Put Yourself Out There: The Myth of the Genius Programmer | Hacker News Reader

Put Yourself Out There: The Myth of the Genius Programmer

joshldavis.com

145 pointsdavis78 comments

This meme of "do things and blog about it" presumes a certain amount of extroversion. I'm rather fed up with it altogether.

Would you rather be thought of as the best developer, or actually be the best developer? The two aren't mutually exclusive, but I find it entirely suspicious that the rise in needing to "put yourself out there" has correlated so strongly with social media becoming pervasive. If I have a finite amount of time available for hacking, why do I want to spend a not-insignificant amount of it signaling?

Hacker culture has been corrupted by this 'need' for influence. Perhaps it isn't the same as the neckbeard culture I looked forward to joining in my youth.

Can't I just spend an evening decoding a binary format? When I'm done, can't I just put it on a wikia without sitting around to make sure my name stays attached to it?

If you did that, how would people know to follow you on Twitter?

The meme used to be "do things and publish in journals about it" now it's blogging because that's easier. But it's always been important to put yourself out there if you wanted others to use your work.

I share your sentiment, and this quote (attributed to Margaret Thatcher) may be relevant:

"Being powerful is like being a lady. If you have to tell people you are, you aren't."

Have you considered blogging to be not about extroversion but instead just being a personal journal that is public?

I think the type of blog you are referring to desires to engage others, while a blog that is indexable simply there to add value for other who might be trying to figure their way through the same random things that you have been.

Yeah, this is the only form of blogging I can tolerate: recording thoughts over a period of time as an individual explores a set of topics. It takes on a decidedly different tone from the more presumptuous, link-baity "why X is the best language ever"-style of blogs. Namely, the latter has no speck of intellectual humility; it is merely a long-form tweet.

I agree with you completely on being bored of posts about "x" that are based on preference, opinion, interpretation that can't see there being more than one equally valid way of solving a problem.

What is interesting is when someone picks up a problem and simply solves it using the tools at hand, be it old, new, preferred, or trying out a new approach. I learn far more from the approaches than the results.

It's not a need to influence, it's to share and spread knowledge. That's one of, if not the, fundamental tenets on which hacker culture was built and depends on.

You do things, find something new or interesting, share it with the world so others can build on your work.

Thank you. This is exactly what it is.

Says the guy with the username "mattgreenrocks".

It's not about being the best programmer. After all, most of us are not the best - yet :) It's important not to discourage each other by pretending that only the best programmers matter. The whole point of the post is that it's sometimes important to "signal" that we're not perfect.

This is a much better comment than how it started out.

Intellectual humility is a good capsule summary of what I'm looking for (despite my username; I never claimed to be self-consistent). But intellectual humility can be at odds with the notion of always presenting the 'best' side of yourself to the market. I'm interested in an individuals steps and missteps, not a carefully-marketed image.

Yeah if it were Reddit I would have left it :)

The whole article is against self-legend-making and you seem to be in violent agreement.

intellectual humility is fake humility.

Seriously, humility is one of the most arrogant things I've ever come across. You do not ever "signal" and be humble at the same time, that isn't how people work.

The theme of all of this isn't just to signal. It's that putting yourself out there will cause you to bend and grow. Over and over I read testimonials of people with low confidence who decided to do something like this and found it incredibly fulfilling. Don't just do the things you've always done. Stretch yourself, and you might be amazed what you can accomplish. Blogging improves your writing ability at the very least, and can also cause you to think more critically about your ideas. If you happen to build some influence or good relationships, even better.

Obviously if you aren't interested in any of that, there is nothing wrong with spending the time the way you want to spend it. But it's not about signaling or influence, at least as I interpret it.

It's unfortunate that someone downvoted you, I've tried to negate that.

The only reason I made a comment to this effect is that I've noticed an uptick in articles about how Github is your resume/marketing yourself/marketing your OSS/how you should only work on 'important' OSS/some other silly notion. I want to believe this isn't a problem, but with open source powering so much, it appears that people want to be the next DHH.

I think there is something to worrying about software shifting to a heavily reputation-based economy (as opposed to one oriented around making things). Namely, in that shift, it rewards people who generate noise, regardless of the amount of signal contained. Polluting the well, in this case, results in everyone having to output a certain amount of signaling just to keep up.

