I tried this for a little while but quickly stopped as a critical mass of websites broke when I tried using it to sign in. Special characters in your email address is an edge case that produces inconsistent results even within a single product
YMMV, I think I only tried to sign up on 3 websites where it was not working. You can fallback to the original email address in those case.
The funniest part was that for one it work great for the signup part, but they used a third party tool for licences that broke because of my e-mail.
For another, only the js code was verifying the e-mail, and I could push it by removing the validation. When the owner had to validate my account, they got a message that the e-mail was incorrect when they tried to submit the form. They called me and had a great discussion about web apps security. We had a good time.
I would point out that it kind of prevents you from checking if your email is in a leak database as you need to test each aliases you used.
Consistency is often helpful, but you also need to be wary of cargo culting. For example, you see a server back end that uses an ORM model and you figure you'll implement your new feature using the same patterns you see there. Then a month later the author of the original code you cribbed comes by and asks you, "just out of curiosity, why did you feel the need to create five new database tables for your feature?"
I know, that's a pretty specific "hypothetical," but that experience taught me that copying for the sake of consistency only works if you actually understand what it is you're copying. And I was also lucky that the senior engineer was nice about it.
As with any of these LLM workflow automation tools, it raises a few questions about each potential use case, and the likely long-term outcomes.
1. Is this working around friction due to a lack of interoperability between tools? For example, is this something that would be more efficient if the owner of the website exposed a REST service? Will the existence of this tool disincentivize companies from exposing services when it makes sense?
2. If there is a good reason for the lack of a service endpoint, perhaps for security reasons, will your automation workflow be used to bypass those security measures? Could your tool be used by malicious actors to disable major services? Are you that malicious actor yourself? Will your tool be used by scalpers to prevent consumers from buying high-demand products?
3. If this is being used to work around deferred maintenance with internal tools and processes, will the existence of these kind of tools be used by management to justify further deferral of that maintenance? Will your tool become a critical piece of the support staff's workflow?
4. If your tool is being used in good faith to work around anti-patterns in website design, will the owner of the website be incentivized to break your workflow? Is your use case just a step in an arms race?
These are the thoughts that go through my head whenever I hear about software being laid on top of complicated processes, where instead of simplifying the underlying processes, we add another layer of complexity to sweep it under the rug. I'm sure that people will find your project useful, but I wonder what the longer-term effects will be.
1. Yes absolutely. But the issue is a little bit more nuanced than that. Websites without APIs don't have them for one of two reasons: (1) They want to protect their data (LinkedIn) or (2) can't be bothered to make an API (boutique websites, government portals). This solves that problem, but also makes it so these websites never have to build an API (after LLM costs go down).
2. We don't want Skyvern to be used on websites that prohibit this kind of behaviour (LinkedIn is the obvious example). Specifically, we didn't open source any of our anti-bot or captcha related code because we get requests to make "Reddit upvote rings" and such. We don't want to support bad actors like that
(3) I think this is a net net good thing. AI browser automations= less need for APIs = no need to maintain both an API and UI = streamlined experience + less code = simpler systems
(4) I'm not 100% sure about this one. We usually just assume companies don't build APIs because they don't have budget for it. Ie for non malicious reasons. Companies like LinkedIn will likely thwart any attempts at automation, but we're not interested in participating in this cat mouse game
I think 100 Gb of GPU memory will always cost multiples of CPU + regular memory.
Using LLMs and computer vision for these kinds of tasks only make sense in small scales. If the task is extensive and repeated frequently, you're better off using an LLM to generate a script using Selenium or whatever, then running that script almost for free (compared to LLM). O1 is very good at it, by the way. For the $0.10 of 1 page interaction charged by Skyvern, I can create several scripts using O1.
Unless we want all those ISPs to be digging up all the streets all the time, that means we need strict regulations on sharing of physical infrastructure then, which brings us back to the CLEC/ILEC wars of the late 90s. Having each company maintaining its own last-mile access is an extremely inefficient use of resources.
I sympathize with your situation, and I appreciate that you're trying to be a cool parent who accepts your daughter unconditionally for who she is, but I wonder if it's even possible to know where the limits are on that, or how much she feels that way. Personally, I grew up in a really supportive family, but I still had a whole lot of hangups as a teenager that meant I didn't always want my parents to see what I was doing or looking into on the internet. Some of those private interests turned into major parts of who I am, and I don't really want to imagine the world where I had to worry about what my parents saw.
Would it have been better if I'd talked these things out with my parents? Maybe, but would I have? Or would I have just self-censored my dreams rather than face that conflict?
I guess maybe as a middle ground, maybe keep watching while she's young but pull back as she gets older?
I saw that demonstrated at the Toyota Museum of Industry and Technology in Nagoya. One of the tradeoffs with that is that you can't use cotton because it won't interact cleanly with the water, so it only works for synthetic fibers.
If you ever find yourself in that part of the world, I highly recommend visiting the textile pavilion there. They basically show you 1000x throughput improvement developed over the course of a century, demonstrated in about an hour.
The cadence and format of this video actually makes it look AI-generated. The content is padded and repeats itself quite a lot, the video is pretty much all stock footage, and the audio lacks any kind of variation in style like you'd expect even if someone were reading off a script. It sounds exactly like what you'd hear if you fed a script into a modern TTS.
Unfortunately that’s how many YouTubers make videos. I was watching a video about Chinese manufacturing the other day by Bloomberg, and 15 minutes worth of content was thinly stretched to 45 in a similar manner.
Came across many YT videos where you can almost smell the lifted content and structure from Wikipedia (or another video), often by people who aren't experts on the topic they're covering. History channels that are unable to tell the difference between England and Great Britain being a personal peeve.
Google is quite good at detecting duplicate text content, apparently duplicate video content isn't so much a thing.
My guess is that startup costs represent the bulk of the cost of a show. Just filming one episode probably costs somewhere on the same order of magnitude as an entire season.
I think this is a straw man. I don't think any politician would ever suggest such a thing at face value unless couched in some other argument. For example, after 9/11 a politician would have said, "I respect your need for privacy, but we live in extraordinary times, and we need to temporarily lift the restrictions on our agencies to effectively combat an imminent threat."
The real argument being made isn't that we should give all our secrets to the government, but that we should trust that the government will comply with our 4th Amendment protections and avoid gathering this data without a valid warrant issued by an impartial judge. If you're reading this, I'm guessing that like me, you don't believe a word of that. But that's the argument that's being made, and that's what needs to be debunked loudly and publicly.
> If this is really their argument they've already lost.
This would only be true if government were in fact answerable to people. It isn't, and hasn't been for a long time, if it ever was. Legislators decide what they want to vote for, and then decide how to brush off people who tell them anything they don't want to hear.
Lost in which arena? Do you trust politicians to not expand the scope of governance? If they are called out, they can simply rebrand their efforts under a new bill until 3 letter agencies get all the goodies they desire. How much patience does a low information public have? Is it as much as the political and security state classes?
It's hard to say without actually knowing the internal workings of the police departments, but one thing that's happened in Seattle is that although the police budget did go up, the SPD is having a really hard time hiring, and a lot of that money has gone to headcount that has remained vacant.
I've certainly read theories that it's a malicious strategy of the police union to make the city council look bad, and the SPOG president has made some pretty shocking political statements, so it honestly wouldn't be that surprising.
I'm guessing it's a mix of both, but I can't really prove anything.
reply