LLMs bring new nature of abstraction – up and sideways
54 points by tudorizer 1 week ago | 57 comments- oytis 1 week agoI don't get his argument, and if it wasn't Martin Fowler I would just dismiss it. He admits himself that it's not an abstraction over previous activity as it was with HLLs, but rather a new activity altogether - that is prompting LLMs for non-deterministic outputs.
Even if we assume there is value in it, why should it replace (even if in part) the previous activity of reliably making computers do exactly what we want?
- kookamamie 1 week agoFunny, I dismiss the opinion based on the author in question.
- Insanity 1 week agoSerious question - why? I know of the author but don’t see a reason to value his opinion on this topic more or less because of this.
(Attaching too much value to the person instead of the argument is more of an ‘argument from authority’)
- kookamamie 1 week agoLet's just say I think a lot of damage was caused by their OOP evangelism back in the day.
- kookamamie 1 week ago
- Insanity 1 week ago
- dist-epoch 1 week agoBecause unreliably solving a harder problem with LLMs is much more valuable than reliably solving an easier problem without.
- furyofantares 1 week agoI'm pretty deep into these things and have never had them solve a harder problem than I can solve. They just solve problems I can solve much, much faster.
Maybe that does add up to solving harder higher level real world problems (business problems) from a practical standpoint, perhaps that's what you mean rather than technical problems.
Or maybe you're referring to producing software which utilizes LLMs, rather than using LLMs to program software (which is what I think the blog post is about, but we should certainly discuss both.)
- dist-epoch 1 week ago> solve a harder problem than I can solve
If you've never done web-dev, and want to create an web-app, where does that fall? In principle you could learn web-dev in 1 week/month, so technically you could do it.
> maybe you're referring to producing software which utilizes LLMs
but yes, this is what I meant, outsourcing "business logic" to an LLM instead of trying to express it in code.
- dist-epoch 1 week ago
- oytis 1 week agoOK, so we are having two classes of problems here - ones worth solving unreliably, and ones that are better solved without LLMs. Doesn't sound like a next level of abstraction to me
- pydry 1 week agoThe story of programming is not largely one of humans striving to be more reliable when programming but putting up better defenses against our own inherent unreliabilities.
When I watch juniors struggle they seem to think that it's because they dont think hard enough whereas it's usually because they didnt build enough infrastructure that would prevent them from needing to think too hard.
As it happens, when it comes to programming, LLM unreliabilities seem to align quite closely with ours so the same guardrails that protect against human programmers' tendencies to fuck up (mostly tests and types) work pretty well for LLMs too.
- dist-epoch 1 week agoI was thinking more along this line: you can solve unreliably 100% of the problem with LLMs, or solve reliably only 80% of the problem.
So you trade reliability to get to that extra 20% of hard cases.
- pydry 1 week ago
- 1 week ago
- darkwater 1 week agoWhich harder problems are LLMs going to (unreliably) solve in your opinion?
- dist-epoch 1 week agoAnything which requires "common sense".
A contrived example: there are only 100 MB of disk space left, but 1 GB of logs to write. LLM discards 900 MB of logs and keeps only the most important lines.
Sure, you can nitpick this example, but it's the kind of edge case handling that LLMs can "do something resonable" that before required hard coding and special casing.
- dist-epoch 1 week ago
- furyofantares 1 week ago
- kookamamie 1 week ago
- smokel 1 week agoAbstractions for high-level programming languages have always gone in multiple directions (or dimensions if you will). Operations in higher level languages abstract over multiple simpler operations in other languages, but they also allow for abstraction over human concepts, by introducing variable names for example. Variable names are irrelevant to a computer, but highly relevant to humans.
Languages are created to support both computers as well as humans. And to most humans, abstractions such as those presented by, say, Hibernate annotations, are as non-deterministic as can be. To the computer it is all the same, but that is increasingly becoming less relevant, given that software is growing and has to be maintained by humans.
So, yes, LLMs are interesting, but not necessarily that much of a game-changer when compared to the mess we are already in.
- bgwalter 1 week agoHow many bandwagons has this guy jumped on? Now he says that LLMs will be the new high level programming languages but also that he listens to colleagues and hasn't really tried them yet.
I suppose he is aiming for a new book and speaker fees from the LLM industrial complex.
- somewhereoutth 1 week ago> As we learn to use LLMs in our work, we have to figure out how to live with this non-determinism. This change is dramatic, and rather excites me. I'm sure I'll be sad at some things we'll lose, but there will also things we'll gain that few of us understand yet. This evolution in non-determinism is unprecedented in the history of our profession.
