 3rd party apps (the FediVerse having several each for iOS and Android, including one in development by Tapbots), an advertising blitz with user attention being sold to advertisers and those who pay Twitter as the funding model (with many FediVerse instances opting for a user-funded or coöperatively-funded model, with organizations that wish an "official" presence funding their own instances for hosting official staff accounts, as Tapbots itself does, and advertisements being banned), and only algorithmic timelines (chronological timelines being the only option in most FediVerse WWW interfaces and apps).
Several users who have disabled their accounts in response to the loss of TweetBot, deleting and archiving all posts, marking the accounts private, and unfollowing everyone, have also reported (in various FediVerse posts) upon checking back that their accounts have had followers re-added to them by Twitter itself.
Unfollowing everyone ensures both that the accounts do not accumulate notifications and concomitant junk e-mail, and that follower counts accurately reflect the real reduced Twitter effectiveness to those others that remain using it. Many (not just Charles Stross) have also stressed the importance of not deleting accounts, and occasional dummy keep-alive posts, in order to prevent impersonation of long-standing public identities. Anecdotal reports over the weekend have already stated much reduced views of Tweets made by those without paid-for accounts, because of the switch to non-chronological viewability only, favouring advertisers and the posts of those who have paid Twitter for accounts. Not paying Twitter is, by those reports, to be no longer read, as much if even at all.
> The chronological timeline has this same weekend been removed from the Twitter WWW interface and 1st-party app
Well, I left a bit ago, but if I hadn't that would be the end for me. That's the only view I used on Twitter, and found the non-chronological timeline to have almost no value at all.
Definitely haven't regretted my decision to leave, but that decision is such a strong reinforcement of that decision.
Thanks for the heads up that chronological timeline is getting disabled. I've disabled auto-updates for the Android app in the hopes that I can keep it for just a little bit longer. We'll see whether that works in the face of Musk's determination to make Twitter profitable.
I don't think chronological timeline is going away though, it's still there on the Mobile Web version at least. There's two tabs: For You and Following. For You is the recommendation timeline and Following is chronological.
But unfortunately (unlike before) you can't default to "Following." You're defaulted to the "For You" tab on every hard-load, or app open (following a close) and you have to click or swipe over.
At least before you could click that "Home" icon and switch to "Latest" and it would reasonably stick.
I gave Twitter a lot more leeway than I should've through all this due to network effect and intertia, and I regret that a lot. When this "For You"/"Following" thing happened I actively went back to Echofon and Tweetbot, and was planning to make a subscription just to avoid what I was seeing on the first-party app/website. And now I can't even do that.
Wouldn't be surprised if Tweetdeck is forced to move over to the "new" Tweetdeck (which is just a wrapper for the web experience) soon.
I am confused. There was indeed a reconfiguration of twitter in the last couple of days.
On WWW, I see the tabs
* For You (which seems to be the recommendation-based feed)
* Following (which appears to be the chronological feed)
* All my favorited lists as separate tabs, which previously had to navigated to on WWW. But which were available as tabs on Android before and still are.
There was also this tweet from EM in response to PG that talked about this update [1]
Mastodon is the name of two pieces of software, a WWW server and (a couple of) iOS/Android apps. There are other WWW server softwares in use on the FediVerse, such as Frendica. There are also other apps, such as (for some examples) Tusky, Fedilab, and Megalodon on Android, and MetaText, Mastoot, Tooot, and Tootle on iOS. The actual network is the FediVerse.
Indeed, the Mastodon people are concerned over the trademark dilution aspects of sysops using "mastodon" in their domain names when it isn't the Mastodon WWW server software in use.
Those familiar with FidoNet will find the FediVerse very familiar in structure, and the distinction between Twitter and the FediVerse is much akin to the distinction between BIX/Compuserve and FidoNet back in the 1980s. Much is the same, including local funding of nodes by sysops and users, local policies set by local sysops, and the fact that not everything gets distributed everywhere.
