A fly-by-night charlatan successfully pushed ticking into our organization in the past year and I would say it was a disaster. I only have the experience of one, but from that experience I am now not sure you can even build good software that way.
I originally hoped it was growing pains, but I see more and more fundamental flaws.
I’ve worked at one, but it required a PM who was ruthless about cutting scope and we focused on user stories after establishing a strong feedback pipeline, both technically through CI/CD/tests and with stakeholders. Looking back, that was the best team I’ve ever worked in. We split up to separate corners of the company once the project was delivered (12 month buildout of an alpha that was internally tested and then fleshed out).
Maybe I had greenfield glasses but I came in for the last 3 months and it was still humming.
Previously? There was an understanding of the problem trying to be solved. The gaps left the pangs of "this isn't right".
Now I have no way to know where things stand. It's all disconnected and abstracted. The ticket may suggest that something is done, but if the customer isn't happy, it isn't actually. Worse, now we have people adding tickets without any intent to do the work themselves and there isn't a great way to determine if they're just making up random work, which is something that definitely happens sometimes, or if it truly reflects on what the customer needs.
You might say that isn't technically a problem with ticketing itself, and I would agree. The problems are really with what came with the ticketing. But what would you need tickets for other than to try and eliminate the customer from the picture? If you understand the problem alongside the customer, you know what needs to be done just as you know when you need to eat lunch. Do you create 'lunchtime' tickets for yourself? I've personally never found the need.
You must be working in projects with a relatively small number of “problems to be solved” at any given time, and with the problems having relatively low complexity. In general there’s no way to keep everything in your head and not organize and track things across the team. That doesn’t mean that a lot of communication doesn’t still have to happen within the team and with the customers. Tickets don’t replace communication. But you have to write down the results of the communication, and the progress on tasks and issues that may span weeks or months.
> In general there’s no way to keep everything in your head
I imagine everyone's capacity is different, but you wouldn't want anyone with a low capacity on your team, so that's moot. Frankly, there is no need to go beyond what you can keep in your head, unless your personal capacity is naturally limited I guess, because as soon as you progress in some way the world has changed and you have to reevaluate everything anyway, so there was no reason to worry about the stuff you can't focus on to begin with.
I find that the current way we do Scrum is way more waterfall-ish than what we had before. Managers just walked around and talked, and knew what each person was doing.
We traded properly working on problems for the Kafkaesque nightmare of modern development.
Thing is, Scrum isn't supposed to be something you do for long.
As you no doubt know, Agile is ultimately about eliminating managers from the picture, thinking that software is better developed when developers work with each other and the customer themselves without middlemen. Which, in hindsight, sounds a lot like my previous comment, funnily enough, although I didn't have Agile in mind when I wrote it.
Except in the real world, one day up and deciding no more managers on a whim would lead to chaos, so Scrum offered a "training wheels" method to facilitate the transition, defining practices that push developers into doing things they normally wouldn't have to do with a manager behind them. Once developers are comfortable and into a routine with the new normal Scrum intends for you to move away from it.
The problem: What manager wants to give up their job? So there has always been an ongoing battle to try and bastardize it such that the manager retains relevance. The good news, if you can call it that, is that we as a community have finally wisened up to it and now most pretty well recognize it for what it is instead of allowing misappropriation of the "Agile" label. The bad news is that, while we're getting better at naming it, we're not getting better at dealing with it.
I don’t think people invested in Scrum believe it’s “temporary” or ever marketed it as such.
And agile teams are supposed to be self-managed but there’s nothing saying there should be no engineering managers. It sounds counter intuitive, but agile is about autonomy and lack of micro-management, not lack of leadership.
If anything, the one thing those two things reject are “product managers” in lieu of “product owners”.
> I don’t think people invested in Scrum believe it’s “temporary” or ever marketed it as such.
It is officially marketed as such, but in the real world it is always the managers who introduce it into an organization to get ahead of the curve, allowing them to sour everyone on it before there is a natural movement to push managers out, so everyone's exposure to it is always in the bastardized form. Developers and reading the documentation don't exactly mix, so nobody ever goes back to read what it really says.
