TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Code Review as a Service

319 点作者 dennisy超过 3 年前

63 条评论

alkonaut超过 3 年前
I&#x27;m all for this if the person reviewing my code will know the context, history and all the details and conversations we had as a team. But in order for that to work, I&#x27;d probably be taking most of this reviewer&#x27;s time. And obviously in order for them to get up to speed with our practices, conventions, architecture, code style and whatnot, they&#x27;d probably need to start by doing a whole lot of development on our project first. At least several months. They could probably not do anyone elses code review then. So we&#x27;d have to pay them... one full time salary to be this coder-and-reviewer type person. I wonder what this service should be called.<p>I think I might be onto something here. &quot;Full time developers as a service&quot;
评论 #29626215 未加载
评论 #29625262 未加载
评论 #29625188 未加载
评论 #29625518 未加载
评论 #29628030 未加载
评论 #29624838 未加载
评论 #29628097 未加载
评论 #29625152 未加载
cannonpalms超过 3 年前
This entirely misses the point of code review.<p>Say it with me: Code review is a knowledge transfer exercise.<p>Finding bugs, security vulnerabilities, and keeping the code maintainable are merely side effects that we appreciate along the way.<p>The primary purpose of code review is increasing the bus factor of the given piece of code and facilitating organic knowledge transfer. That&#x27;s it.
评论 #29624875 未加载
评论 #29624888 未加载
评论 #29627438 未加载
评论 #29624723 未加载
评论 #29625255 未加载
评论 #29625667 未加载
评论 #29625445 未加载
评论 #29631390 未加载
评论 #29625694 未加载
评论 #29627997 未加载
评论 #29630261 未加载
评论 #29625901 未加载
评论 #29626439 未加载
评论 #29627299 未加载
评论 #29624726 未加载
评论 #29624736 未加载
评论 #29629804 未加载
评论 #29624758 未加载
hankchinaski超过 3 年前
Having on-demand engineers look at code, without broader context on the project it is in my view the same as something that can be automated either or both via static analysis and custom ci&#x2F;cd workflow checks. This can probably makes sense on a project with more junior engineers where many basic improvements can be supposedly suggested without needing broader context? I would be keen on hearing the use case
评论 #29625824 未加载
评论 #29624685 未加载
评论 #29624587 未加载
评论 #29624428 未加载
dirtbag__dad超过 3 年前
I used this service at my previous employer to launch a small django&#x2F;nextjs app and it was mostly fantastic.<p>Background is that I worked at an VC-backed startup as a dev after doing General Assembly’s full stack bootcamp. Left that job to do ops&#x2F;growth, and ~2.5 years later volunteered to build the web app when my company put the project on their roadmap.<p>As the only developer at the company, pullrequest was great for: - a general gut check on how I was doing - recommendations on how to better write js&#x2F;python. linters help but nice to have a person offer feedback on more advanced ways of doing things - sourcing documentation on best practices. I found a lot of typescript&#x2F;JavaScript resources to be inconsistent&#x2F;confusing. Was great for someone to find and vet guides for me. - help with bugs&#x2F;errors - basic library choices and architecture decisions<p>It was also fantastic to have several people reviewing my code at once. Gave me a perspective on the type of engineering manager I’d want to work for. Some folks focused more on technical details but struggled explain their changes in plain English, while others seemed to be the other way around.<p>pullrequest was not great for: - doing things fast. They didn’t have a real-time messaging feature so I’d get hung up on waiting for feedback. They do have a 24(?) hour turnaround, but when several folks are commenting on the same PR it gets hard to track what changes _really_ matter vs what they are throwing out there as a nice to have. - Anything that required context outside of the files committed. Though with some extra long PR comments I could manage.<p>I would 100% recommend pullrequest for small teams who are heavy on more junior devs or migrating to a new stack. It is an inexpensive way to ramp up learning.<p>Last thing here — when I worked at the startup my code was rarely reviewed, and I’d have to actively ask for it. Not all companies follow best practices (even if they have the resources to). I would’ve loved this at my past job, too.
aneutron超过 3 年前
I will admit that I thought this was a joke on the &quot;SaaS&quot; everything trend.<p>I can see how this works for fairly limited web applications for example, but as soon as the application grows in complexity and interacts within a bigger system of systems, I am doubtful that it would be logistically possible to outsource the code review (legally, knowledge transfer wise, and a plethora of other angles I&#x27;m a tad bit lazy to consider).<p>Overall, why not, if it&#x27;s priced correctly, then it&#x27;s probably a set of additional eyes for small projects. But for anything medium or higher, yeah, I don&#x27;t see this realistically working.<p>Maybe OP (?) can explain if I&#x27;m wrong (very likely). There&#x27;s probably something I&#x27;m missing.