(There aren't quite FidoNet-style local echoes, but there is a marked difference, even across instances in a single country such as, say, the U.K., as to what can be found on each instance. Quite a lot of discussion in the #TweetBot hashtag can be seen at https://tapbots.social/tags/TweetBot , and markedly fewer -- on Saturday almost none -- of the same posts can be seen at https://know.me.uk/tags/TweetBot .)
To those familiar with Twitter, one way of thinking about it is that it is not one big central Twitter that carries everything on the planet all mushed together, but lots of little separately owned Twitters that can cross-post, and that don't carry everything.
Jack Dorsey did try funding a competing federation protocol called Bluesky[0], but I imagine anyone at Twitter who was in favor of that has either left or been fired.
Mine is missing tweets. It is more of "what twitter algorithm thinks is interesting from the people I follow in roughly chronological order". The latest used to show everything people you follow tweeted.
I’m surprised that acquisition isn’t part any strategy to constrain 3rd party apps. Acquiring the top few apps would soften the PR blow and retain users. It’d bring back on board some crack dev/product people who’ve already managed to create better apps than Twitter Corp was able to forge from it’s earlier acquired official app. The cost would have been nominal in the grand scheme (the devs would have had little choice but to acquiesce). It may yet happen.
To me, that it hasn’t, we’ll this indicates the block is temporary, showing who’s boss before introducing a new rule like paid users only, or ad exposure.
Either way, it’s an Emperor level decision. Would love to have seen the internal heat when the decision was mandated.
I've had active accounts since 2009. Never had a super active presence, but it's been an incredibly useful tool for communicating support and user outreach.
I just announced that I will be deactivating my accounts.
Direct contact will be through charles@getjsonip.com.
I just can't abide this blatant behavior of firing thousands of people and arbitrarily disabling access keys for various client apps with absolutely no prior notification.
Not sure if I should make a post on HN about the change for jsonip.com or not. I have a lot of users but don't know if there enough here to justify a post.
Most of them were ad-less free apps that offered authenticated access to free APIs, so pure costs to Twitter, but Twitter hadn't been able to shut them down due to feature deficiencies in official apps.
I can confirm at least some of the free apis are still in working order. Twitter started to throw an error yesterday afternoon that my app. wasn’t in a project but I was able to resolve it.
It almost seems like Elon Musk has forgotten about path dependency, which maybe makes sense since he other companies basically don't require you to think about path dependency, you build a new factory or design a new car or rocket. I think most people agree there's a problem for Twitter as a business that these third party apps exist. These apps sap Twitter's ad revenue, they become an intermediary between Twitter and it's highest value users. What happens, for instance, if one of these apps suddenly start mixing Facebook/Instagram/Mastodon into users feeds? Ouch! Also, it impacts how your new feature roll outs happen. So, you need to own the app that all your customers use.
But you need an actual plan to get people to move from the third party app to the first party. You know, figure out why the third party apps are used, build some equivalent features and lure users back. Instead Elon Musk's plan is just "let's break that". Well ok, but you haven't actually managed to get people to come back to your first party app, you've just broken their experience of twitter. Well done, you've broken twitter for your most valuable users. Do you think, having broken the service with no notice at all, and with no viable app that they want to use, that they're going to use your first party app? No! They're just going to fucking leave! What do you think all those developers who were writing great twitter clients are going to do now? You could have hired them, you could have had them writing great twitter clients for you. No instead, you've forced them all to go off and write clients for your competitors social networks. This move has single handedly created the largest developer community for Mastodon.
So yes, we're finally at the point where Twitter is best used through the first party app. We've achieved that by losing the majority of the high value users that you were worried about on the 3rd party app in the first place.
I wonder if Twitter considered charging users of third party apps to be able to use them? i.e. in order to access Twitter via a third party app, you have to pay a monthly subscription to twitter.
I know Twitter has discussed subscriptions, and this seems like a way to leverage the value of third party apps to directly drive revenue. Maybe even give the third party app developers a cut. Then you have a situation where your developer ecosystem has incentives to drive usage and revenue for you.
