> With these changes and reinforcements, I believe the situation has been fully resolved. Fingers crossed.
I get that this is a difficult situation for a small developer, but ending with this line did not fill me with confidence that the problem is actually resolved and make me trust their software on my system.
That's the most honest assessment you can expect from any small-scale developer. What do you expect them to say or do? Their adversary is presumably a national intelligence agency of a superpower.
The odds may be better if you operate the way OpenSSH does: move slow, security first, architect everything to be very difficult to attack. But if you're building a text editor, it's not your mindset, and probably never will be.
> The odds may be better if you operate the way OpenSSH does: move slow, security first, architect everything to be very difficult to attack. But if you're building a text editor, it's not your mindset, and probably never will be.
I mean, if you look at the Notepad++ website this developer seems just as concerned at spamming political messaging all over everything as much as he is with writing the software he's distributing. It's pretty crazy he apparently didn't think to take more basic precautions given he is basically permatrolling Russia and China with his messaging. Big brain moment for him. And meanwhile, after reading that disclosure nonsense none of us even know what's going on - like, should we be formatting machines that were affecting during that timeframe? Was the attack targeted and specific only? Who the fuck knows!
Binaries or source, it's pretty much the same unless you thoroughly vet the entire source code. Malicious code isn't advertised and commented and found by looking at a couple of functions. It's carefully hidden and obfuscated.
Before Tesla came along there were a small number of EVs but they were all pretty bad because their only purpose was to serve as “compliance cars” in states like California so automakers could sell more gas cars. (See the documentary Who Killed the Electric Car? for more on this.)
So Tesla deserves credit for building the first electric cars that people actually wanted to buy. They also deserve credit for building the largest and most reliable charging network - a key factor in making electric car ownership more feasible.
But they’ve made a lot of poor decisions recently and all the money and power went to Elon’s head. I think it was beneficial to the world for Tesla to exist and do that important work early on, and now it’s beneficial to the world for the company to die.
The sequel Revenge of the Electric Car is a very interesting follow up, especially now with hindsight. It followed the stories of the Tesla Roadster, Nissan Leaf, and Chevy Volt all in about the exact same time period and stages of development.
Of those, the Leaf is the only model that has continuously existed since then, and from the documentary there is a sense of that. GM admits the Volt was a stepping stone and not the final product. Tesla's part of the documentary involves a lot of trials and tribulations and even Tesla seeming unsure about their manufacturing problems. (Though the documentary itself spins a hopeful tune.)
Of the figures in the documentary the most prescient seems to be Carlos Ghosn, then in charge of Renault-Nissan. He very much insisted that EVs weren't just the future, they were the scramble for the present. Renault took that message to heart and seemed to be the side that won it in the messy divorce that also eventually wound up with Ghosn getting charged for treason and embezzlement in Japan. Which is an incredible and weird story on multiple levels and maybe the documentary makers will get a chance to include that in a third movie for the series.
Given that this is an experiment and the website says they want to treat Claude as a “true collaborator”, they should follow the AI’s directions EXACTLY. Claude alone should make decisions and no human should be allowed to deviate from its instructions, even if they know better. That’s what would make this a valuable experiment, otherwise if there’s a human moderating Claude then it’s no better than Googling.
Vimeo’s prices are insanely low for what they’re offering. You could not host on AWS and match them on price, nor probably even Hetzner. I never knew if they built out custom infrastructure or if they’ve just been losing money this whole time.
Lots of custom infrastructure. A bit of losing money as well, but moderately profitable on the whole. They were a publicly traded company in their own right from 2021 to 2025, so you can look at the 10-Ks. There was a massive boost in business from COVID in 2020 and early 2021... which meant that the spinoff in May 2021 left investors in a position to be perpetually disappointed, sort of like Peloton.
The spinoff definitely came at a particularly bad time, but that was always how it was going to end up because that is how IAC operates. Vimeo was probably doomed from the moment IAC acquired Connected Ventures. It was never going to be the kind of business that could really operate as a public company, but totally could have been a profitable private company.
The custom stuff was all on the transcoding and player sides. Actual video file storage and delivery was using a variety of well-known CDNs depending on when and where, but primarily Akamai.
Sorry, but Palantir doesn’t get off that easy. They know full well how their technology is used. Just because a market exists that doesn’t mean you need to fill it. The tech industry could have taken a moral stand like the chemical industry did with execution drugs.
If you watch any entrepreneur-focused channels, the entire premise of Palantir was "what if we just didn't care about what people think is ethically dubious? What if we went into business in all the places that people have traditionally shied away from for moral reasons?" It's part of Thiel's "Monopoly is good/You want to build the 0 to 1, not jump into a crowded market" mantra.
I started a company in that market 10 years ago. We compete with palantir. It’s a competitive market with lots of actors.
On of their strengths is the ability of thiel to raise lots of money, and win huge gov contracts by convincing everyone that what he built is magic. it is not.
palantir is regular enterprise software. morally, they are vilains for sure, but their superpower is being excellent at marketing themselves.
What I meant is that they espoused that attitude in the Silicon Valley world, which traditionally has not really invested in Defense. I imagine that's also why they're able to raise lots of money and build hype trains, they have one foot still in SV and SV VC.
Observability vendors massively overcharge. I got tired of paying an ever-increasing amount of money per month, so my solution now is a self-hosted SigNoz instance on a cheap Hetzner box. It costs me $30/month and I can throw large quantities of data at it and it doesn’t break a sweat.
There are features they are planning to make exclusive to the subscriptions. I don’t know if they’re planning to make the one-time purchase go away completely, but it seems like it’s going to be approached as the “lesser” option.
I noticed Apple’s software quality decline the moment they committed to 1-year release cycles. Because an x.0 release inevitably has issues, it offers less than a year of stability (sometimes only a few months if it takes until x.4 to be fully stable) before things get broken again in y.0. And because Apple stops signing old versions pretty quickly, you’re often stuck on an unstable new version if you take the risk and upgrade.
Additionally, it is hard on all developers (Apple included) to release updates for all of its many platforms on the same day, which IMO reduces software quality across the ecosystem.
(Apple also has the luxury of only supporting the latest OS versions with its software. Customers often expect third-party developers to support a wider range of OS versions and devices than Apple does.)
I have been using OS X since 10.4 Tiger. I still remember standing in line at midnight trying to get a copy on DVD. Getting to test all the new features back home in the middle of the night was so exciting! Well worth the €129/€29 they charged for it. Nowadays the yearly releases are more of a "meh". I hit install, they added a new grouping feature to Reminders and that is about all I use from what they added.
Still bitter that my 2006 Core Duo MacBook only had support up to 10.6 Snow Leopard but back then that was over 4 years of being able to use the latest OS, so comparable to four releases with the current cycles.
I used it up till 2011. It had multiple top-cases replaced under the extended warranty, display CCFL was changed a few times due to flickering, disc-drive got swapped once, new logic board because the audio-jack was stuck on SPDIF, new power adapter.
The only device I ever got Apple Care on and I got thousands in repairs covered for free. This was from before Apple would just replace the entire device.
All my other MacBooks have been trouble free luckily.
I get that this is a difficult situation for a small developer, but ending with this line did not fill me with confidence that the problem is actually resolved and make me trust their software on my system.
reply