评论 #29624530 未加载
评论 #29624105 未加载
Rhodee超过 3 年前
:wave: Pull Request reviewer here with over 1000 reviews.<p>I&#x27;ve reviewed code across languages, team size and maturity. A good static analysis tool is not code review. The word &#x27;context&#x27; came up 37 times in this thread (by the time I hit submit) and it is worth digging in to how I build and maintain context with teams. First up, it is my responsibility to uphold, not define a team&#x27;s best current practices. If you believe a linter and static analysis tools are best current practices, you probably do not need code review. The code review process is an exercise to affirm how well a pull request meets the expectations (context) of a team and suggest remediation where appropriate.<p>I assume &#x27;context&#x27; used in the threads to mean &quot;how we do things here&quot;. As a reviewer, I conduct review understanding the problems the team wants its code reviews to optimize for or to avoid. This is due to the people creating the platform. It&#x27;s a solid platform for reviewers to get things done. Every line of contributed code I review is done with an eye towards ensuring it affirms a team&#x27;s stated objectives. If those objectives somehow falls outside of what I know to be true from my experiences as a professional and best current practices (BCP), I am empowered to engage the team.<p>I&#x27;ve had teams request to never provide guidance for coding style issues. Others want to know if how they modeled a React component tree could be improved. My success on the platform relies on always building context. Without that rapport it limits the depth of the review. Because code reviews are interactive, they tend to get better over time. When things go well, the relationship between team and reviewer is seeking a pareto optimum between the pull request and the team&#x27;s best current practices. When needs change I adapt my reviews. It is the same treatment if you&#x27;re a one person shop, SME or a listed company.<p>k thx bye
评论 #29632701 未加载
andrewmwatson超过 3 年前
I&#x27;ve been a reviewer on PR for a couple of months now and it&#x27;s been amazing. I&#x27;ve gotten to help dozens, maybe hundreds of developers improved their code, fix security, readability, testing and lots of other issues. Sometimes the PR is short and the review is short but sometimes it&#x27;s much more involved and I&#x27;ve had the opportunity to really explain things and educate people in a way that no lint checker could. I&#x27;m able to apply the wealth of my experience and knowledge with specific guidance based on things I&#x27;ve learned from building software for 22 years. I can&#x27;t think of another way that a company could gain access to that kind of guidance for only $699 a month.<p>I can go into detail explaining how to safely and reliably structure something and the feedback I get from the developers is positive and appreciative. I also collaborate with other reviewers about things we&#x27;re seeing and I&#x27;ve learned a ton in the process because the other reviewers they&#x27;ve all got such a depth of experience and knowledge in different areas.<p>I don&#x27;t do reviews full time, I have a day job, but I&#x27;ve been earning about $1000 a week doing it just some nights and weekends and that has been really great too.<p>Overall, I think PullRequest is an enormously positive thing for its customers and the reviewers. We&#x27;re not trying to replace people&#x27;s existing review processes if they have them. Our goal is just to add to our customers&#x27; capabilities and I think we&#x27;re going very well at that.
评论 #29626279 未加载
niros_valtos超过 3 年前
Saw this link here in the past. Didn’t pickup. The reason why I don’t like it is that random people, regardless their expertise, cannot just review PRs and understand the impact of the change on the system without being deeply involved in the product.
评论 #29623985 未加载
评论 #29624641 未加载
nkmnz超过 3 年前
I&#x27;m currently the only developer in my startup and I crave for high quality feedback, so this service sounds like it&#x27;s a gift sent from heaven – especially for the $699 price tag per month. I guess they probably make a mixed calculation: take a small company with 10 devs. Two of them are part-time, on any given week one of them is on holidays&#x2F;sick leave, one is very junior and doesn&#x27;t contribute much yet, another one is more concerned with project management than with writing code and one has in fact transitioned into management but remains on the dev team because it&#x27;s part of their identity to still &quot;get their hands dirty&quot; (even though they don&#x27;t have the time to produce much code) – so you might end up with sth like 4-7x more code than I write, but 10x the payments. My productivity is not above average, but I have very little overhead besides coding. Unfortunately, I cannot find any information about a minimum team size for the PRO plan. I think I&#x27;ll give them a try.