> I wonder if Twitter considered charging users of third party apps to be able to use them? i.e. in order to access Twitter via a third party app, you have to pay a monthly subscription to twitter.
Or even better, third-party developers pay for an API key set on a subscription plan to use the Twitter API in their apps, which makes sense and is a win-win-win for Twitter, third-party developers and users.
It's a rather funky phrasing, and I don't think it's an issue with your English. "per" is usually used to mean "in accordance with". So, "per your request" means "in accordance with with your request". It looks the submitter was trying to use "per developer" to mean "according to the developer". While "according to" and "in accordance with" are synonymous when meaning that two things are in agreement, I wouldn't use "in accordance with" or "per" to mean "as told by".
per my understanding, ie according to my understanding, per developer here means according to the developer. as in, the developer reports that their keys have been disabled by twitter. given that it’s the developer reporting/complaining, we can infer that they didn’t issue any commands or requests.
I wonder if commenters who thought that Musk would usher in an era of freedom and an end to bad design choices at Twitter are eating crow about now. It seems like the previous executive team was walking a tightrope of satisfying several competing interests, whereas the current leadership has gone all-in on monetizing to the maximum extent possible.
Not one thing that Musk had done prior to his purchase of Twitter ever suggested to me that he would do anything but exactly what he's doing @Twitter. Everything about him suggested "freedom for me (and my ilk), but not for thee" would be how things would proceed. Which flavor of kool-aid do I need to drink before I see Musk as the great savior instead the person I see now?
> Which flavor of kool-aid do I need to drink before I see Musk as the great savior instead the person I see now?
Presumably buying a Tesla and believing that its the best car in the world, also that Elon musk did everything alone and he's the reason Tesla exists(not true, elon didn't create Tesla).
One oft-overlooked factor is that when you look at the people Musk surrounds himself with deliberately (so not the cult, which is mostly just generic personality cult nonsense, TSLA owners desperately trying to deny that their stock is collapsing and people who deluded themselves into thinking their Tesla car is better than it actually is) are all people who fundamentally do not care if Twitter goes down.
They're the right-wing commentators and conspiracy theorists who see Twitter as everything wrong with modern society[0]. To them, no matter what Musk does, it'll be beneficial for them. Either Musk reworks the platform to be amenable to them or he drives the company in the ground. Right now he's doing both, so for them it's an absolute win.
That's the flavor of kool-aid Musk is deliberately drinking from and once you realize that, most of his actions make some amount of sense (whilst also having a distressingly low concern about his personal fortune, something which makes it clear that when it comes to right-wing radicalization, he's firmly within the "sucker" group, rather than the "grifter" group; the latter is flexible with their beliefs if it risks losing personal profits, Musk isn't), like the utter nothingburger that is the "Twitter files"[1].
[0]: Point of clarity - this is the people who Musk surrounds himself with, I am not saying anyone who thinks twitter is negative on society is a right wing commentator or conspiracy theorist. Musk specifically is talking and listening to policy advice from these two groups and this is what those groups generally believe.
[1]: Which so far have been extremely boring and generic content moderation requests rather than proof of any "inside jobs" and restatements of old news.
>> Which flavor of kool-aid do I need to drink before I see Musk as the great savior instead the person I see now?
The things he said about freedom of speech ("free speech absolutism", etc.) made a great impact on people. To be fair, there was also some merit in that narrative as censorship was indeed happening in Twitter 1.0.
It's quite easy for a commoner or pleb to think that Elon Musk is single handedly taking on the "mighty establishment" on behalf of the common people. The "Modern day Robin Hood" is a kool-aide stereotype which isn't easy to shake off and Elon seems to have been successful so far in that narrative pitch.
The fact that the person who tried to silence multiple employees and journalists convinced anyone that he is a "free speech absolutist" really baffles me. But then again the same people shouting for this brand of "free speech absolutism" had absolutely no problems with protestors (often violently) being escorted out of Trump rallies.
>I wonder if commenters who thought that Musk would usher in an era of freedom and an end to bad design choices at Twitter are eating crow about now.
