My team tried coderabbit and qodo and they are both trash compared to a tool we quickly built in-house that is more or less a thin wrapper around claude/codex, along with per-repo skills. PR review is triggered by webhooks from github to the review tool's web app. The tool shared by OP from alibaba certainly does some things ours does not and appears more sophisticated, but we have never had the problems they mention.
"The agent can read full file contents, search the codebase, inspect other changed files for context, and produce deep reviews — not just surface-level diff feedback." our tool does all this too. It catches dumb typos as well as more complicated bugs. Not to mention it is great as a ratchet (https://qntm.org/ratchet). It is not a substitute for reviews from other engineers though, since obviously it does nothing to achieve one of the main goals of code review, which is to socialize knowledge of the codebase.
Alibaba's work here is almost certainly more advanced than what we've done, but ours has been perfectly satisfactory and better than the paid offerings we've tried. I think most teams should not be paying SaaS fees for AI code review, that is the kind of business that mostly should not exist any more.
This project argues that with appropriate harness, the performance gap between frontier and much smaller open weight models shrinks dramatically: https://github.com/antoinezambelli/forge. I haven't kicked the tires yet.
Do you happen to have any links to more info about this schedule system. If I Google “sears schedule system” I get a bunch of results about vaccines, and if I append “logistics” the top result is your comment.
Claude Design was iterating on the plans page and decided to remove clutter and their review bot LGTM’d it as “minor copy change human review not required” and auto-merged it.
I don’t have much experience with pony but it seems like it addresses the core concerns in this article by design https://www.ponylang.io/discover/why-pony/. I wish it were more popular.
Evidence is often contradictory, especially in the social sciences--that is not a terribly damning charge in this case. Additionally, there is evidence that relationship between social media use and anxiety/depression is not just an association, see Meta's own internal research from 2019: https://metasinternalresearch.org/#block-2e15def2e67a803a83e....
"Meta’s own researchers found — in an experiment they believed was better designed than any external study done thus far — that reducing time on their platforms improved mental health and well-being, specifically depression, anxiety, loneliness, and social comparison."
The ad is hilariously bad but McDonald’s has done many terrible ads over the years where “creatives” were involved eg the infamous random red couch ad.
I would like this to be true, but am a bit skeptical.
I am what the article calls an "industrial software engineer" and I work on "low- to medium-assurance" projects, but have used various formal methods (alloy and TLA+) in my work to prevent and discover bugs.
I've experimented with using LLMs to generate both Alloy and TLA+ a couple times over the past years, and the problems I see are:
- They have gotten better over the last few years, but still can only produce useful results in the hands of someone who is moderately competent. Becoming moderately competent requires many hours of investment in these tools, and you will lose much of this competence if you don't keep it up. For example, I can still read TLA+ and Pluscal but can't write them without lots of referring to the docs because I only write them like once or twice a year.
- They suffer even more from GIGO than other aspects of software development. If you can't really rigorously define your problem you will get a bad model/output that only gives you false confidence. A large part of the value of doing formal methods is building the muscle for thinking rigorously. Hillel Wayne says this in several places, that doing enough TLA+ (e.g.) work gives you a much better innate sense for where there will be race conditions.
- There will still be a cultural and technical problems with integrating formal methods, and their artifacts, into the rest of your codebase and team. For example, how do you prevent drift? Will you have a CI automation that uses an LLM to detect when the spec has diverged from the code?
I'm not saying it is impossible that this will happen, and I would love to be wrong, but the general tendency I see with LLM use is to make software developers less intimately familiar with their tools, and less invested in deeply understanding their code. That bodes ill for formal methods even more than regular programming.
Roll back is not always the right answer. I can’t speak to its appropriateness in this particular situation of course, but sometimes “roll forward” is the better solution.
Here are the article's main points, as I see them:
1. The "modern American self" is best defined by (the tension between) Franklin and Rousseau.
2. Rousseau believes X and Franklin believes Y.
3. "Modern America" (society? politics? government?) flip flops between these two, though they are "almost entirely incompatible".
4. The author claims one of them scales, and says he likes it.
I engage directly with claims 2 and 3.
I think 1 is another completely absurd simplification. I do not address it, or claim 4. I don't see how that constitutes lack of engagement or quibbling. Perhaps I could have written an essay refuting OP with many citations, but I don't think that level of work is required to constitute legitimate engagement.
I guess you're probably right that my comment is more shame than content, maybe 60/40 shame to content, I should have dialed that down a bit. Fwiw I think it's fine to be simple-minded and ignorant, I am both of those things about many topics, but then your writing and argumentation should reflect your lack of knowledge and certainty. OP's article is, otoh, full of hot air.
I don't know much about Franklin, but this strikes me as a gross oversimplification of Rousseau, to the point where I wonder whether the author has actually read much Rousseau, rather than just other lightweight "thinky pieces" on Rousseau. For example The Social Contract is significantly concerned with how people can and will act in accordance with the general will.
Also the idea that these philosophies are "almost entirely incompatible" reveals the author's complete ignorance of one of the most important influences in Western philosophy, Aristotle, for whom concordance of action and "intention" (arguably not an ancient Greek concept, but close enough for an hn comment) must be united in ethically good action.
But if your goal is not actually to understand anything and merely to sound smart on a causal reading, and perhaps try to get people to "not think so damn much and just do stuff" I guess this piece achieves its goal.
"The agent can read full file contents, search the codebase, inspect other changed files for context, and produce deep reviews — not just surface-level diff feedback." our tool does all this too. It catches dumb typos as well as more complicated bugs. Not to mention it is great as a ratchet (https://qntm.org/ratchet). It is not a substitute for reviews from other engineers though, since obviously it does nothing to achieve one of the main goals of code review, which is to socialize knowledge of the codebase.
Alibaba's work here is almost certainly more advanced than what we've done, but ours has been perfectly satisfactory and better than the paid offerings we've tried. I think most teams should not be paying SaaS fees for AI code review, that is the kind of business that mostly should not exist any more.