> And agile teams are supposed to be self-managed but there’s nothing saying there should be no engineering managers.
The Agile Manifesto is quite vague, I'll give you that, but the 12 Principles makes it quite clear that they were thinking about partnerships. Management, of any kind, is at odds with that. It does not explicitly say "no engineering managers", but having engineering managers would violate the spirit of it.
> not lack of leadership.
Leadership and management are not the same thing. The nature of social dynamic does mean that leadership will emerge, but that does not imply some kind of defined role. The leader is not necessarily even the same person from one day to the next.
But that is the problem. One even recognized by the 12 Principles. Which is that you have to hire motivated developers to make that work. Many, perhaps even most, developers are not motivated. This is what that misguided ticketing scheme we spoke of earlier is trying to solve for, thinking that you can get away with hiring only one or two motivated people if they shove tickets down all the other unmotivated developers' throats, keeping on them until they are complete.
It is an interesting theory, but one I maintain is fundamentally flawed.
I've realized it's a different paradigm in (very loosely) the Kuhn sense. You wouldn't track tasks if you're fundamentally not even thinking of the work in terms of tasks! (You might still want a bug tracker to track reported bugs, but it's a bug tracker, not a work tracker.)
What you actually do is going to depend on the kind of project you're working on and the people you're working with. But it mostly boils down to just talking to people. You can get a lot done even at scale just by talking to people.
I think the trick with Haskell is that you can write types in such a way that the APIs that get generated are idiomatic and designed well. The implementations of individual functions might be messy or awkward, but as long as those functions are relatively small—which is how I tend to write my non-AI-based Haskell code anyhow!—it's not nearly as important.
From what I've heard—and in my own very limited experiments—LLMs are much better at less popular languages than I would have expected. I've had good results with OCaml, and I've talked to people who've had good results with Haskell and even Unison.
I've also seen multiple startups that have had some pretty impressive performance with Lean and Rocq.
My current theory is that as long as the LLM has sufficiently good baseline performance in a language, the kind of scaffolding and tooling you can build around the pure code generation will have an outsize effect, and languages with expressive type systems have a pretty direct advantage there: types can constrain and give immediate feedback to your system, letting you iterate the LLM generation faster and at a higher level than you could otherwise.
I recently saw a paper[1] about using types to directly constrain LLM output. The paper used TypeScript, but it seems like the same approach would work well with other typed languages as well. Approaches like that make generating typed code with LLMs even more promising.
Abstract:
> Language models (LMs) can generate code but cannot guarantee its correctness often producing outputs that violate type safety, program invariants, or other semantic properties. Constrained decoding offers a solution by restricting generation to only produce programs that satisfy user-defined properties. However, existing methods are either limited to syntactic constraints or rely on brittle, ad hoc encodings of semantic properties over token sequences rather than program structure.
> We present ChopChop, the first programmable framework for constraining the output of LMs with respect to semantic properties. ChopChop introduces a principled way to construct constrained decoders based on analyzing the space of programs a prefix represents. It formulates this analysis as a realizability problem which is solved via coinduction, connecting token-level generation with structural reasoning over programs. We demonstrate ChopChop's generality by using it to enforce (1) equivalence to a reference program and (2) type safety. Across a range of models and tasks, ChopChop improves success rates while maintaining practical decoding latency.
As a huge proponent of constrained decoding for LLM reliability, I don't quite think it's the right approach for code. This is because in many programming languages, it is legal to use a function before its declaration. Since this is common in existing code, LLMs will also try to write code that way.
You might be right, but I think you must take into account that (I think) you're not super familiar with these languages as well, so you might not notice all the warts a programmer with a lot of experience in these langs would, and overrate the skill of the LLM.
Nowadays, I write C# and TS at work, and it's absolutely crazy how much better the LLM is at TS, with almost all code being decent the first try, but with C# I need to do a lot of massaging.