The whole point of computers is that they were deterministic, such that any effective method can be automated - leaving humans to do the non-deterministic (and hopefully more fun) stuff.
Why do we want to break this up-to-now hugely successful symbiosis?
- nmaley 1 week agoI'm in the process of actually building LLM based apps at the moment, and Martin Fowler's comments are on the money. The fact is seemingly insignificant changes to prompts can yield dramatically different outcomes, and the odd new outcomes have all these unpredictable downstream impacts. After working with deterministic systems most of my career it requires a different mindset.
It's also a huge barrier to adoption by mainstream businesses, which are used to working to unambiguous business rules. If it's tricky for us developers it's even more frustrating to end users. Very often they end up just saying, f* it, this is too hard.
I also use LLM's to write code and for that they are a huge productivity boon. Just remember to test! But I'm noticing that use of LLM's in mainstream business applications lags the hype quite a bit. They are touted as panaceas, but like any IT technology they are tricky to implement. People always underestimate the effort necessary to get a real return, even with deterministic apps. With indeterministic apps it's an even bigger problem.
- CraigJPerry 1 week agoSome failure modes can be annoying to test for. For example, if you exceed the model’s context window, nothing will happen in terms of errors or exceptions but the observable performance on the task will tank.
Counting tokens is the only reliable defence i found to this.
- danielbln 1 week agoIf you exceed the context window the remote LLM endpoint will throw you an error which you probably want to catch, or rather you want to catch that before it happens and deal with it. Either way, it's not a silent error that goes unnoticed usually, what makes you think that?
- diggan 1 week ago> If you exceed the context window the remote LLM endpoint will throw you an error which you probably want to catch
Not every endpoint works the same way, I'm pretty sure LM Studio's OpenAI-compatible endpoints will silently (from the clients perspective) truncate the context, rather than throw an error. It's up to the client to make sure the context fits in those cases.
OpenAI's own endpoints do show an error and refuses if you exceed the context length though. I think I've seen others use the "finish_reason" attribute too to signal the context length was exceeded, rather than setting an error status code on the response.
Overall, even "OpenAI-compatible" endpoints often aren't 100% faithful reproductions of the OpenAI endpoints, sadly.
- CraigJPerry 1 week agoInteresting, the completion return object is documented but theres no error or exception field. In practice the only errors ive seen so far have been on the HTTP transport layer.
It would make sense to me for the chat context to raise an exception. Maybe i should read the docs further…
- diggan 1 week ago
- danielbln 1 week ago
- CraigJPerry 1 week ago
- gdubs 1 week agoThe Star Trek computer – I grew up with TNG – is the example I always think of. They ask it in human language to pull up some data, create a visualization, run an analysis. You get the sense that people also still write programs in the future in a more manual way – but, done well these LLMs are the building blocks for that more conversational way of getting the computer to do what you want it to do.
A lot of the complaints that come up on Hacker News are around the idea that a piece of code needs to be elegantly crafted "Just so" for a particular purpose. An efficient algorithm, a perfectly correct program. (Which, sorry but – have you seen most of the software in the world?)
And that's all well and good – I like the craft too. I'm proud of some very elegant code I've written.
But, the writing is on the wall – this is another turning point in computing similar to the personal computer. People scoffed at that too. "Why would regular people want a computer? Their programs will be awful!"
- felineflock 1 week agoIt is a new nature of abstraction, not a new level.
UP: It lets us state intent in plain language, specs, or examples. We can ask the model to invent code, tests, docs, diagrams—tasks that previously needed human translation from intention to syntax.
BUT SIDEWAYS: Generation is a probability distribution over tokens. Outputs vary with sampling temperature, seed, context length, and even with identical prompts.
- dcminter 1 week agoSurely given an identical prompt with a clean context and the same seed the outputs will not vary?
- furyofantares 1 week agoYou can make these things deterministic for sure, and so you could also store prompts plus model details instead of code if you really wanted to. Lots of reasons this would be a very very poor choice but you could do it.
I don't think that's how you should think about these things being non-deterministic though.
Let's call that technical determinism, and then introduce a separate concept, practical determinism.
What I'm calling practical determinism is your ability as the author to predict (determine) the results. Two different prompts that mean the same thing to me will give different results, and my ability to reason about the results from changes to my prompt is fuzzy. I can have a rough idea, I can gain skill in this area, but I can't gain anything like the same precision as I have reasoning about the results of code I author.
- diggan 1 week ago+ temperature=0.0 would be needed for reproducible outputs. And even with that, if it's actually reproducible or not depends on the model/weights themselves, not all of them are even when all those things are static. And then finally depends on the implementation of the model architecture as well.