GrahamCampbell超过 3 年前
I have performed reviews on pullrequest.com, and lack of context tends not to be an issue. Code review is an interactive process, where questions can be posed to the PR authors, and PR reviewers have access to search the codebase. Customer success at pullrequest.com also provide context for the repo and organization working practices, attached at the top of PR descriptions, to help reviewers with context and to know what kinds of feedback are useful. Sast and Dast are still not sophisticated enough or widely used enough to be good enough. In-house senior and principal engineers often do not have the time to be on top of every repo and every PR in their organization, and need to outsource an extra pair of eyes. It is highly common that I am reviewing codebases that do not have any Sast or Dast, or have workarounds to silence the tools, resulting in bugs and security issues slipping through. It is very rare that I am able to approve PRs without comments, even after an internal review has taken place!
评论 #29631795 未加载
评论 #29625936 未加载
评论 #29625533 未加载
评论 #29655844 未加载
deeveloper超过 3 年前
I&#x27;m a reviewer and a firm believer in PullRequest. I do believe that some of the concerns expressed here are valid but I think they belie a fundamental misunderstanding of &quot;what you&#x27;re getting&quot; from PR. Having an under-experienced contractor review your code without context, pointing out things that static analysis could&#x2F;would catch sounds like a terrible proposition, but that is NOT what PR provides.<p>Speaking from personal experience, the reviewers are very knowledgable professionals who are not doing this for money as the primary motivation. The pay is great but is more of a perk&#x2F;reward for doing something we enjoy, and doing a great job at it. The staff at PR does an amazing job of coaching and guiding reviewers on how to provide the most value to the client and also encouraging us to do our best work. It is never anything like &quot;You need to be faster and bill less, do more, etc&quot; it is quite the opposite, we are encouraged to &quot;keep the clock going&quot; while we research, gain context, etc.<p>Everything is focused on providing value to the client, and IMHO a stellar job is done. Every review comes with detailed information on who the client is, the make up of their team (seniority, etc), recommendations on what to look for (and what NOT to look for) etc. Essentially you are putting your code in front of people who have been there, done that for a long time and are acutely aware of not only risk but also pain points and how to avoid tech debt. We are providing insight, recommendations and even code snippets on how to avoid repetition, speed things up, or make them easier to maintain.
评论 #29655874 未加载
Kocrachon超过 3 年前
I understand NDAs are a thing, but I still don&#x27;t think I&#x27;m too comfortable with the idea of letting a bunch of third-party people look at a bunch of my internal information. And I understand this is targeting startups a lot more than large established companies. But to me that&#x27;s even more concerning because as a startup you&#x27;re really trying to move fast and hoping someone doesn&#x27;t beat you to the punch, and you&#x27;re handing a bunch of people you don&#x27;t know how much of your secret internal information, and hope that they don&#x27;t go talking to other people about it.
评论 #29631762 未加载
评论 #29628898 未加载
ssully超过 3 年前
I see a lot of comments regarding lacking context&#x2F;knowledge of the full system&#x2F;integration. I get that, but I think there is real value here. I have absolutely worked on teams where they were fire fighting for months at a time, where there were 1-2 developers total, and just not enough time or resources to do proper code reviews. While this service could miss big picture things, it could absolutely catch low-mid tier issues.<p>My biggest issue would be from a security perspective. The background checks and everything are nice, but there are some systems I would just never let this service touch.<p>Overall, this is interesting! Wish the team behind it the best of luck.
mytailorisrich超过 3 年前
This implies that you have no-one in your team to do code reviews and that you&#x27;re fine allowing random third parties access to your infrastructure and a peek into your projects and code base... Both the premise and the proposed solution sound very odd to me.
