Live data from Hacker News

Munder Difflin – Agent harness to run an office of your clones

munderdiffl.in

141–150 of 162 posts

Re: Munder Difflin – Agent harness to run an office of your clones

#141
post #139
post #99

Earlier quoted context omitted.

[flagged]

The people moderating this website are deranged, sadly

I never did Digg but I remember the issue there was the "power users".

Power corrupts, almost always. Most people are horrible when given power. The only thing holding them in-check is their powerlessness.

Re: Munder Difflin – Agent harness to run an office of your clones

#142

Earlier quoted context omitted.

Am I the only one who doesn't want a virtual assistant? I just struggle to understand this goal we apparently are all aiming for I want to be an active player in my own life. I want to do things even if there's some friction. All I can imagine is that future from WallE where everything is so easy no one tries or does anything at all

I'm just not that annoyed by having to search for a flight, or train, or bus. Or make plans. Or find a restaurant. I'm not annoyed enough by these trivial tasks to want to outsource them. I don't want to ask Siri to book me a restaurant, when the alternative is a quick google search.

It's helps to remember that most of the folks pushing this are ambitious but fundamentally lazy.

Re: Munder Difflin – Agent harness to run an office of your clones

#143

Earlier quoted context omitted.

In areas of applicability where GenAI to now considered to be good/great, the pattern I've observed is a frighteningly rapid progression from "abjectly terrible", to "middling and mostly more trouble than it's worth" and then "good/great". Agentic coding and image generation are probably the best examples of this. There's no guarantee that this trend will be seen with virtual assistants, but if it does, we're on the…

Am I the only one who doesn't want a virtual assistant? I just struggle to understand this goal we apparently are all aiming for I want to be an active player in my own life. I want to do things even if there's some friction. All I can imagine is that future from WallE where everything is so easy no one tries or does anything at all

Can’t say you’re the only one but I certainly want an assistant.

I’ve been putting off some dental work for a year or more because I have an implant that’s come lose, and the two dentists I saw about it said they’d have to remove the crown, send pictures to some lab that might be able to identify the manufacturer (or maybe not), maybe get the right tooling, maybe have to replace the entire thing (thousands of dollars, removing and resetting a post in my jaw), etc etc. all because I didn’t know the manufacturer of an implant I got 20 years ago.

Claude cowork was able to understand the problem, interview me, find dental insurance paperwork in my email, identify the prosthedontist used, find that that office had closed years ago, find where the practitioner moved to, their new practice, and get me a number to call.

Within a couple of ours I had the manufacturer, procedure notes, and X-rays from the installation.

Is that WALL-E style laziness? Maybe. But I’m good with it, as I have line of sight to solving an annoying problem that seemed insurmountable.

Re: Munder Difflin – Agent harness to run an office of your clones

#144
post #142

Earlier quoted context omitted.

I'm just not that annoyed by having to search for a flight, or train, or bus. Or make plans. Or find a restaurant. I'm not annoyed enough by these trivial tasks to want to outsource them. I don't want to ask Siri to book me a restaurant, when the alternative is a quick google search.

It's helps to remember that most of the folks pushing this are ambitious but fundamentally lazy.

What an odd claim.

Is it possible they have families, aging relatives to care for, health issues or other reasons that time is precious and “spend an hour correlating amazing restaurants with their available bookings against my crazy schedule” is not a good solution?

IMO the greatest failure of the HN-style tech community is complete lack of empathy, or even awareness, for people whose lives have different shapes.

Re: Munder Difflin – Agent harness to run an office of your clones

#145
post #140
post #68

Earlier quoted context omitted.

I can’t imagine having so little empathy that you can’t put yourself in the shoes of the creator of the IP. They didn’t even bother to tweak the characters. It’s not “inspired by” or riffing on a theme, it’s a straight ripoff being used to market some AI product. Do you think anyone who created The Office would want to be associated with AI, something that’s threatening to drive them out of a job? What pisses me off…

No man, if it's a "straight ripoff", it would be a TV show. Not an programming application.

Right, because a Temu backpack with Darth Vader on it isn’t ripping off Star Wars either

And not everyone on the internet is a man

Re: Munder Difflin – Agent harness to run an office of your clones

#146
post #141
post #139

Earlier quoted context omitted.

The people moderating this website are deranged, sadly

I never did Digg but I remember the issue there was the "power users". Power corrupts, almost always. Most people are horrible when given power. The only thing holding them in-check is their powerlessness.

[dead]

Re: Munder Difflin – Agent harness to run an office of your clones

#147
post #142

Earlier quoted context omitted.

I'm just not that annoyed by having to search for a flight, or train, or bus. Or make plans. Or find a restaurant. I'm not annoyed enough by these trivial tasks to want to outsource them. I don't want to ask Siri to book me a restaurant, when the alternative is a quick google search.

It's helps to remember that most of the folks pushing this are ambitious but fundamentally lazy.

frankly, this is just a lazy and ignorant pov

Re: Munder Difflin – Agent harness to run an office of your clones

#148
post #137

Earlier quoted context omitted.

If you want to over-analyze people's reactions, sure I'll help: They're pigeon-holing because it can't be both. This isn't a situation where there's real overlap. Ambiguity is a lot more fun when there is substantial overlap, and on top of that ambiguity for the question of "is this useful" is bad ambiguity much more often than it's fun ambiguity.

> it can't be both Really? Do you have a proof of that? Looking outside the realm of computer software, are you familiar with the concept of satire? "A Modest Proposal" is both funny and deadly serious at the same time. And it's been studied for hundreds of years. So I guess we'll just have to agree to disagree.

A modest proposal isn't serious about what it suggests. It's not a real way to solve either problem it talks about. If that's the best comparison, then this is straightforwardly not a productivity tool.

Re: Munder Difflin – Agent harness to run an office of your clones

#149

Earlier quoted context omitted.

I don't think it's an oxymoron at all. An expert climber still uses a harness for safety, even though they won't (typically) try to jump off the wall.

They would 1 times out of 10 if they were an LLM. Then say sorry, you’re absolutely right I shouldn’t have jumped off the cliff.

You know - that's why we use 2 agents for any important decision.

1 in 10 becomes 1 in 100. And you can get any degree of accuracy you desire. If only humans were so reliable.

Why are you ignoring basic principles of engineering?

Post reply on HN