jedberg

No it isn't. I've put human written material in there (my own unpublished material) and it says 90%+ AI. Then I put some AI material in and it said 30% chance of AI.

I only tested it with the two pieces, but was not impressed.

I'm curious if you've tried Claude. Subjectively, I've always preferred Claude's writing over ChatGPT and Gemini.

> Currently we don't allow for multiple model selection, but it is on the roadmap.

Cool! What model does it use now? Enterprises will definitely want to know.

And the other big question: Why not just use Claude or Codex (via CLI on a desktop or droplet)?

There is a lot of competition in this space, both commercial companies and open source. It seems like the biggest advantage for using one of these and not Claude or Codex is that you can use multiple models. Otherwise, those two have all the same features, or probably will in the next week.

Are you able to use multiple models? Or I guess the first question is, what model(s) are you using? At this point, many enterprises care so that should be front and center, especially if you're using models hosted in China, because a lot of non-Chinese companies care about that.

You mention some things about unique ways to store memory. Do you have any data or use cases on how that improves performance? If you do, you should get those on your website too.

I'm not sure you have a very strong answer to "Why not just run Claude or Codex on my laptop?" Sure, you provide compute, but so do they, and most enterprises and even regular people are getting desktops/droplets to run their agents now because of this problem. This is kind of solved already.

Also there is a small nit with your website. The graphic next to "Proactively gets things done for you" gets bigger and smaller, so if you're trying to read anything below that, it keeps jumping up and down.

My last few jobs I started with a listening tour. I talked to everyone who was relevant to the job and asked them:

- what is going well,

- what is not going well

-what do hope that I will fix

-what should I do

-what should I not do

I put those question in the agenda for the meeting invite so they could come prepared. Then we took the conversation from there.

I put it all into a google doc, then synthesized it into summary that hid the identities of the people who said it and used it as my guide for the first few months.

That's a good point. The other examples didn't involve copyright.

Hopefully since it's so old, the copyright holders are cool about it.

ps. Hi Seth, good to see you around!

This is pretty well trod legally. By the user taking a specific action, it makes the user responsible. Comma.ai uses the same technique. Technically their self driving hardware comes with no software. You have to click a button and point it at their GitHub to get the software on it, but it's automatic after that.

Remember, Nvidia sells services too, not just GPUs. They also sell full racks of servers. This is just sales, not profits.

Amazon is actually pretty good about open sourcing core technologies like that. Firecracker is a big one. They also support Valkey and pay some of its maintainers.

> His companies have made many monumental engineering achievements at his direction.

If you know anyone who has worked at one his companies, especially at a high level, you would know that they made those achievements despite his directions. The SpaceX team was ecstatic when he bought Twitter, because it meant he would be around a lot less.

The policy at SpaceX was basically avoid Elon, and if you had to talk to him, just agree to whatever he said and then not do it. He almost always forgot by the next time he talked to you, and if he didn't, you just said it was really hard.

Since he doesn't understand rocket engineering, it was pretty easy to get away with that.

And the people who worked with him at PayPal would tell you that he knows nothing about software engineering and they basically threw away everything he wrote, but they had to be nice because he had a lot of stock after his company was acquired by PayPal but called a merger.

Elon's main skill is marketing himself. He next best skill is marketing his companies' jobs to talented people and getting them to agree to work there.

That is certainly a critical skill for a CEO, but he is not a great engineer.

They must have fixed it because I drive my two main monitors from a dock connected to one port and then my 3rd monitor from the other port. Been set up this way for at least two years now.

Yes but if your debate is desktop or laptop, you're probably not looking at anything but a pro.

My M1 Pro drives 4 screens no problem. The trick is that you can only attach two through one Thunderbolt port, so you have to attach the 3rd one via the other port instead of your dock. (The 4th screen is the laptop's built in screen)

There used to be a class at Berkeley that was cross listed between the CS department and the business school. When you were in the class, the CS kids and B-School kids were teamed up, and the B-School kids would write the business plan and product description, and the CS kids would make the prototype.

Then they would pitch in front of investors with both. Sometimes the ideas got funded and the kids would actually keep working on them.

> what's going to stop the average lazy student from ignoring whatever harness the university recommends and instead use an unrestricted LLM, thus learning nothing?

Nothing until they start failing exams and maybe seeing the error of their ways. My suggestions wasn't for you, it was for the smart student who wants to use AI to enhance their learning but not have it do all the work.

But cheating isn't new. People have been cheating in school for centuries. It's easier now, but the consequences were always the same. At some point the chickens come home to roost and you pay the piper.

Claude assumes humans won't look at code anymore. The comments are for the next Claude that comes in to change something.

When I was in college, we had office hours with the prof and the TAs, we had group study sessions, and private tutors. Or you could ask your friends. All of those required going somewhere at a specific place and time and asking someone else to give up their time for you.

AI gives you a way to get the same help without asking another human. Say of that what you will, but not everyone was comfortable asking other humans for help even back then.

"I need help with my homework, but I don't want answers, just guidance on how to get to the answer".

Put that in your agents.md.

I have to track daily data for an illness. What motivates me to stay on top of it are the insights gained from the tracking. Being able to go to the doctor and answer questions definitively instead of anecdotally.

We've actually changed the therapy procedures based on the data and then been able to show it was effective with the data that came after.

So think about the insights the tracking will provide you and your doctor, and that might help motivate you.

We use old school paper and pencil. I have a clipboard with a tracking sheet created in excel with rows for each thing that needs tracking and columns for the day. Then I don't have to think, I just fill in boxes, and the pencil and paper are always right there on the wall in the bathroom, where the data is "generated" :)

This also lets me scan quickly across for daily and weekly trends. I have room at the bottom for longer notes if necessary. I put a [1] in the box and then a note that matches.