Absolutely not. Musk fans tend to decide good/bad based off of “Musk idea”/“not Musk idea”
Somewhere there’s somebody crafting some version of an “Uptime and usability are ideas from a beta cuck mentality, in this thread I will show that slow-loading brain pill ads on a website facing a mass exodus are the sign of virility and strength…” post right now.
Exactly. They still think after he pumped and dumped Doge coin that he’s going to pump it again one day, and they’ll be rewarded for faithfully HODL’ing. The older I get the more I think established organized religion may not be such a bad thing compared to the alternative
Common mistake. He doesn't mean "free speech" when he says "free speech" and neither do those defending him. There are like three people who genuinely believe in free speech above all else, everyone else is fine with a subset of speech being censored.
Note that e.g. reinstating far-right accounts banned for hate speech is "free speech" but banning leftist or progressive accounts doesn't contradict that. Only one side of the political divide loves to portray itself as "centrist" to hide its actual beliefs and values.
I'm not sure if Twitter burning down is bad or if it's a blessing, Musk's Twitter so far seems to be generating less hate-driven politicisms than it used to. That is a plus, short term.
It's a cult of personality rather than anything else, so presumably not. If Musk said he was going to save Twitter with homemade submarines, the fanbase would eat it up.
Just like Musk himself, they did not wanted general freedom. They wanted to stick it to the libs, to suppress their enemies and to restore hardcore right wing accounts. Musk delivered on all counts.
Given twitter’s massive headcount increases in the previous couple of years and the decline in ad spend that came last year, the previous executive team was about to fall off their tightrope. I’m sure they would have handled it with much better communication while treating employees better, though.
Do you want to guess how much it lost in 2022? Hint: Elon Musk's buyout forced Twitter to take on $13 Billion debt... estimated to be $1.3 Billion in interest payments per year.
This "Headcount" fallacy seems to be just the pro-Elon Musk crowd ignoring the fact that the #1 and #2 sources of Twitter's financial woes are:
1. The aforementioned Loan.
2. The $2.5 Billion (estimated) loss in advertiser revenue.
Twitters losses in 2023 will be astronomical. It will not be possible to ignore Twitter's loan issues, or their loss in advertiser revenue.
Are they still required to report those losses to the public under the new structure? Or can they wear a facade of success until a sudden "Our Incredible Journey" at some point?
Losing money in 2021 for a tech company in Twitter’s position is dire. That was the most intense boom year the industry has experienced since at least the 90s, maybe ever. Meta had all time high gross revenue and net income, both about 35% higher than in 2020. They haven’t lost money and are still reducing headcount by 12%. Twitter’s layoffs were bound to be much greater.
I am not a fan of Elon Musk, for what it’s worth, and I’m sorry that you felt the need to make this some kind of tribal battle instead of a productive discussion. Twitter’s financial issues were well known before Musk became involved, and it is widely agreed that he greatly overpaid for the company. Nothing he has done with the company should impress anyone at this point. The point stands that the company was going to be experiencing major pain even without him.
> The point stands that the company was going to be experiencing major pain even without him.
$3.8 Billion / year of that pain is directly attributable to Musk, in a company that only had $5 Billion/year revenue.
Its really that simple. Look at the advertising losses due to Elon Musk's right-wing redpilling of the world... and add the $1.3 Billion loan structure Elon Musk is directly responsible for. There's no need to overthink this. We can actually calculate how much revenue Twitter is losing directly related to Elon Musk's behavior.
---------
If you want to argue otherwise, you need to start putting down at least some napkin-level estimates to where the other losses-of-revenue came from for Twitter. Or you need to chip away at the $3.8 Billion number I posted above.
Coming into a $5 Billion revenue company and immediately causing $3.8 Billion (combination lost revenue and new liabilities) is very impressive. I don't think I've ever seen someone wipe out... 76% of a company's revenue in just 3 months before.
> I don’t know what this has to do with what I posted.