It's entirely possible I am projecting some here, as I've struggled with this myself. I just don't want to always concerned with this tripe, because it's not the same at all as actually making software.

samspot is dead on here, at least from my perspective. I am very private and have suffered from social anxiety for years. I don't use Facebook because I am overly sensitive to the reactions of others (no likes? everyone hates me and thinks I'm stupid). I spend far too much time thinking about what others think of me, so my natural inclination is to completely remove myself from public view at all opportunities. I don't use GitHub for this reason. I'd like to contribute and become part of a community, but I'm so terrified of being judged that I never step forward.

So I completely understand the author's attempts to go public. It isn't about being the next DHH, it's about overcoming the fear of rejection. I have been considering "going public" for years for this very reason.

Just do it.

Seriously. What's the worst that could happen about pushing some open source code on GitHub? You don't have to start by contributing to openssl or something of similar magnitude. Just push some stuff to your personal one.

Trust me, I graduated college with an art degree and some of my earliest GitHub stuff shows this. But I keep pushing, literally. One of my babies has 1300+ stars and that's awesome to me. I don't think of it like validation, but more like "holy shit a lot of people like this." Some of my libraries are sitting at 0 stars, and that's ok.

Freaked me out a bit saying I was dead, I appear to be alive again, at least from checking with some random devices.

I'm not sure if this describes you, but I can be pretty thin skinned on the internet, which often discourages me from posting. I write things and delete them more often than I hit submit.

One nice thing about going public is it's unlikely anyone will notice you for a while. Even patio11 says only his brother looked at his blog for a few years (IIRC).

Even patio11 says only his brother looked at his blog for a few years (IIRC).

If only getting visitors were as easy as being related to them!

No, seriously speaking, my blog had perhaps 200~300 visitors for the average article from 2006 through, hmm, late 2009 or so? (What changed? A combination of discovering HN, going full time, changing my writing style to exclusively focus on the 4k to 8k word essay that I feel is generally my best work, and a minor refocusing on topics.)

I find blogging for blogging's sake to be just noise.

When I have something to share, I'll share; when I don't, I don't. Confidence has little to do with it.

Not to mention "put yourself out there" usually means the same rehash of the groupthink, and/or sometimes baseless opinions ("here's why I hate MongoDB because of blah blah blah, but I never used it actually")

There are a lot of useful blog posts and social media interactions and they're usually useful (especially with google), but no, I don't need another blog post saying how to install RabbitMQ

This is a whole separate topic, but I'm glad you mentioned it.

I don't see enough heresies entertained in a supposedly-progressive-minded crowd like HN. That is a problem, because it represents a real stumbling block to progress.

I think it is not just about "need for influence".

To learn somthing, is to 1. Read, 2. Do and 3. Teach. So blogging can be "teach" part.

But to get hired in the first place you have to show overconfidence in your knowledge and abilities.

Slightly annoying how that works.

I disagree. I could concede and agree that a bit of confidence is needed, but certainly not overconfidence.

Aye: good experience breads opportunities to gain good experience.

Though depending where you are starting from and trying to head to, there are ways that you can try to short-circuit this. If trying to move sideways into a technical role from another (or something else entirely) for instance then taking professional certifications in your own time can be useful for proving to a potential new employer that you really know+understand the skill that you believe you might have. The trick is finding something that actually means something to your target, which can be a black art (in the Microsoft ecosystem for instance the MCSA and MCSE qualifications can mean a lot to some, but little to others, and sometimes nothing without some commercial experience to back it up).

> overconfidence

I don't think this is about promoting overconfidence, more avoiding the underconfidence that comes from misjudging how capable other people are relative to yourself.

> But to get hired in the first place you have to show overconfidence in your knowledge and abilities.

I've had interviewers literally tell me that "Ya, that was a trick question to see if you thought you were god's gift to programming. The fact you didn't rate yourself a 9 or 10 proves that you are sensible."

So I'm not sure that is true.

I think it is a question of being confident enough that you don't second guess yourself yet humble enough that people don't think you have an ego that will cause workplace drama.

I've asked that. "On a scale of 1 to 10, how fluent do you feel in X?" And the questions asked get partially tailored to that.

Anyone answering 9 or 10 is in for it.

I'm sure it is different between interviewers. I was just providing an anecdote where it could be a bad idea.

I'm pretty sure the reason I elicited that reaction was I have a statement on my resume about a project I work on at my current job that handles as many transactions [in terms of $$] as that entire company does.

That is just a guess on my part tho.