评论 #29625894 未加载
swalsh超过 3 年前
I&#x27;m not sure this is a good idea for enterprise startups building closed-sourced software. As a software architect, some things you just don&#x27;t delegate. Code Reiews are some of the most important things I do. But (and please don&#x27;t downvote me for this, I know there&#x27;s an instant reaction to downvote anything blockchain related) for DAO&#x27;s this would be amazing.<p>A DAO (Decentralized Autonomous Organization) might be running a software as a service. But it might not have any full time employees. It might not have any employees at all. A lot of updates to the software might come from random people (or bots). The DAO will need to evaluate and pay for any of those updates before it decides to merge the pull request. A code review as a service would be an absolutely invaluable tool for a DAO with a software product that doesn&#x27;t have any full time architects to perform the service.
评论 #29625410 未加载
schnika超过 3 年前
I wanted to sign up as a reviewer but it seems like one needs a linkedin account to do so. Since I don&#x27;t have a linkedin account it seems like there is no other chance?
评论 #29626555 未加载
softwaredoug超过 3 年前
On the one hand, lacking broader context might make this service seem silly.<p>However I think it can help bust groupthink<p>It’s still valuable to get more general feedback and ideas explicitly without context to challenge our attachment to current practices in existing codebase schools.<p>It’s nice to get a pair of eyes with a completely different background give feedback. With context we may have blinders on. A fresh perspective might uncover things we haven’t thought of and bust groupthink<p>Even though I’ve been coding in Python for years, I still might not have awareness of the best most effective way to do something in general. Imagine all the projects started in the last few years from people learning Rust for the first time?<p>It may not make sense for the largest, most complex code bases, but I can see it valuable for medium to small projects to get outside perspective.
pelasaco超过 3 年前
I&#x27;m imagining that to make such service profitable, offering something like $699 per developer per month, you are not hiring reviewers from USA, right?
评论 #29624609 未加载
评论 #29624623 未加载
评论 #29625561 未加载
评论 #29625843 未加载
评论 #29624648 未加载
speby超过 3 年前
Gosh, building any tools that target developers as customers is a gargantuan task. Kudos to you for attempting something new here and giving it a whirl. We developers are some of the harshest critics around, no matter how legit or unfounded our criticism may be. Keep it up!
cptaj超过 3 年前
I&#x27;ve never seen outside consultants make useful contributions to a project. This is even worse.<p>Thanks, I hate it.
throwaway984393超过 3 年前
I really love this idea. It&#x27;s like static code analysis but with more specific feedback based on context. This could dramatically improve code written by junior and middle level engineers. I&#x27;d use it!<p>The only stumbling block seems to be that a lot of devs seem to be resistant to it. I&#x27;m not sure why that is; there&#x27;s many forms of code review and feedback, from shallow to very deep. And I&#x27;ve seen many teams that fail to do proper code review. This could be an excellent introduction to proper practice for immature teams.
dennisy超过 3 年前
Interested to know if anyone has used this service or similar and what the quality is like.
评论 #29623830 未加载
soheil超过 3 年前
I lot of people are saying CR is a way to transfer knowledge within a team. I think they&#x27;re missing the point. Sure, you can transfer knowledge within a PR but don&#x27;t you think that&#x27;s a bit late? By the time CR is requested, the code has been written, the architecture has been thought of and so much time has been wasted doing the &quot;wrong&quot; thing. It&#x27;s a horrible way to learn&#x2F;train.
jph超过 3 年前
Good code reviews are superb for helping teams accelerate into new technology areas, frameworks, languages, and integrations.<p>As a top of mind example, when a team wants to spike on migrating from C++ to Go or Rust, including using libraries and porting services, then I see very high value in paying for skilled contractors to do code reviews-- because what your team is gaining on-demand upskilling.
seneca超过 3 年前
This post has a number of what look like astroturf advertising comments.
评论 #29631860 未加载
评论 #29626686 未加载
nathanielarmer超过 3 年前
I&#x27;m a reviewer on PullRequest, and thought I&#x27;d share some perspective. Happy to answer any questions in comments.<p>Background - I&#x27;ve built and led engineering teams at multiple fast-growing startups. In doing so I&#x27;ve seen the incredible value PR&#x27;s can provide, but also the huge cost of them on small teams. I review on PullRequest part-time as I work full-time building a startup.<p>Many of the critics here are right. The PullRequest service won&#x27;t catch every bug. As reviewers on this service, we lack context* to fully understand the impact of every code change.<p>However, this can be a blessing in disguise. As an outsider, I bring an entirely different set of context to the project. I can see errors or improvements that teams have become blind to, I don&#x27;t have the pressure of shipping for X release, and I&#x27;ve often been not only where these teams are, but where these teams want to be in X months.<p>This is all done without using any man-hours on the client team - which is often a critically short resource.<p>Ultimately the proof is in the pudding, I raise comments on almost every single review I do, raising from best practice to architectural to security vulnerability, and the majority of the time teams take that feedback onboard.<p>Other QA:<p>Where are you located? I&#x27;m located in San Francisco.<p>Who approves PR&#x27;s? I tend to consider my role to advise and sometimes mentor. I will give opinions, but it ultimately the client&#x27;s job to approve&#x2F;reject PR&#x27;s.<p>My code is great, I don’t need this! Have you tried it? As an outsider, I catch peoples blind-spots. That said, PR is always looking for great reviewers to join the team!<p>* It&#x27;s worth noting, that between seeing the code, optionally having access to the full code base, and asking questions of developers - I develop a decent mental model of most projects.
