Today, the difficult shit is how to get the resources from the people who use them to turn difficult shit into an easy shit. Basically, a war. I am not sure I want people to move on to that.
People who understand the vast gulf of technological advancements (some bordering on fantasy) between what we currently have, and what needs to exist for it to happen?
There is a finite capacity/time for a human mind to do difficult shit, if it can be slop forked in a microsecond before you even get to flesh it out there is no point
An AI smart enough should act like a senior engineer gathering requirements, it should start with assumptions and poke at different areas with questions until it has a complete idea, when I talk with a client I dont expect him/her to really formalize all the details its my role to question them until all the sharp corners are covered
Yes, but also, who do you gather requirements from? Other people. But if we're talking AGI, then these other people, i.e. users - or at least those who define the requirements - could be replaced, too.
>My app has 200k LOC + 100k in tests, and there is no sign of the AI slowing down or being unable to maintain it.
May I ask what does this app do ? 200k is like GTA vice city level, is like sqlite level, If its a financial tracker or a to-do list or an agent coordinator I am going to kill myself
I'm not going to share exact details about my paid app, but you could think media management, with a lot of extra features, an in-app browser with custom controls, and multiple AI powered features with on device processing.
He's not. That's how things will be from now on. Because there's plenty of hacks who think prompting will make them millions if they could just find that one idea, yet aren't willing to do the work to find, vet, build, and iterate.
It doesn't matter when you post it; some kid with a $200 claude sub (or cheaper) is going to clone it for far cheaper than you, and sell it for far cheaper than you.
There's no friction to what you produced, so why do you think that you won't have 100s of clones, all cheaper?
We're now in a market for lemons as far as software goes.
I spent hundreds of hours on this, so there is still some friction.
If I told you the exact market and use case, I'd still have a decent head start, but indeed, someone without a job could invest more time than I do. People in a cheaper location can work for far cheaper than I can afford to.
With head start I don't meant that the idea is so novel that I need to keep it a secret. However, I believe that if there is an established app that does the same thing, and your app cannot beat it on quality, it will be hard to attract users.
A conundrum I see is that once someone releases a “solution” to a problem, it becomes much easier for someone else to come up with prompts to have AI make a clone of that solution.
Basically, the moat for many products is disappearing, and it would be relatively cheap for someone to make a clone that undercuts the cost of the first to market. The original product is charging $2? Cool, the clone will cost $1 - maybe it’s not as good due to less QA, but consumers will see the lower price and buy it.
> I spent hundreds of hours on this, so there is still some friction.
I'm afraid your hundreds of hours of prompting will reduce to "clone this product, here are 200 screenshots of every screen."
I don't understand why people aren't getting this: there is no skill in prompting! The CC leak literally showed Anthropic engineers developing Claude with broken half-english prompts like "gud, now make better".
Hundreds of hours doing what? If you're not reviewing the code, then what are you doing, just planning features? Because that would be trivial to copy in a fraction of the time by just feeding screenshots of your app to one's AI of choice.
That's the thing. The moment you publish, the design you spent 100s of hours on also becomes public, at which point it becomes entirely replicable with little to no recourse. Hence the theme of this entire thread: a devaluement spiral.
Despair is a choice some people have made, and they want to drag others into it to justify their inaction. Take your best shot. Maybe it takes off, maybe it doesn't.
Either way, it's a hell of a lot better than being a replyguy.
Vice City was released nearly 25 years ago! Also it's a game, so it's not really an apples-to-apples comparison. If you count the unique 'features' (or systems) of Vice City, you might find it's a comparable count to the number of features of a medium-sized mobile app.
You are joking right ? renderer, audio, math, physics, animation, AI, asset streaming,.... in addition to the gameplay itself, and you compare it to a medium-sized mobile app ???
Apps usually have tons of boilerplate, so pure LoC may be similar. But in density, no contest. I recently dug up an old objc app of mine and was shocked how few lines it took to send a REST request, parse and process it. Nowadays it is split into multiple files, usecases, repositories, protocols to accomplish the same. All in the name of testability or architectural “correctness”. Write that in a game and you’re fired because it kills performance.
There's a different bar for maintenance though. Many games exist in a single point in time, they ship and are done, whereas the goal with SaaS is to support the code forever.
Maintenance, sure, but also: games have one chance to be good, and that is at release. With a multi-year development cycle, what we get at release (of the good ones) is like a version 4 or 5 of an app. Everything is connected and changed (and thus maintained) over the years.
It's not like a simple embedding search takes that much longer to implement. Especially on short descriptions where you don't have to deal with chunking. And if you let an LLM write the code it's even less of a difference. Combine that with embedding search promising to solve all your search problems, and I understand why people often skip over full text search and go straight to embeddings
Even that is an oversimplification unless you are doing something very basic.