I'm genuinely curious what you're testing with that. Is the idea to see how their scale of fluency matches up with yours, or to see if--assuming they use a reasonably similar scale as you--their self-evaluation if is accurate? If it's the latter, I agree overconfidence is a shortcoming, but it seems odd to structure the interview around sussing that out, especially given that the candidate may be only artificially overconfident because they think that's good for interviews. It feels more like a "gotcha!" than a genuine attempt to determine the candidate's quality. If it's the former (finding out what the candidate thinks "really good at X" means), that sounds interesting, but I don't quite get it.

It actually doesn't matter what the goalposts are very much. I've seen it asked with explicit goalposts, but it didn't seem to change anything except readjusting the goalposts; you could still infer the intent. "1 is a complete newbie, 10 means you helped write or at least have memorized the specification" just means the real scale is now 3-8, and they'll pick from there, scale accordingly. That said, someone who asks gains a point right off, as it were.

It's intended to view their confidence. As someone else posted, the older they've gotten, the lower they'd score. That's because as they've gotten more experience, they've had more of those "...wow, this code someone else wrote is so much better than mine" and "Oh man, this code I wrote 6 months ago is -terrible-", etc.

Someone who says "I'm a ten!", it doesn't matter what they think the goalpost is, the are the best they can imagine, which either means they are great (hence seeing how they answer on the followup questions), or it means they aren't, but they've never seen better...which means they aren't out learning, be it from other devs they work with, open source code, whatever. Throw them some hard questions, see how they answer. If they do well, snap them up, as they're good and they know it; if they do poorly, you may want to pass (a lot of the worst performing candidates I've seen ranked themselves 8-10 in a technology).

Someone who says "I'm a 3 maybe" probably means they are passably fluent, but they haven't done much with it, and they recognize it. Feel them out some easy questions first, and expect that if you need them to use this language there will be some mentoring and/or time needed to get them up to speed, but they'll probably be willing to learn given that they applied, and recognize they don't know it well.

Someone who says "I'd like to think I'm an 8, but most days I probably code more at a 6" means they're probably quite good with that language, but also quite realistic, and have had experience with requirements changing, making hard decisions due to delivery dates, have worked with a lot of libraries and people and seen some who are better, some who are worse. Start with intermediate questions and go from there.

It has no direct bearing on passing/failing the interview, it's just used to set the tone and expectations of the interview. If we have multiple positions open, it might be taken in conjunction with how you do to determine what position is offered, as well.

well you have to calibrate it what does a 1 a 5 or a 10 look like ?

I once interviewed a candidate whose resume listed his skills alongside a rating for each, from one to ten.

He rated himself 10/10 for Javascript, so I asked what he'd done with the language.

He replied "I've validated forms."

I was expecting more to say the least.

0 - 10 is a pretty horrible scale. If I was rating myself on such a scale I would have got lower as I got older. At 21 you can be a "10" by being the best in your CS class. At 35 you have to compare yourself to people with serious achievements.

That's actually what I'd expect. If someone rates themselves a 10, it means they don't know what they don't know.

No, the same standards apply regardless of how old you are. If this question comes up in an interview, you are being asked to rate yourself on an objective scale, not on a curve based on where you're at in your career. A 21-year old (or 35-year old, for that matter) who rates himself a 10 is almost certainly way overconfident and will run into (or cause) all sorts of problems. What you're complaining about is actually one of the main things this question is trying to determine.

Put another way, someone who rates himself a 10 because he's the best in his CS class is someone who doesn't know what he doesn't know, and is also likely to be a primadonna. Someone who shows some thoughtfulness, skepticism, and humility when answering this question is more mature and self-aware, and is much more likely to be an effective part of the team. The rest of the interview is supposed to establish how much they actually know.

But not knowing what you don't know is really to be expected of people who are recent grads in particular. I think this reveals why the question is problematic unless it is worded very specifically. A 10 from a recent grad could legitimately mean "I've had no problems learning everything that has been thrown at me so far and have very high confidence that I will easily be able to learn more" as much as it could mean "I know everything".

Then we will have communication issues and it will be difficult to work with you. The question was worded "On a scale of 1 to 10 how fluent do you feel in X". This is explicitly -fluency- (or phrased another way, "how well do you know X"), not something completely undefined like "How good are you at X?". The question is asking you to project your own perceived scale over a reasonably definable axis, not an ambiguous one.

Project the question into another domain - "I passed my intro Spanish class", "Oh yeah, on a scale of 1 to 10 how fluent are you in Spanish?", "I'm a 10". Nope.