评论 #29627862 未加载
评论 #29655784 未加载
nyanpasu64超过 3 年前
I noticed the example image: &quot;It looks like this method needs to use the read lock because it&#x27;s accessing the `dataKeys` map. I would recommend documenting which methods are and are not meant to be safe for concurrent access.&quot; To me, this is a problem addressed by Rust&#x27;s &amp;&#x2F;&amp;mut separation, which can be imitated in other languages (eg. in multithreaded code, wrap types in Mutex&lt;T&gt;&#x2F;MutexGuard&lt;T&gt; or RwLock&lt;T&gt; with read guards lending out `const T&amp;` and write guards lending out `T&amp;`, if you&#x27;re in a language without const, tough luck). This creates code which self-documents concurrency in the type system like Rust, though it&#x27;s less watertight since you can hold onto references for too long.
dawnerd超过 3 年前
I had to flag this after seeing several obvious spam replies trying to boost the service. Come on now.
评论 #29626732 未加载
napolux超过 3 年前
LOL. This will become in less than 1 month a &quot;LGTM&quot; feast. 3rd world countries developers (organizing themselves in groups to pass the intro test from the website) giving green lights to random people around the world for peanuts. This is a clever business model if you ask me.<p>Let&#x27;s be clear for a second. Nobody, not event the legendary 10x developers can review properly code without context or everything mentioned here: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=29624787" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=29624787</a><p>If we want our code to be approved by random people doing it only for money, this will work for sure, but it&#x27;s not how it works in the real world.
评论 #29625516 未加载
评论 #29629433 未加载
评论 #29625576 未加载
vemv超过 3 年前
I&#x27;d have some amount of healthy skepticism over obvious concerns (most prominently, keeping IP safe) but it also it seems worth a shot; it solves a real problem because often for teams code reviews are a bottleneck <i>and</i> bit of a battlefield.
gorgoiler超过 3 年前
The best code reviews are ones that leverage experience from the reviewer to stop logic errors. Your code may well compile and the tests you concocted may pass but that doesn’t mean your code should ship. Without a second pair of experienced eyes looking at your crap, how are you to know that:<p>- <i>(general experience with modelling)</i> what you felt should be a single class would be better if it were factored into two classes; and<p>- <i>(specific experience with this particular codebase)</i> the second of those classes already exists in the codebase and can be found in com.clown.util?<p>The former can be helpful but it’s just noise compared to the value of the latter.<p>How does this service avoid being just a human powered style linter?
8organicbits超过 3 年前
I noticed this requires LinkedIn for signup, that&#x27;s too bad. I deleted my account years ago once it became a major source of spam. I think this is the first place I&#x27;m seeing it required. Strikes me as an odd requirement.
siliconc0w超过 3 年前
It&#x27;s a bit crazy to me you&#x27;d outsource this to third party. Code reviews are the single best way spread context on a team so more than one person understands (and consents) to the change that is being made. Giving this up to &#x27;free up engineers&#x27; for reviews seems like short term thinking. On the other hand, using this to screen hires using projects instead of inane leet code interviews seems like a good idea - especially if they worked with multiple large companies so it&#x27;s many to one.
grenran超过 3 年前
There&#x27;s no way that Google would actually use this service.