I write Haskell professionally and I can tell you that Opus 4.5 can do a great writing industrial Haskell code consistent with an existing code base.
I don't think it is capable of writing galaxy brain Haskell libraries, it absolutely missed the forest for the trees, but if you have an existing code base with consistent patterns it can emulate then it can do a surprisingly good job.
I built a library (without Claude) that wraps in an opinionated way Servant and a handful of other common libraries used to build Haskell web-apps and then have let Claude use that to build this site. There is absolutely some hairy code and I have done a ton of manual refactors on what Claude produces, but claude has been highly effective for me here.
GitHub has at least some support for org-mode rendering. Which is a good reminder that a lot of this stuff really doesn't have to be a stark either-or choice—a lot of standardization isn't inevitable or natural or necessary, but just a choice we happen to make, often for systemic reasons that aren't inherently justified (legibility/etc).
what I've done is avoid the sort of boss who would mandate AI use
in a past job I did tell a boss that I wasn't going to be doing the whole tickets/estimates/schedule tetris thing, and that actually worked out... because the leaders I worked with understood the value of being flexible and trusting their lead engineers
Different programming languages come with different schools of thought about programming and different communities of practice around programming.
If you take a group of people who are squarely in the enterprise Java school of thought and have them write Rust, the language won't make much of a difference. They will eventually be influenced by the broader Rust community and the Rust philosophy towards programming, but, unless they're already interested in changed approaches, this will be a small, gradual difference. So you'll end up with Enterprise Java™ code, just in Rust.
But if you hire from the Rust community, you will get people who have a fundamentally different set of practices and expectations around programming. They will not only have a stronger grasp of Rust and Rust idioms but will also have explicit knowledge based on Rust (eg Rust-flavored design patterns and programming techniques) and, crucially, tacit knowledge based on Rust (Rust-flavored ways of programming that don't break down into easy-to-explain rules). And, roughly speaking, the same is going to be true for whatever other language you substitute for "Rust".
(I say roughly because there doesn't have to be a 1:1 relationship between programming languages, schools of thought and communities of practice. A single language can have totally different communities—just compare web Python vs data scientist Python—and some communities/schools can span multiple languages. But, as an over-simplified model, seeing a language as a community is not the worst starting point.)
If you're doing something forgettable, what makes you think the workaday Java or Python programmer would find it innately motivating?
Alternately, if you have the sort of work or culture that taps into people's intrinsic motivation, why would that work worse with Haskell or Clojure programmers than anybody else?
People are interested in different things along different dimensions. The way somebody is motivated by what they're doing and the way somebody is motivated by how they're doing it really don't seem all that correlated to me.
The last six decades of commercial programming don't exactly bear this out...
The real lesson is that writing software is such a useful, high-leverage activity that even absolutely awful software can be immensely valuable. But that doesn't tell us that better software is useless, it just tells us it is not absolutely necessary.
Caring about craft in programming is more like a photographer caring about light and composition and creativity and taste than a photographer caring about equipment.
In some ways yes. Many “engineers” obsess over “idioms” and other trends to the detriment of performance, correctness and usability. So this analogy is a bit too charitable.
I'm not sure that's a valid analogy. Light, composition and creativity are all experienced directly by viewer, and essentially describe what it is that we notice and appreciate in photography (even if subconciously). The best analogy I can think of to programming is the UX/UI of the application. Given equaly competent developers, nobody is going to notice or care if your application was written in Rust or Cold Fusion.
But the original analogy is flawed too. I wouldn't consider caring about the craft of programming to be similar to obsessing over your photography equipment. GAS is about consumerism and playing with gadgets, at the end of the day.
Caring about the craft of programming is more about being an artist who takes pride in crafting something beautiful, even if they're the only ones experiencing it. I am most definitley not one of those programmers, but have always had nothing but immense respect for those that are.
It's amazing how such a short comment manages to betray a fundamental misunderstanding of stimulants, tinkering, human nature and, implicitly, neurodivergence.
reply