$2700/mo is about 1/3 of an engineers' salary (cost to the business of a mid-level engineer in the UK)...
But, there's the time to set all of this up (which admittedly is a one-time investment and would amortize).
And there's the risk of having made a mistake in your backups or recovery system (Will you exercise it? Will you continue to regularly exercise it?).
And they're a 3-person team... is it really worth your limited time/capacity to do this, rather than do something that's likely to attract $3k/mo of new business?
If the folks who wrote the blog see this, please share how much time (how many devs, how many weeks) this took to set up, and how the ongoing maintenance burden shapes up.
One thing that is interesting about this is that I wasn't able to get good results from smaller/faster models like claude haiku, so I opted to use a larger model instead. I found that the small delay of about a second was worth it for more consistent results.
I never got the valuation. I (and many others) have built open source agent plugins that are pretty much just as good, in our free time (check out magenta nvim btw, I think it turned out neat!)
I recently finished some updates to my neovim ai plugin:
- context tracking
- multiple threads
- compaction
- sub-agents
I decided to record a video of myself using the plugin to implement another feature. Aside from a demo of the plugin itself, I think it's a good view into what a typical workflow of using AI for development might look like, and I think is a good illustration of what these agents can and cannot currently do.
I agree - there seems to be talking past each other about some very fundamental things:
How extensive is the violence of the protests? I saw some images shared of cars that were burned, maybe some buildings damaged. But also lots of images from other protests from previous years. Are the images of the same 3 cars and storefronts or many? Trump says the riots are out of control, Newsom says the protests are largely peaceful.
How much of a strain do undocumented immigrants place on the US? You can answer this question from a financial and criminal point of view. From the point of view of crime, Trump and ICE are parading every violent undocumented immigrant they can, but that is not statistics. Do undocumented immigrants account for a significant portion of violent crime in the country?
From a financial point of view, what resources are undocumented immigrants straining, and is it to a significant degree?
The economic picture is much more nuanced. On the cost side, a criticized study (FAIR) reported the cost at about $182bn annually (this is likely an over-estimate). For comparison, undocumented immigrants pay about $100bn in taxes, boost the GDP, and create jobs. Mass deportation is estimated to cost $315Bn.
From what I've seen the burned cars are Waymos, which have their own set of issues and the burning is probably more opportunistic than related to the protest in general. People have been protesting Waymo for years now, obstructing them with cones and other such vandalism. One big thing is that since Waymos are driverless the violence is not being perpetrated against people.
Its LA. When the dodgers won the world series they burned a metro bus in the streets. And those events were way bigger and drew a lot more chaos and crowds. But of course downplayed due to a lack of a political angle at the time to milk out of the event, unlike now.
I was at OSU when they won the 2018 natty. They deployed the horse cops. The riot squad. Teargassed most of High st and campus. Arrested dozens. But it was legendary, and I will never forget that night and its energy.
Excellent, so it's working. Waymo has been nonconsensually testing their unready "autonomous" cars on people for far too long.
As recently as last year a Waymo stopped at a stop sign and then accelerated directly towards me in a crosswalk.
These vehicles aren't ready and have already injured people. Lighting them on fire seems a valid defense strategy since the government's unwilling to intervene to protect its citizens.
Personally I can't wait until Waymo reaches the east coast. Within the past year, my neighbor was hit in a crosswalk pushing a baby stroller (minor injuries to mom and baby, but could have been much worse). And while walking my daughter our of an kids art school, an impatient car sped past the cars stopped at the crosswalk and were a foot or two from hitting my kids. These were both human drivers. And both cases would have been avoided if they were Waymos. In the first situation, the driver was only looking forward instead of the direction they were turning.
Any mode of transportation will cause injuries, especially since other humans are on the road. So just saying "Waymos have already injured people" is kind of a meaningless comment. I do think the type of accidents matter as much as the number of incidents. For example, the video that went viral recently of a self-driving Tesla randomly making a dramatic left turn into a tree on a rural road (possibly because it misinterpreted a shadow in the road) is not a mistake a human would have made.
> And both cases would have been avoided if they were Waymos.
This is speculative. Waymos hit people too.
I agree that pedestrian infrastructure in the USA is sorely lacking. The proven solution that worked in other countries is to take measures to reduce the number of cars on the road, not try to replace every driver with a computer that can only be trained by putting stupider computers on the road first to experiment on the population.
Public transit. A subway moves literally millions more people than car infrastructure can with significantly fewer injuries - basically 0 if the platforms are built with doors or gates. Busses and cable cars, driven by professional drivers, have far lower incidence per capita of injury as well.
What I was hoping to see from your post was some stats showing Waymo incidents where Waymo is at fault. I'm not actively following this stuff, but after some googling it's hard to find anything. There's tons of links with click-bait titles and then once you click into them you see things like:
- "Waymo drove through a red light (while being driven by a human)"
- "Waymo involved in a hit and run (it recorded a hit and run by a human using it's cameras)"
- "Waymo was involved in a multi-car accident (while it was stopped with other stopped cars)".