There's a difference between being fluent in something and mastering it. You might be capable of reading and writing Java code, but that says nothing of your knowledge of JVM internals or of any particular framework etc.

It's like the difference between being fluent in Spanish and being able to use the language to write moving poetry or having a very extensive vocabulary.

I think it is much better to be as specific as possible about what skills you expect the candidate to have going in and what you expect them to be able to learn on the job.

My response would be see ya! thanks for interview(not really)

Which doesn't mean the question didn't serve it's purpose... :)

The entire point of an interview is figuring out the answer to these questions: "Will this person deliver what I want if I give them X amount of resources?" Second, "Can I stand working with them for at least one year?". Unfortunately, establishing a trusted conversation with someone takes a lot more than two questions! Fortunately, it's well worth it when you find them.

yeah, but the whole interview situation is completely unlike a work environment.

That's a fantastic point. I've spent a lot of time trying to figure out how to make it like one though.

Trick questions in interviews -- of any kind -- are a huge turn-off.

I never said I was the interviewer or that I agreed with it.

I'm pointing out it does happen and its specifically intended to locate potentially overconfident people and needle them.

I've noticed that isn't always true. I've sat on the other side of the interview desk, as it were, and one of the things that was always, ALWAYS a red flag, was someone speaking with too much confidence about themselves.

Obviously, yes, an interviewer prefers someone who knows the answers to every question asked. But an honest "I don't know/am not sure, but I think..." is a much, MUCH better answer than trying to bluff. If you're right after prefacing it with that, it's nearly as good as having bluffed correctly (the interviewer knows you didn't know or were not sure, but your knowledge of the tech is sufficiently good for you to deduce what unfamiliar constructs do), and if you're wrong, it counts nowhere nearly as badly against you as if you bluffed incorrectly (as you at least are showing you know and admit to what you don't know).

Now, obviously this is also partly dependent on the level of knowledge needed to answer the question. If you are, say, applying for a Java job and don't know what private means, you're probably not getting the job. But if you're applying for a Java job, and are asked for the difference between soft, weak, and phantom references, and respond "I'm not sure...I know the first two have to do with keeping references to objects that should not prevent garbage collection, like when adding items to a cache or something, but I don't remember the difference between them, I'd need to look that up. And I've never heard of a phantom reference", and then asking about them if there are no follow up questions, is still pretty damn good. And "I don't know; I'd imagine from the name they have to do with referencing an object in a way opaque to the garbage collector?" is also very good. "I don't know" is okay, it just might mean we'd not be willing to hire you on as a senior dev, but if you nail everything else maybe one level below that.

> "I don't know/am not sure, but I think..."

Coincidentally, this answer takes more confidence than bluffing. Bluffing is actually a signal of low confidence in a person. Too afraid to be wrong.

A bit off-topic, but I've always avoided language-familiarity questions. If a candidate has coded a lot of, say, Java, then I do expect them to know the details and their importance, mostly because I expect people to care about the tools they use. But to begin with, I don't at all mind hiring people who have literally never once coded in the programming language my team uses. They'll learn, and that investment is worth it. Moreover, I suspect that trying to connect "how well do they understand Java's reference system?" directly to "what seniority of engineer should this person be?" is not very wise. The things to optimize for in a senior engineer are problem solving skills, organization, experience and comfort with technical tradeoffs, architectural wisdom, and so on. Over the course of the next, say, two years, you'll get a lot more mileage out of that; having deep familiarity with your language of choice is just a bonus.

In general I agree.

I only ever asked language familiarity questions if either, A. The job is explicitly frontend and we need someone already well informed (Javascript has a lot of gotchas, and HTML and CSS can be complex; backend experience doesn't directly translate), or B. It's about a technology (not just language) on the person's resume. If you put down you know X, you better expect questions about X. The questions were generally prefaced with asking the person how well they felt they knew the tech, too; it was as much a "does this person know what they say they know" as much as it was a "does this person know X".

Reference type in Java isn't some esoteric library knowledge, either. It's not "tell me everything that implements the List interface" or something else as equally useless. If you want to build any sort of caching mechanism, you -really- ought to know about it, and if you're claiming great Java knowledge, and want to be in charge of delivering a Java product, it's reasonable to expect you to be familiar at least to when you might need to go read the Javadocs on those classes.

