There's a lot of tools now with a similar feature set. IMO, the main value prop an official OpenAI client could provide would be to share ChatGPT's free tier vs. requiring an API key. They probably couldn't open-source it then, but it'd still be more valuable to me than the alternatives.
I've ended up with $5K+ in a month using sonnet 3.7, had to dial it back.
I'm much happier with gemini 2.5 pro right now for high performance at a much more reasonable cost (primarily using with RA.Aid, but I've tried it with Windsurf, cline, and roo.)
I didn't know this, thank you for the anecdata! Do you think it'd be more reasonable to generalize my suggestion to "This CLI should be included as part of ChatGPT's pricing"?
Too expensive for me to use for fun. Cheap enough to put me out of a job. Great. Love it. So excited. Doesn't make me want to go full Into The Wild at all.
I don’t think this is at the level of putting folks out of a job yet, frankly. It’s fine for straightforward changes, but more complex stuff, like concurrency, I still end up doing by hand.
And even for the straightforward stuff, I generally have a mental model of the changes required and give it a high level list of files/code to change, which it then follows.
Maybe the increase in productivity will reduce pressure to hire? We’ll see.
I've been trying to get back into RSS recently. The problem I keep having is the dramatic weight difference between a news site and a personal blog; I'm occasionally interested in what the news site posts, but its volume is so overbearing that it's all I see in my feed. I just walk away with FOMO every time.
I'm currently thinking about trying Feedly AI as an algorithm that could surface good content for me.
I'm working on a feed reader that separates articles into Inbox and Library.
New content shows up in the inbox, where you can bookmark or archive. Bookmarked content shows up in the library.
Going through the Inbox and bookmarking interesting content is fast, and in the Library the high volume feeds don't matter as much (because you curated before and only have interesting content there).
I use TinyTinyRSS - it has very powerful filters that support regex. So I spend the time to write Regex filters that run over all the news sites, so as to only surface stuff I'm really interested in on the "fresh articles" page. I can still go in and look at every article etc.
You can also apply weights etc, so for the small blogs etc I follow, I give their new articles a high score they so float to the top of the reader.
It works really well and I don't feel like I'm drowning anymore - I have a massive amount of content still get imported, but only the stuff I want to see is what I'm presented with.
Twitter launched a paid tier, Twitter Blue, before the Musk takeover. It was a sensible move, offering additional features such as edit functionality for power users, but not harming the experience for those who chose not to buy into it.
Most importantly, it was entirely separate to the use of blue ticks to denote verified and authentic accounts. In fact, there was as far as I know no visual indication that someone was a Twitter Blue user.
When Musk took over, one of his first changes was to combine Twitter Blue and the verified programme, removing any actual verification in favour of a claim that anyone who had paid was 'verified'. The company also artificially boosts the posts of those who have bought this new premium tier, which disadvantages everyone else.
Buying a blue tick is seen, quite rightly now, as buying an artificial boosting via the algorithm and that's just pure vanity.
> Most importantly, it was entirely separate to the use of blue ticks to denote verified and authentic accounts
Blue ticks had nothing to do with being "verified" or "authentic." They were handed out randomly to some celebrities. bigcorps, and friends of Twitter employees. As far as I know a bluecheck was never handed out to anyone who wasn't who they said they were (although you could get them under pseudonyms too), but the "wrong" celebrities or normal people could not get them. It was a mark of social cachet, not authenticity.
That’s false. The programme was explicitly launched to highlight accounts that Twitter the company had verified as actually representing the person. [0]
It was launched after the company was sued by a sports star who was impersonated. Until Musk took over it could always be relied on as proof of an account’s authenticity.
What you’re complaining about isn’t that it represented something else, but that as with any large scale verification process it doesn’t scale well and the process was opaque.
Twitter cycled through various approaches to solve the problem. At first they merely reached out to people, based on the popularity of accounts representing them, their level of fame, and their risk of impersonation (or evidence of prior impersonation) and as and when the team had capacity. Therefore celebrities, politicians, and journalists were prioritised. However, this was a flawed process that was too US-centric and too US West Coast-centric on top of that.
They later opened up public applications but closed it down when they were overwhelmed with requests and couldn’t process them all. They ran into all the usual verification challenges, including language differences and the difficulty of verifying across countries. So they reverted back to the previous model while working on a way to use external factors to allow for a scaled up verification process in the future. That process stopped when Musk bought them.
