omarhaneef

At the heart of this article is the claim that buying equity is a form of theft.

That is an extreme claim (in the sense of surprising, remarkable, unusual, and one that needs a lot more support than ordinary claims).

It is inadequately defended here. The argument that it violates fair exchange is tautological.

Try the “best sellers” or new arrivals: https://anycrap.shop/best

There are a few dozen of these out there. That’s the first question anyone should answer.

I know you’re trying to see where the puck is headed but I think there is a lot of work between where we are and where LLMs replace 4 Junior devs.

The workflows we have are not quite right for it. Coding has always been 10% coding and 90% debugging but I think the rate at which we generate the 10% will grow exponentially.

This means that the debugging has to grow. We will generate errors at an unprecedented rate.

LLMs trained on previous errors and methods won’t catch them. They’ll be more complicated and spread out over the code.

We need new tools to visualize the code and track errors. I think what it means to be a programmer will change. More testing, thinking and less klocs.

If you found even a single job off of it, doesn’t that mean it’s great? This is its main purpose.

Perhaps it could be better, but a lot of people would pay a lot for a single good job offer. And you accepted it.

For every application of semantic search, I’d love to see what the benefit is over text search. If there a benchmark to see if it improves the search. Subjectively, did you find it surfaced new papers? Is this more useful in certain domains?

I think this is probably right but so far it seems that the areas in which an LLM is most effective do fine with the lower power models.

Example: the 4o or Claude are great for coding, summarizing and rewriting emails. So which domains require a slightly better model?

I suppose if the error rate in code or summary goes down even 10%, it might be worth $180/month.

One of the best pieces on here recently (and didn’t realize it was by Sinofsky!)

Anyway, one other insight I would add is that the issues tend to come up at the interface between the systems. Your automation has to get some input and send some output and both those are pain points. That’s why we sometimes prefer imperfect monolith software.

That’s (an) awk word.

All these suggestions are great but I would add two more:

1. find a good text book in the field as narrowly construed as possible, and just browse the table of contents.

This will give you a quick map of the field, and a place to put all the info you gain.

2. Read the management discussion in the 10k of a public version of the company you work at and 2 leading customers, and the transcript from the earnings call.

Fastest way to get up to date on challenges.

If you read between the lines, it would seem no one was keeping the place clean.

I think the exciting part is this:

“BNT211 continues to show encouraging antitumor activity in patients with CLDN6-positive relapsed or refractory advanced solid tumors” <- interesting because they finally demo some effect in solid tumors in humans

“Follow-up of efficacy data at 1x108 CAR-T cells with or without CARVac shows an overall response rate (“ORR”) of 59% and a disease control rate (“DCR”) of 95%, with the CARVac cohort demonstrating a prolonged persistence of CAR-T cells” <- interesting because those are high response rate. Let’s hope it lasts and they can increase those rates but as a patient, you would take this.

This should not be a debate.

Take a note from the pet shop boys: they never argue.

They also never calculate the currency they spend.

Ugh. Any clue as to why?

I think question should include the size of the dev team.

Solo dev vs 50 devs might need different tools just to get out of each others way.

Yes, good point and I like your final partial solution. I think it’s a good compromise.

Microsoft Authenticator syncs across all instances so you just have to log in to a new Authenticator on the new phone and the codes are there. (I think it uses OneDrive).

I am sure that is less secure than a local only copy but this may be least bad of all alternatives.

You might even give a trusted person a login and have it on their phone so you can use theirs in an emergency.

Can’t it just scan the client? For email at least, I feel it can just read your inbox or whatever and store the results of the analysis locally.

You were very active with this friend for n years and it dropped off. Drop them a line.

There have been several iterations of this sort of idea — a personal CRM — and the main issue I have is they should automatically scan my emails, texts and chats to figure out my friends and suggest that I reach out.

Also, want it to be 100% local and privacy first.

You may have heard people say that “business” is not a deep discipline and I think the theory of business is pretty accessible.

Go to any business school site and just look at the MBA curriculum:

https://www.hbs.edu/mba/academic-experience/curriculum/Pages...

I would then focus on the introductory courses (typically Accounting, Valuation, Marketing, Org theory etc).

Find the syllabus for those courses or the text book and just browse it.

That won’t teach it all but it will sort of give you a map of the territory so you can look it up when you need to. About the level of memory most MBA grads have 5 years after graduation.

Firstly, I am sorry to read that.

Secondly, if they removed all visible signs, that is great and very hopeful.

If my loved one had a serious medical condition, and there wasn’t a known cure — for instance if it comes back — I would look at the government database that lists current trials in the U.S. (assuming you live in the U.S) and then I would research each potential treatment in turn.

You don't know if other humans do, but you know at least one human that does: yourself (presumably).

Same.

1. Ask chatgpt to write the code

2. Debug the obvious errors

3. Ask stack overflow to find the non-obvious errors

4. Repeat

You accidentally become a better programmer like the old days when people would learn by typing in listings from compute.

N+1 theory on why this may be the case:

Cognitive ability is forged in reading and discussions. After you’ve seen the arguments rehearsed hundreds of times you’ve become inundated and exhausted by the arguments and counter arguments from both sides of many debates.

If only something like this were part of markdown.

(And here on HN, I hope someone is going to say: yes, this vscode extension will add it to markdown)

It is one thing to pursue something out of curiosity, and another to spend hours upon hours every day for weeks on the same thing.

I think your curiosity might give way to other obligations at some point which is why you have to be somewhat goal oriented.

Worth noting that I have spent hours pursuing a hobby as a kid: programming. And I think that is probably true of many of the readers here.

I think the computer lends itself to this in a weird -- perhaps unique -- manner. The act of programming, or writing, is like a video game in that you get this reward just for the act. And it is easier to spend hours for days for weeks programming (or writing) than say mulling over other matters.

One trick for turning a task you cannot spend time on into a task you can spend time on is to turn it into an activity. This is why mathematicians have to write out their equations and diagrams when they're mulling plates.

I have to disagree.

The quality of decline has really gone downhill in the last few years. Used to be that when things fell apart, they really fell apart (think revolution, collapse) but now they just peter out.

He writes whatever he wants to write about.

And he would argue, he can't control what he wants to write about.

What makes you want to make him not want to write what he wants to write about?

Well, unfortunately, you can't control it either!

At this point I am basically an unpaid shill for this company but:

Udacity

While courses from Coursera and EdX are also excellent, Udacity consistently makes me work hard and actually expend the energy to understand non-trivial tech.

There are three insights that David Allen had with GTD that I appreciate:

- one inbox to absorb tasks you distribute later

- lists go into working contexts so you do them while in that context

- short term and long term are different

The issue I have is that we also have different “recording” contexts. These days I can use an app on my phone, a note in my computer or a notepad in a meeting. I have multiple inboxes that have to be routed.

In addition I think the relationship between project and tasks needs to be fleshed out.

However I think those 3 insights persist across all good productivity systems.