评论 #29624131 未加载
adamqureshi超过 3 年前
Im a 1 MAN SHOP and i hire contractors ( engineers). I am not sure if this a good fit for me? My main concern is of course cost. Do you provide feedback on what todo with the software stack? ( can my guys implement a new plan provided by you?)and 2nd question is do you provide pay for service, for example if i am interested in a new software architecture and my guys just code it up &#x2F; follow your plan &#x2F; recommendation. Thx
loloquwowndueo超过 3 年前
“LGTM” as a service :)
game_the0ry超过 3 年前
How is this not the same as contracting a freelance dev? You might as well contract through wipro &#x2F; infosys &#x2F; tata &#x2F; cognizant &#x2F; hcl.<p>I am annoyed at how replacing software engineers is somehow viewed as a business problem that can be solved as a SaaS business or no code tooling.<p>PR&#x27;s customers need to hire more developers and pay them well. It is expensive and hard, but that is the market business people need to accept.
joshgrib超过 3 年前
Can someone give a solid and succinct review of doing this as a side-job? What was 1) the hourly pay, 2) language(s) you reviewed, and 3) how was the work?
评论 #29633622 未加载
nn3超过 3 年前
Why would anyone ever apply to a place that has such a graph on its website?<p><a href="https:&#x2F;&#x2F;www.pullrequest.com&#x2F;images&#x2F;figures&#x2F;reviewers&#x2F;screening-graph.png" rel="nofollow">https:&#x2F;&#x2F;www.pullrequest.com&#x2F;images&#x2F;figures&#x2F;reviewers&#x2F;screeni...</a><p>From a customer perspective, and if it&#x27;s really true they probably will have to lower their standards to serve you too.
srhaegrfes超过 3 年前
Putting aside IP and feasibility concerns, what is the longterm roadmap? This is obviously not a VC scalable product business if it acts as a broker to (albeit expensive) human consultants.<p>Are they hoping to get enough training data from the consulting practice to bootstrap an AI code review product?
skeeter2020超过 3 年前
This reminds of the first (and only) time I had a full 360 degree review: contradictory, zero-context opinions from people I don&#x27;t know very well who haven&#x27;t earned the right to interact with me in a way required for this minefield of feedback.
walterbell超过 3 年前
<i>&gt; expert engineers backed by best-in-class automation.</i><p>&quot;Automation&quot; = static analysis or something like <a href="https:&#x2F;&#x2F;codeql.github.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;codeql.github.com&#x2F;</a>?
lmilcin超过 3 年前
Except I don&#x27;t believe in code reviews. Over the years I have experimented with pair programming and I think it is way better solution.<p>The one catch with pair programming is that some people just prefer to work alone and it is really draining to be talking constantly for couple of hours. I have modified the system to do on&#x2F;off pair programming (ie. split up a little bit of work, rejoin later in the day, share what we have done, and work a little bit together).<p>The basic, unsolvable issue with code reviews is that the review is done <i>after</i> the code has already been written. As you know, the cost of fixing a problem is larger the later in the process you catch it. Pair programming aims to accomplish the correction while the code is being written.<p>Another huge problem is that, because the reviewer is not taking part in the development, he&#x2F;she does not have the same level of understanding of what was supposed to be done.<p>Also, code reviewers are typically disincentivized from doing review well:<p>* They have other tasks to accomplish, review takes their time away from those tasks but the deadlines are not pushed automatically,<p>* The review tends to land at a random point in time disrupting their flow -- they have something else in mind already and they want to switch to their work as quickly as possible -- meaning they will not want to get into great detail with understanding the problem.<p>This causes reviews to usually be very shallow and focused on trivia. Usually, I see reviewers read the code file by file line by line, hundred times faster than it was written. It is absolutely impossible to verify a large change like that and this guarantees that they will not actually verify it thoroughly.<p>Yeah, you may find superficial flaws, but that&#x27;s about it.<p>Other problems:<p>* The developer feels resentment because he&#x2F;she thought it was all done.<p>* A lot of effort was spent on a wrong solution which is lost productivity.<p>* From project management PoV it is a problem because we can&#x27;t tell how much time&#x2F;effort it will take until last second.<p>* Reviewers feel pressure to find <i>something</i> so they will just report trivia if they can&#x27;t find real issues.<p>* Reviewers tend to not want to report huge issues that would require complete rewrite because they are developers themselves and wouldn&#x27;t want the same happen to them, and also because they are typically members of the same team.<p>* and so on.