The claim that it was only handed out to people for social cachet doesn’t hold water. For one, Republican politicians, Fox News and other conservative publication journalists, and sufficiently notable right-leaning celebrities all had blue ticks. The company even verified Jason Kessler, though that stirred up a public controversy that forced it to pause the programme for a bit. [1]
It wasn’t a perfect programme of course, but it did what it said it would: If you saw that an account had a blue tick you could trust it meant it was an authentic account that Twitter had verified. That aspect of the system never failed.
Yes, bluechecks were who they said they were. But just because you were who you said you were did not mean you could get a bluecheck.
Bluecheck -> authentic.
Authentic !> bluecheck.
There was no path for most people almost all of the time to get a bluecheck, and they could be removed for arbitrary reasons having nothing to do with your identity. The "process" consisted of someone at Twitter deciding you were worthy, based on arbitrary criteria including "you're my friend."
> The claim that it was only handed out to people for social cachet doesn’t hold water. For one, Republican politicians, Fox News and other conservative publication journalists, and sufficiently notable right-leaning celebrities all had blue ticks.
Having bad politics doesn't mean one doesn't have social cachet. You yourself say here "notable" - but again, just being notable was not enough, though it was a big help. And of course, if you had the right friends, you could be completely non-notable.
Your post begins with "that's false", but the rest of it agrees with what I said.
In many old Twitter circles, having a bluecheck made you an object of derision for precisely these reasons - it didn't signify authenticity, it was a social marker that frequently came with inane tweets and thin skin that was perceived to arise due to an idea they thought they were "elite."
Your post said Blue ticks had nothing to do with being "verified" or "authentic." and It was a mark of social cachet, not authenticity..
That is clearly false. The blue check was an accurate signal of authenticity through verification by Twitter. What it was not was a universal verification mechanism that scaled to being able to verify everyone who used the platform. Those are two different things, and the lack of universality does not mean that the programme was not useful in valuable in providing a means to prove authenticity and avoid impersonation of the many people it did cover.
Moreover, the company was actively working on ways to scale it up into being able to reliably verify many more people before Musk bought it.
Now all that is gone, and a blue check means nothing to other users of the platform other than as a sign that the holder is paying for premium features. It's no longer a trustworthy verification mechanism.
And yet somehow, impersonation is no more a problem on X than it was on Twitter.
The fact is nobody paid attention to the blue check except the people who had one. The outcry was because a set of people felt elite because of their blue checks and were upset that anyone could get one. Hilary Clinton or Donald Trump or Kanye West impersonators are not an issue on the platform.
There have already been high profile impersonations, one of which caused a company’s stock to dive. Another was the infamous situation where Musk freaked out over people impersonating him.
The only reason it’s not worse is because of inertia and the fact that most public figures are still using the accounts they had when verification was in place. That won’t last. It’s also why public figures who are leaving for other platforms usually opt to keep their X accounts in place but dormant.
Kanye was in fact impersonated on Twitter in the past, it was one of the key reasons behind the introduction of Verified Accounts.[0]
I have also seen impersonators for all three of those you mentioned, especially in terms of crypto bots for Trump. However those are also three of the most heavily policed and watched profiles by the company’s understaffed moderating team, who proactively monitor for impersonations. That’s not true for lesser public figures: I just searched for a hand full of well-known journalists and found several impersonation profiles for them all.
What’s worse, their impersonation reporting function does not let you report the impersonation of someone who is not an active user. I know another public figure who doesn’t have an account there, doesn’t want one, but now hasn’t been able to get them to do anything about an impersonator who is using his name to scam others.
The current model is badly broken and decidedly worse than the verification system that existed before.
What X gets mocked for is that the feature of being "verified" by Twitter was worth paying for. But X turned the "blue tick" into "this person has paid" which totally removed any value for the blue tick.
It was yet another self-own, along with things like getting promoted on the feed.
X could have made money by selling features (eg tweetdeck) and other useful things, but they didnt.
Of course, then politcally it became a "this person is a MAGA person" so Musk had to start handing out "free" blue ticks to big accounts to avoid that appearance. To the point that people were putting in their name "didn't pay for blue"
But they did. Nobody pays for X premium just to get a blue tick. Being verified I guess is part of it, but they're paying to get other things including less ads, Grok access, longer posts & videos, communities, media studio, monetization, and other things.
The blue tick doesn't cause the reply boost. The paid plan level does, from small to large boost. Blue tick means you supplied a valid phone and credit card - a strong indicator of a genuine account.
I'm trying it to check out Grok, which is impressive now that it can handle current events and return posts and web links.
On the negatives, they claim "download videos" as a feature of paid plans. This is false advertising. Many videos cannot be downloaded. The uploader must specifically set the video as downloadable, and many do not. X is deliberately misleading about this. It was a key reason I paid.
reply