What can be asserted without evidence can also be dismissed without evidence. The benchmark creators haven't demonstrated that higher scores result in fewer humans dying or any meaningful outcome like that. If the LLM outputs some naughty words that's not an actual safety problem.
How stopping using hyperscalers models on their infra would "get as much of this capability into the open as possible"?
Either "we" create models better than commercial state of the art (by using whatever means).
Or we use open models AND fund organisations building such models (could be by purchasing service from these orgs or donations - in which case would these orgs be different than hyperscalers?).
But i dont see how just hosting the models on some private servers would give us an edge?
I use a film camera, which removes the possibility of infinite shots. It works for me! And since I must print the pictures, I make duplicates and share with friends and family.
You can give it a try: there are very easy cameras!
I was looking for a similar experience, where I take a picture only to immediately go back into the moment, without disrupting anything.
All those compact cameras were still too large for that purpose, because as the saying goes: "The best camera is the one you always have on you"
After some searching I found an AWESOME gem of a little camera: The DxO One, a camera from back in 2016 from DxO, a company actually specialized in benchmarking camera-quality.
The device was built to be sold at ~700 USD, but flopped and can still be purchased as old-stock for ~110 USD now.
Size is fantastic, Pictures are great, I can only recommend it if someone just wants to capture a moment like it is without people getting pulled out of it.
The device is also somewhat hackable [0], as it's based on the Ambarella platform (RTOS and Linux).
I started collecting infos about it to preserve it [1], there's still alot of potential in this little gem.
Don't know what country you're in, but in general people has been picking up analog cameras again. Madrid probably has at least doubled or tripled the number of print shops in the last 5 years.
The real problem is the price of film has skyrocketed. Since factories has been closing in the last ten years, the offer is low, and the demand is high... Even low quality Chinese stock is going for prices higher than professional rolls where 5 or 6 years ago.
for someone who wants to try that without investing in film camera and film development, there are small handheld printers that can print directly from a phone (and I guess a computer). It doesn't remove the possibility of infinite shots but it allows to focus on the shot you really want to keep and print
This goes well beyond historical precedent. The legal system is complex, but even when Parliament approves a law, it can still be struck down if it violates fundamental principles.
For matters concerning IT and privacy, the CNIL (French Data Protection Authority) could spearhead such cancellation proceedings.
And companies doing business in France should watch out—CNIL sanctions are no joke!
Are there specific events making you feel it has been neutered?
At university, we used this extension to teach our classmates about good security practices, such as locking their computers when left unattended. It was fun, especially when professors didn't lock their computers. And my former classmates did learn to lock their computers :)
I once pranked a coworker/friend with a Windows installation screen after lunch break. He was … astounded. The thing is, we were all using Debian in this company.
A roommate of mine in college used to leave his laptop unlocked all the time, and I found an app that would put an overlay on the screen that looked like a kernel panic. This went on for months, and he became convinced that his laptop had some issue where it would panic if he left it idle for too long. One day he happened to be going through his apps folder, and he saw something with a name like "iPanic.app", and watching his dawning comprehension as he realized what just must have been going on was probably the satisfying conclusion to a prank I've ever experienced.
violating security policies in order to “teach a lesson” is a sure fire way to get people to lose trust in you.
Accessing someone’s computer and manipulating the software was instant termination at my old company. Some new security guy joined and tried to do what you did. Find unlocked computers and mess with them to prove a point. He lasted a week.
There is a time and place for everything—and you should not assume a business environment is the only possible setting in which colleagues might pass by unattended workstations.
Ideally the prank is pulled in a high-trust, low-stakes environment like a college campus or high school computer lab, before corporate policies are part of one's life.
It is also a rich tradition, from the days of yore, before robust security practices became standard:
I would much rather my colleagues be taught this lesson (even if just through a verbal reprimand) than work with someone who is allowed to remain ignorant of the risks of their behaviour.
At the same time, a new hire could actually be a pentester, investigator, or corporate espionage actor. I know people who’s job this was to take over employee computers while the target went to lunch
It depends on the company and probably even the team. At least when I was running an IT team I generally viewed a colleague doing something like this as more effective than me nagging some sysadmin about them leaving their computer unlocked. Would have never tolerated someone on my team doing it to someone outside the team though.
I worked at a place where if you left your laptop unlocked, anyone could use your slack account to announce you were buying breakfast for the team tomorrow. That was more effective than any training video they could have made us watch. But I obviously wouldn't do something like that as a lone wolf.
> Accessing someone’s computer and manipulating the software was instant termination at my old company. Some new security guy joined and tried to do what you did. Find unlocked computers and mess with them to prove a point. He lasted a week.