Volume of documents, size of documents, versioning, frequency of update, documents similar or overlapping information, how much or exactly what you need for the LLM to understand, AI friendly documents, who has access and at what level, blue teaming, red teaming, multi-lingual, does the LLM know the domain language of the user and documents.
I think Bitwarden implemented some vector search in their password search feature ... totally annoying it gives me back all kinds of stuff that I don't care.
I want fuzzy search like 95% of time and then I might consider having additional list of things that can be suggested by vector search.
Bandcamp has had legendarily bad semantic search for as long as they've been around. It's often completely impossible to find an artist or album or song even when you type the exact name.
ahh now I realize why I get so much completely irrelevant search results in many sites recently. I mean I'm searching for betel and you're giving me nuts. haha
A good UI could do these and also exact match, give some point system to the results, then order them and perhaps use a bold highlight to reflect what parts of the input query reflected in each result.
Bitwarden has lost the plot. The most recent Windows update is so bad. It has way lower information density in the UI, more buttons to click for the same use, no longer puts focus on the search field by default (this one makes me irrationally angry), and on one of my Win 11 installs can't lock the vault, manually or automatically. How could they mess up such a simple app that worked fine for so long?! What perverse incentives caused this nonsense?!
Yeah I cant imagine my job without it anymore, in my hobby projects I dont really use it because its more about learning new stuff but for work I dont really care
That's so depressing, and a real loss for your employer. Even if they neither know nor care that you are now sandbagging them, you still are. You aren't making real gains because you're tossing out value of equivalent or greater worth (your learning and engagement)
Most places I worked at cared very little for quality code, there was much more pressure on shipping fast even if it meant incurring technical debt, crap performance for end users, or developers leaving due to legacy code accumulation that nobody understood anymore.
I'm glad there is a tool that let's these companies have even shittier code shipped even faster. The faster they burn down the better.
You can only learn and engage so much in a day. In my head, it can be healthy to shift from tiring yourself out via work to tiring yourself out at home. You're not sandbagging them, you're setting boundaries on how much you're willing to do. If this is a problem for the employer, they can act accordingly.
I'm amazed you think typical corporations care about staff learning and engagement in the first place.
There's a reason they're trying to replace humans with AI: they just want outcomes. People learning and growing is only useful if it generates better outcomes. If you can get the same thing with a bunch of reverse centaurs, from a corporate perspective that's better: labour becomes more fungible and you can spend even less on training and investment in staff (assuming you ever did). And do it right and those folks will just train your machines to make them even better (see: Meta capturing worker activity and reallocating large numbers of staff to data labelling).
Sure, but that's a big big big big big big big big big big "if".
All AI has really succeeding in proving so far is that humans CAN really, very much "keep up", despite repeated repeated repeated repeated assertions that in six months it will not be so
Oh I'm aware. I'm not saying it makes any g-d sense. I'm just saying that's the corporate logic. So for those of us stuck in AI pilled orgs, the best bet is to play the game and keep your skills sharp on the side in case the whole thing goes pear shaped.
But it isn't a loss for any employers. They're not being sandbagged in any way. It's the path they've chosen, knowingly and willingly.
I'm with you except that the economy will crash when it turns out everyone stopped creating real value. ...and in the mean time every real product I have to use is getting worse.
Where they created value wasn't just the CRUD app but developing their skills, supporting communities that asked and answered questions to build public knowledgebases. They adapted to painfully repetitive tasks by developing tools that made expressing common ideas simple. In losing the pain of repetition, we lost the incentive towards abstraction. The forces in opposite directions cancelled out, creating growth for AI companies but overall stagnation due to the immense amount of value lost because before AI there was never a dollar value put on it.
Its fun but is it really effective ? I mean I checked the LLM one and I came out more confused about a topic I already know about, I find the best way to to learn using LLMs is to just generate an example try to somewhat get a mental model of how it works and then ground my understanding with traditional documentation and resources, its an iteration of a technique I used to do in college where I would read the textbook questions first to understand what is important and then read the chapter
I assume different ways of learning work for different people. For me personally, it's taking a piece of paper and drawing the diagram of how things work together; of if it's some math, then, again, using the pen and paper to follow the text. I can very much accept that for some people playing the simulation is a good way to touch the new problem space. I can easily imagine that for some topics, let's say, traffic signal automation, a careful simulation game will probably give more information than reading papers or manuals.
Game-based learning, described by Comenius, works if someone else prepares “a game” for you. E.g. like a dungeon master. :)
Otherwise you probably get more confused as you have mentioned.
On the other side, Peter Diamandis describes a situation where a bunch of kids were given a internet-connected computer and they had no teacher. Instead of it there was a “grandma” that checked kids from time to time.
After that there was a knowledge test that revealed “no teacher” approach was more efficient.
a- understood it and all the background information it requires
b- internalized techniques and methods that are helpful in problem solving in general
Now it just means nothing
reply