I think the tricky part is that we tend to think that prompts with similar semantic meaning will give the same outputs (like a human), while LLMs can give vastly different outputs if you have one spelling mistake for example, or used "!" instead of "?", the effect varies greatly per model.
- dcminter 1 week agoHmm, I'm barely even a dabbler, but I'd assumed that the seed in question drove the (pseudo)randomness inherent in "temperature" - if not, what seed(s) do they use and why could one not set that/those too?
To your second part I wouldn't make that assumption - I can see how a non-technical person might, but surely programmers wouldn't? I've certainly produced very different output from that which I intended in boring old C with a mis-placed semi-colon after all!
- smokel 1 week ago> I think the tricky part is that we tend to think that prompts with similar semantic meaning will give the same outputs (like a human)
Trust me, this response would have been totally different if I were in a different mood.
- dcminter 1 week ago
- furyofantares 1 week ago
- genidoi 1 week agoThis is too abstract and a concrete example of what this looks like in output is needed.
- dcminter 1 week ago
- stpedgwdgfhgdd 1 week agoI’m programming software development workflows for Claude in plain english (custom commands). The non-deterministic is indeed a (tiny!) bit of a problem, but just tell Claude to improve the command so next time it won’t make the same mistake. One time it added an implementation section to the command. Pretty cool
This is the big game changer: we have a programming environment where the program can improve itself. That is something Fortran couldn’t do.
- stpedgwdgfhgdd 1 week ago(bit off topic, but I wished someone told me this a few months ago; for regular programming generation, use TDD and a type safe language like Go. (Fast compile times and excellent testing support). Don’t aim for the magical huge waterfall prompt)
- stpedgwdgfhgdd 1 week ago
- alganet 1 week ago> This evolution in non-determinism is unprecedented in the history of our profession.
Not actually true. Fuzzing and mutation testing have been here for a while.
- diggan 1 week agoI think the whole context of the article is "program with non-deterministic tools", while non-deterministic fuzzing and mutation testing is kind of isolated to "coming up with test cases", not something you constantly program side-by-side with, or even integrate into the (business-side) of the software project itself. That's how I've used fuzzing and mutation testing in the past at least, maybe others use it differently.
Otherwise yeah, there are a bunch of non-deterministic technologies, processes and workflows missing, like what Machine Learning folks been doing for decades, which is also software and non-deterministic, but also off-topic from context of the article, as I read it.
- alganet 1 week agoI just have a problem with his use of the word "unprecedent".
This is not the first rodeo of our profession with non-determinism.
- alganet 1 week ago
- TZubiri 1 week agoRight, in testing, but not in the compiler chain
- alganet 1 week agoI don't understand what you mean. Can you elaborate on your perception of what a "compiler chain" is and the supposed LLM role in it?
- TZubiri 1 week agoA C compiler outputs x86 or ARM or whatever assembly. C is the source, x86 is the target code.
Javascript is source code that might be interpreted or might output html target code (by Dom manipulation)
Typescript compiles to javascript.
Now javascript is both source and target code. If you upload javascript code that was generated by ts to your repo and you leave out your ts, that's bad.
Similarly, an LLM has english (or any natural language) as it's source code and typescript (or whatever programming language) as its target code. You shouldn't upload your target code to your repo, and you shouldn't consider it source code.
It's interesting that the compiler in this case is non deterministic, but it doesn't change the fact that the prompts are source code, the vibecode is target code.
I have a repo that showcases this
- TZubiri 1 week ago
- alganet 1 week ago
- diggan 1 week ago
- dtagames 1 week agoI respect Martin Fowler greatly but those who, by their own admission, have not used current AI coding tools really don't have much to add regarding how they affect our work as developers.
I do hope he takes the time to get good with them!
- diggan 1 week ago> have not used current AI coding tools really don't have much to add regarding how they affect our work as developers
I dunno, sometimes it's helpful to learn about the perspectives of people who've watched something from afar as well, especially if they already have broad knowledge and context that is adjacent to the topic itself, and have lots of people around them deep in the trenches that they've discussed with.
A bit like historians still can provide valuable commentary on wars, even though they (probably) haven't participated in the wars themselves.
- TZubiri 1 week agoI agree, I don't use coding tools, "except to ask for a script to chatgpt every once in a while". But I experience it by reviewing and detecting LLM generated code by consultants and juniors. It's easy to ask them for the prompts for example, but when they use autocompletion based LLMs, it's really hard to distinguish source from target code.
- TZubiri 1 week ago
- diggan 1 week ago