That's a very strange policy to apply to your security team. They have good reason to make a point about leaving your workstation unsecured.
Working for NCC Group, the expectation was that if you left your computer unsecured, something would happen to it, and you, not the person who followed office policy by highlighting your mistake, would look bad.
I’m of two minds about it. I agree that these days it’s by far the safer choice to steer clear of such antics.
But I do sort of miss the days when we had a little more fun with computers even at work. Twenty years ago it was pretty ubiquitous to get a goofy desktop background if you left your machine unsecured all the time and I never saw any harm come from it.
It is definitely a better CYA move to just have a policy that nobody touches the unlocked computers, but is it actually more effective? If the company mostly employs adults that can be trusted to keep their pranks reasonable, it seems like a good way of self-policing.
If calling out somebody’s unlocked computer gets them punished for real, nobody will call out their friends…
At Amazon there was a "unicorn game". If you find an unlocked computer, you could send "I love Unicorns" message using the credentials of the logged on person.
There was even an internal site with the unicorn image.
I guess it’s a company cultural thing. In one past company, the SECURITY guys were the ones to do this to us teach us a lesson.but rather than a panic screen, it was porn.
To this day a few milliseconds before I stand up I wiggle my mouse to lock the screen. Muscle memory because lessons were learned
At my office it was either a picture of a shirtless David Hasselhoff as your desktop background, or an email sent to the networking+devs list announcing that you were giving away $20 bills at your desk, lol.
There's definitely a difference in company culture. One place I worked at you'd shout donuts into the office chat from your coworker's unattended laptops (and they'd be on the hook to bring in donuts or equivalent).
Always easy to catch the people who usually work from home.
One jnr dev at a place I worked left his desktop unlocked and a very elaborate email about his love for my little pony and wanting to start a company my little pony fan club was sent from his account to whole company lol.
Ironic, given that a ton of the security dogma these days is "don't trust anyone" --- you can guess why that started happening; precisely because of people like him.
Yeah I lean on this side - avoid doing pranks and other practical jokes.
When there is any actual malware or security incident, you don't want your colleagues to think of you and go "Maybe this is just Dave pulling one of his clever pranks".
This article is extremely well written and energizes me... Having a framework that allows separating the graphical work from the logic would be great. But I have trouble imagining this idea passing the brutal test of production -- except for showcase websites and blogs. I'm specifically thinking about apps (linear?)
But
I'm looking forward to seeing what comes next and hope to get more information when the templates arrive :)
You're not alone! The feedback form shows overwhelming relief from developers tired of complexity. Single-page applications will be the true test of the standards-first approach, as that's where the contrast with current practices becomes most stark.
LLMs contexts are fast to overload, as the article states.
That's why he writes smaller, specific packages, one at a time, and uses a web UI instead of something like cursor.
I had the same issue as you a few days ago. By separating the problem in smaller parts and addressing each parts one by one it got easier.
In your specific case I would try to fully complete the business logic one side. Reset the context. Then provide the logic to a new context and ask for an interface. Difficulty will arise when discovering that the logic is wrong or not suited to the UI, but i would keep using the same process to edit the code. Maybe two different contexts, one for logic, one for UI?
I'm french. Live in Paris. I do not understand the excitement the world has for Notre Dame.
Question for you, reader: what does the fire and/or the rebuild means to you? Why do you care?
I'm guessing it's because of "The Hunchback of Notre Dame", which if I recall correctly, Victor Hugo wrote to (successfully) stimulate desire to repair/restore the cathedral around 1800.
For me, personally, I think that Gothic architecture is the height of beauty, especially the interior. Notre Dame is one of the first examples, and was built quickly enough to have a consistent architecture (also Chartres and Amiens). Many Gothic cathedrals took 100 years to build and you can see variations in the architectural style in the different pieces which lessons the beauty somewhat. I also like the French circular apse rather than the English square apse (although that does allow for a gorgeous, huge window).
I also think beauty is important, especially in buildings, and Notre Dame is just a great example. It's beautiful inside and out. I've even spent quite a long time just enjoying the beauty of the back side.
For a lot of cultural/literary reasons, I think it's just iconic in a way that many other (at least equally beautiful) cathedrals--that most people probably couldn't even name, even the ones in Paris--are not.
I consider it to be a more recent (but still very old) entry in a category of human constructions that also includes the Parthenon, the Colosseum, the Taj Mahal, Angkor Wat, Machu Picchu, and Stonehenge.
I visited Notre Dame the first time I was in Paris, stood in the queue, climbed the stairs and took the photos everyone takes and have positive associations of the trip. The age, location and the view all help make nice memories that I guess many visitors feel. Victor Hugo & Disney probably add to that for others.