The problems you're trying to discuss are almost entirely irrelevant to Twitter's current woes. The actual problems are measured in the $Billions/year size. The problems you discuss are perhaps only a few dozen million/year.
> You are shadowboxing.
You are making a mountain out of a molehill. The $million mistakes should be ignored for the $Billion mistakes happening right now.
No, it's the other way around. The article from The Verge that you substituted points to and is based upon (after it was updated) the original source post by Paul Haddad that I gave, which didn't point to anything. Notice that the post referenced is the one in the large font, not the one at the top. (-:
I was going by the general principle of pointing right to the source, which is Paul Haddad's actual post, rather than secondary coverage of it by someone else, which only got updated after Paul Haddad had posted, as well.
They do not point to each other. M. Haddad's previous post point to The Verge. The post that I hyperlinked to came afterwards, and The Verge points to it. It does not point to anything.
No, that is https://tapbots.social/@paul/109695455993367527, which is the preceding post. Unlike Hacker News, Mastodon makes the post linked-to by the URL have a different background and a larger font. And https://tapbots.social/@paul/109695822047176004 is very clearly a post that contains no hyperlink at all. That's because it was posted before The Verge then went back and edited its article.
You see it mentioned on HN “twice within a day” because it’s likely one of the more popular twitter client, and it’s one of the client which got hit by twitter deactivating api keys unilaterally and without warning or communication.
Twitter never fed third-party apps ads. Part of why this decision is so questionable is because there were numerous ways Twitter could have monetized these apps more directly, yet never attempted to do so.
Most Twitter users don't use a third party client, but anyone who's anyone uses a third party client, and most of Twitter's features (retweets, quote tweets, pull to refresh, the use of a bird, the use of the word tweet....) were invented by third party apps.
If Twitter wants to be... like, a really bad instagram clone, leaning into the following celebrities thing all the way, then this... maybe works?
In other words, the people who are most lucrative to advertise to are not seeing any ads, and are not paying Twitter for the privilege of an ad-free experience.
It never occurred to me how strange it is that a company providing people an online socialization medium prioritized non-human entities’ abilities to use their platform.
This sucks, but I don't find it surprising when many of these third party clients are hostile to Twitter. No ads, taking UX out of Twitter's control, and promoting other backbends. It's not clear to me that it benefits Twitter to support these clients.
The Twitter API doesn’t give them the ads. They literally can’t display them.
As for the UX thing — Twitter owes a huge amount of its own UX to third-party clients. Hell, Twitter’s own iOS client was originally a third-party client they acquired called Tweetie. The developer of Twitteriffic made a list of all the things they pioneered for Twitter: https://furbo.org/2011/03/11/twitterrific-firsts/
There are plenty of good reasons why Twitter might not be comfortable giving ads as an API, and that means they're giving away their resources for free by providing these APIs.
Also, Twitter is a more mature platform now. They probably don't feel like they need the innovation from these apps anymore. But that misses the point, small changes like the view count on tweets are added to improve engagement / profit, so Twitter wants full control of what is seen.
Again, it probably does not help Twitter to support these people, even if they are a small part of the user base. Same mindset as banning competitor links in your bio.
You said “many of these third-party clients are hostile to Twitter”, and gave these as examples. If none of them matter, why did you mention them in the first place?
Sure, if the number of power users is low, not increasing, and these power users drive a lot of engagement. I suspect that at least one of the above is not true.
Also you should rethink your life if you consider yourself a Twitter/Reddit power user.
Even better: access to use third party apps is a second tier that costs $10, and give $1 to the app maker. All the incentives are nicely lined up: the user won't choose to pay more if it's not worth it so the alternative app is guaranteed to offer some added value, the app maker gets paid for their efforts and is inclined to closely follow Twitter's increased rate of feature development so at least they don't fall behind and ideally innovate, Twitter gets value in not just a bit of extra money but also platform interaction experimentation which they can mine for good ideas to integrate back into the official client, they're inclined to make the API good, etc. I don't know how most people would take to this idea, but I like solutions where the incentives are all aligned. Am I missing something?