I’ve read plenty of literature. So has the AI training. That won’t help. In fact, it will make it worse.

I even studied writing in college. I’ve written published books and articles. I can write well. But thanks for the tip!

Finally a use for copy/pasting AI!

1668 — Old ’76 House — Tappan, New York

1673 — White Horse Tavern — Newport, Rhode Island

1681 — King George II Inn — Bristol, Pennsylvania

1742 — Black Horse Tavern & Pub — Mendham, New Jersey

1747 — Reynolds Tavern and 1747 Pub — Annapolis, Maryland

1753 — The Pirate’s House — Savannah, Georgia

1776 — The Griswold Inn — Essex, Connecticut

1779 — Old Talbott Tavern — Bardstown, Kentucky

1779 — The Tavern — Abingdon, Virginia

1780 — Warren Tavern — Charlestown, Massachusetts

1789 — The Inn at Hancock — Hancock, New Hampshire

1790 — Ye Olde Tavern — Manchester, Vermont

1806 — Old Absinthe House — New Orleans, Louisiana

1827 — Ye Olde Trail Tavern — Yellow Springs, Ohio

1831 — New Hudson Inn — New Hudson, Michigan

1835 — Knickerbocker Saloon — Lafayette, Indiana

1842 — O’Malley’s — Weston, Missouri

1847 — The Village Tavern — Long Grove, Illinois

1850 — Landmark 1850 — Milwaukee, Wisconsin

1852 — Golden Gate Saloon — Grass Valley, California

1852 — Breitbach’s Country Dining — Balltown, Iowa

1853 — The Genoa Bar and Saloon — Genoa, Nevada

1857 — Hays House Restaurant and Tavern — Council Grove, Kansas

1859 — The Buffalo Rose — Golden, Colorado

1860s — Buckhorn Saloon and Opera House — Las Vegas, New Mexico

1862 — Miners and Stockmen’s Steakhouse — Hartville, Wyoming

1863 — Bale of Hay Saloon — Virginia City, Montana

1864 — Kelly’s Logan House — Wilmington, Delaware

1866 — Scholz Garten — Austin, Texas

1876 — Glur’s Tavern — Columbus, Nebraska

1877 — The Palace Restaurant and Saloon — Prescott, Arizona

1877 — Buffalo Bodega — Deadwood, South Dakota

1879 — Shooting Star Saloon — Huntsville, Utah

1883 — Pioneer Saloon — Paisley, Oregon

1887 — Neumann’s — North St. Paul, Minnesota

1887 — Bluebird Inn — Bickleton, Washington

1891 — Imperial Bar — Juneau, Alaska

1896 — Eischen’s Bar — Okarche, Oklahoma

1896 — Springwater Supper Club & Lounge — Nashville, Tennessee

1899 — North End Tavern and Brewery — Parkersburg, West Virginia

1903 — The Palace Saloon — Fernandina Beach, Florida

1905 — The Ohio Club — Hot Springs, Arkansas

1906 — Ford’s Bar — Idaho Falls, Idaho

1932 — Antlers Bar — Blowing Rock, North Carolina

1933 — Henry’s on the Market — Charleston, South Carolina

1935 — Town Hall Bar — West Fargo, North Dakota

1936 — Forest Gardens — Portland, Maine

1946 — Callaghan’s Irish Social Club — Mobile, Alabama

1948 — Blue Front Cafe — Bentonia, Mississippi

1957 — La Mariana Sailing Club — Honolulu, Hawaii

Source: https://www.businessinsider.com/oldest-bar-every-state

Sure, that's one of those obvious tells. But when I sit and spend 10 minutes making a well thought out and researched comment, and then properly format it in markdown, and then the first response is "this is AI", it's really frustrating.

Why would a bar in Alaska close during prohibition when they weren't part of the USA yet?

Every time this comes up I have to ask, how do you know it is AI? Sure, sometimes there are strong tells, like if they forget to take off the "Do you want me to change anything" at the end.

But AI was trained on good writing. Good writing looks like AI. I've had to reduce my use of em dashes for example because they are an "AI tell". I used to like using the term "load bearing".

And the list gets longer every day. Eventually every mark of a good writer will be an AI tell.

I don't know how to resolve this. But I've been accused multiple times of copy/pasting AI when I did no such thing, and it's getting equally frustrating.

> The big thing about monorepos in this case is making a change atomic across a set of projects.

The thing microservices is supposed to solve is eliminating the need to make atomic changes across services. There is an inherent overhead in doing that, like making your service support both new and old APIs as long as any service exists using the old API. This is the cost of microservices.

But a monorepo wouldn't make any of that easier.

You have to make the change to service A and B, and then test both and deploy both. You haven't saved any time or effort in a monorepo.

A library needs updating, you still have to update it and then test and deploy every service that relies on it independently. Again you haven't saved any time or effort. In fact you've made it worse, because if those services are maintained by different people, you just forced them to test and deploy on your timeline and priority, not theirs. You actually made the coordination problem worse.

> The tedium of making branches / commits / pull requests across 6 repos to land a feature is less problematic to an agent.

You're probably not using your microservices correctly if you need to change more than one service at the same time. The whole point of microservices is independently developing and deploying the services. Sweeping changes like that should be done in pieces, one service at a time.

Which is why microservices are best for larger organizations, because it reduces coordination between dev groups.

While you are technically correct (the best kind of correct), I would challenge you to find an organization that is doing the monorepo/microservice combo correctly. And what I mean by that is with microservices, the API is the only contract. Every service should be deployable independently.

If you have a monorepo, you're almost certainly violating that somewhere, using a shared library, or a shared database, or even just blocking deployment because some other service has turned the repo red.

They have wok adaptors for induction now.

They have wok adaptors for induction now, FYI.