评论 #29625924 未加载
antonpuz超过 3 年前
I&#x27;ve greatly improved as a developer by reviewing code, this improved my project and improved team alignment, that&#x27;s much more valuable than getting inputs from human linter.
ralusek超过 3 年前
A service like this would only be able to find a certain class of issue. Basically syntax, but not semantics. There is no substitute for deep knowledge of a particular codebase.
评论 #29625880 未加载
rpunkfu超过 3 年前
I wish person (people) involved in making this would do some more research, it&#x27;s completely missing a point of PR review.<p>This is to me a set of linters with human error involved..
gandutraveler超过 3 年前
On top of what others have already flagged this is a big no for any companies security &amp; compliance. Why would any company share their private codebase ?
评论 #29625747 未加载
imglorp超过 3 年前
I hope my boss doesn&#x27;t think, &quot;It&#x27;s cheaper to pay for this service than for us to review our own PRs. Outsource ++!&quot;
评论 #29625536 未加载
booleandilemma超过 3 年前
And I’m guessing their reviewers work for free, like on Code Review Stack Exchange?<p>Or are these people working for relative poverty wages overseas?
评论 #29625335 未加载
评论 #29624933 未加载
lambic超过 3 年前
The only time I can see this being useful is on solo developer projects, and solo developers probably can&#x27;t afford it.
jcadam超过 3 年前
I received an invite to pullrequest nearly a year ago, filled out the application, and never heard from them again.
b1nj0y超过 3 年前
I think this service only applies to those open source projects, or personal learning project.
smalter超过 3 年前
i&#x27;ve used pullrequest a bunch and had great experiences with the product and company.<p>mostly used it in small companies with 1-2 devs where it helps to get another pair of eyes on it. also, helpful as career development &#x2F; learning for a junior dev.<p>happy to answer any questions.
z3t4超过 3 年前
&gt; Unlimited lines.<p>I wonder how long it will take them to go though my entire code base
alexashka超过 3 年前
What we need is software architecture review, not git diff review.
est超过 3 年前
I hope it can review popular code on github.com automatically.
评论 #29625809 未加载
akhil-ghatiki超过 3 年前
How does this play with the Non disclosure agreements ?
评论 #29626807 未加载
cheriot超过 3 年前
This smells like lead gen for a consulting shop.
whoomp12342超过 3 年前
very often code review requires domain knowledge. This seems like it would devolve into arguing over coding preferences
racl101超过 3 年前
Human linter?
natded超过 3 年前
Teaching sand to think was a mistake.
privacyonsec超过 3 年前
Isn&#x27;t paper peer reviewing broken ? Want to do the same thing with software?
gitdiff超过 3 年前
As a PullRequest reviewer, I think many on this thread are missing the forest for the trees.<p>First, I think many here may be viewing Code Review as a Service from their frame of reference: from unicorn start ups, FAANG tech companies, prestigious universities, etc. Many of the companies that we help aren&#x27;t coming from this world. They&#x27;re small startups, looking for technical guidance. They&#x27;re entrepreneurs who need to ensure they&#x27;re not being duped by app developers. They&#x27;re older, less tech-savvy companies that are looking to modernize. These companies need help, and they can get help from people who have technical expertise. (And by the way, some companies aren&#x27;t sophisticated enough to set up linters or code scanners at their stage of development)<p>In a similar vein, many of the developers we help may not have had the same level of education, learning, or coaching as you or I may have. For example, I&#x27;ve helped introduce more modern syntax options to developers, such as string interpolation and extension methods in C#, to Options and Streams in Java, to filter&#x2F;map&#x2F;reduce functional patterns and optional chaining in Javascript. These developers may have never seen high quality code or had mentors who insisted on a high bar for code quality.<p>Even for more sophisticated customers, I&#x27;ve left comments ranging from security vulnerabilities (e.g. SQL injection), errors in boolean logic, recommendations for improved test coverage, recommendations for simplifying code (e.g. creating reusable functions), preventing race conditions, and more. I&#x27;ve also reviewed candidate assessments to help unburden senior engineers so they can focus on writing code.<p>Sometimes, the proof is in the pudding. There&#x27;s a market for these services and that&#x27;s why some companies pay for them and why I get to review code on demand. I periodically get feedback from the teams I help that I&#x27;ve done &#x27;Nice Work&#x27;, receiving positive ratings from the developers I review for. I&#x27;m proud that I can lend my expertise to make code better.
评论 #29632367 未加载