I'm not saying they don't happen. I'm sure they definitely do. As a father of two little girls, I would feel much safer with them getting into a Waymo for a short trip than an Uber. With that said, I'm strictly talking about Waymo. I would never get into a Tesla Robotaxi.
I understand that what I'm basically saying is that, if Cruise, Waymo, etc, want to develop self driving cars, they need to basically figure out a way to do so without testing in live environments against a non-consenting population. I understand that for you, you're down, and it's frustrating to hear that someone who might be your neighbor not wanting that means you don't get self driving cars.
I don't have a solution for this annoying aspect of democracy, other than to try to convince you that I'm not crazy for not wanting self driving cars at all - really, truly, it's just not the path we need to go down, we should be getting rid of cars almost entirely in cities and switching to public transit, with private vehicles driven entirely by professional drivers.
I chased up the old email. First, apologies, it was a Cruise driverless car, not Waymo.
I informed them of the incident, and here was their response, about 24 hours after my email to them:
> Copy of CUSTOMER SUCCESS (2)
> Hi Caleb,
> I'm sorry to hear you experienced this. At Cruise we take safety very seriously and this is not the experience we’d like you to have. This issue is being escalated to the appropriate team to be looked into further. Your input is greatly appreciated as we grow our ride-hail service, We appreciate your assistance.
> Best Regards,
> Jamilla
> Cruise Support
I never heard back from them again. The incident was on sep 4, 2023, at just about exactly 4pm, in San Francisco around potrero hill, perhaps Mariposa street. The vehicle was named "Bruschetta," I believe I saw that on the back or something.
The key tell is.....CNN for example is mentioning with specificity how many cops and military are on the ground but only uses the word "large groups" when talking about how many protestors and rioters there are. Mentioning the actual small numbers of protestors / rioters doesn't allow them to sensationalize this.
The police and military are giving numbers, so you can easily publish those. I don't think the protesters are counting themselves, or maybe you could use AI and drones to do a survey of how many protests are out?
Even with you trying to be factual, how you present this still shows your bias. Plus you rationalize the opposite arguments away (“overestimate”) and boost the ones you agree with.
From the first Claude link:
> Damage remains far below 2020 George Floyd protest levels
Ugh. Is this the bar now?
Aside from quantifications, I see the disagreement largely on the qualitative/philosophical/ideological (tribal) side. An inverse of the Jan 6th insurrection (which also caused much less damage than the George Floyd riots, and much less than now in LA, so was it ok in retrospect?). But the damage was not in property, but that the state and its laws itself was attacked. The sides are just now switched again between left/right: the former shout rule of law, the latter want resistance/insurrection.
I do have a bias, and I don't think that negates the points. I think that's the benefit of establishing a common factual basis. You can argue with the bias by presenting competing facts, or a different interpretation of the facts, but at least we can agree on what we're starting with.
I think the question is - why is the national guard and the military being sent to CA without the governor's consent?
Part of the justification for this from the administration is that the riots are out of control, are posing an immediate risk of violence and property damage. Based on what I've found of the actual violence and damage being done, this justification does not hold up, as the violence and property damage are lower than previous protests in which there didn't seem to be a need for interference.
The rule of law is a different argument. What is the rule of law that is being undermined? I think here too you can have an argument about the operations that ICE is conducting, are they lawful - given that they are being conducted in sanctuary cities? Who has jurisdiction in this case? Is the administration lawful in sending in the national guard without the consent of the governor? What about the military? Newsom is now suing Trump over deploying the marines and the national guard despite his wishes, so there is a claim that such actions were unlawful.
Based on my brief research into this, ICE was operating in an unconstitutional way and making many procedural violations. City Sancutary status is lawful and has been upheld in the court of law. Newsom's challenge of Trump's deployment of the National Guard also held up in court. [link](https://claude.ai/public/artifacts/99a64b2d-e3b6-4d37-956f-c...)
When you say that "the law is being attacked" in this case, what do you mean?
There are lsp tools available to the agent, so in theory it should be able to ask for types, references and diagnostics.
In using it I've found that the agent doesn't really make use of these tools unless I instruct it to. I think I need to do some messing around with prompts to encourage the agent to use them more.
I think the next major thing I want to add is commands to allow you to send symbol types / diagnostics to the agent before it asks for them, which should help speed up some of these workflows.
Thanks for the pointer to take another look at aider - the DIFF format sounds really interesting. Though it seems like the license is Apache 2.0 so I think I'll have to learn a bit about what that means since I made mine MIT licensed.
But, there's the time to set all of this up (which admittedly is a one-time investment and would amortize).
And there's the risk of having made a mistake in your backups or recovery system (Will you exercise it? Will you continue to regularly exercise it?).
And they're a 3-person team... is it really worth your limited time/capacity to do this, rather than do something that's likely to attract $3k/mo of new business?
If the folks who wrote the blog see this, please share how much time (how many devs, how many weeks) this took to set up, and how the ongoing maintenance burden shapes up.