If you were a junior programmer who just coded tasks as they were handed to you, and had code reviews, and you don't know it (and you admit that), that's not the kind of question I'd expect you to know. And if you don't claim to have much experience with Java, obviously asking you a Java question is silly; I'd ask you something about what you do use.

I often leave languages off of my resume for exactly that reason. I've used Scala before, and I'd feel comfortable using it again, but I certainly don't know it well enough to handle on-the-spot Scala questions.

And that's probably a good thing. That, or indicate it separately. "Some experience with..." or something, to show that "Hey, I've seen this, I've played with it, I've got some concepts down, but if you ask me about this I am totally going to demur". I can't speak to interviewers at large, and certainly, it ends up with recruiters trying to hire for those languages which is irritating, but I'd like to see it just to ask in what context you've used it. "Played with it in my spare time" is a plus in my book; it means you're actively curious.

Having had a look at some resumes recently, I see some people who list every language / framework under the sun. I question how much knowledge they have of each of these. I would say learning Django alone to a decent level took me 6 months. I am still learning more 3 years later. If you claim to be learning 5 new languages a year, your knowledge of each can't be so deep.

I feel like that these days. I used Java, then Perl for years. For the Last three I have been doing mainly Django. I never feel I learn much about the intricacies of Python, as I am using Django, and coding at a higher level most of the time.

I think I may add your description of a senior engineer to my CV. I can see that these are the qualities I have built up over the years, while my intimate knowledge of programming languages has gone down, as I have used more of them. No one ever list these skills in job ads though....

The two are not mutually exclusive. You can be overconfident and ask for help. I mean - I'm an expert on what I know, not what I don't know, ya know?

I started a blog some years ago and here's what happened: nothing.

hahah, that's pretty funny. I've done some blogging myself and had no traction. The main reason i did it was to document progress of the startup I was building. Would be investors could check out the blog and get a high level view of my thinking and my output. I guess it isn't a terrible idea, if my startup had interest from investors, I think it would have been useful. It was just a wordpress blog, so it didn't take tons of time to maintain. But, I feel like the heart of the matter is, my startup didn't have much traction. And if it did, I don't believe this blog would have been a make or break factor in an investor deciding to invest. In my current venture, I just tweet and send emails when a new feature is released. It basically serves the same purpose and takes mere minutes to produce.

I have started to enjoy nothing happening to my blog. Since I am sure that nobody actually reads my blog I write fearlessly. Some times the posts are so silly that I would have been embarrassed if some one actually read it.

Any way I go back to my old posts from time to time and have a good laugh about them. Nothing is the best thing that has happened to my blog.

Same. I always think someone is going to see it and laugh... "Oh, that's so cute! You tried."

My recent favorite reminder of why I need others in order to program well is the list of things I am unqualified to say about my own work, including, but not limited to

- Is the user interface intuitive?

- Will this library be efficient for its most common uses?

- Have I given the simplest explanation for what this thing I wrote does?

It helps to be able to say, "I can't know that yet, not until I ask someone." That might happen in use cases, watching over someone's shoulder, showing a conference poster, or getting complaints on a message board two years later, but it's always collaborative, even when I have my act together.

I have a lot of ideas, many of them good, some great, but quite a few are horrible. Just one person to bounce them off of (and who isn't afraid to say "no") is incredibly valuable.

Why do people who mention Linus Torvalds, Theo de Raadt, Rich Hickey, Fabrice Bellard, Simon Peyton Jones, Satoshi Nakamoto (ok, that may be several people) etc. get downvoted to oblivion?

This seems especially weird seen most people here are using Git and many of the things we have wouldn't exist without these people.

If I'm not mistaken Rich Hickey took one or two years off, pulled the plug and went on to create Clojure.

Saying "But that's only one in ten million" ain't an argument: there are genius programmers out there and we're all using daily piece of code from these genius programmers.

Watching videos from Linus Torvalds (the talk about Git he gave at Google is amazing), Theo de Raadt (anything about security), Rich Hickey, Simon Peyton Jones, etc. is one of the thing I prefer to do.

I've also personally met several genius programmers: including one who had been working on at Adobe and who (while not at Adobe anymore) then single-handedly wrote a Java bytecode to Objective-C source code converter. I know for a fact that this converter allowed to write an app which reached top 3 in the appstore.

I've also competed ten years ago or so versus plain geniuses in the TopCoder algorithmic competition. Some of the best ranked coder there had incredible achievements, like for one of them having won the mathematics olympiad several times.

That's the thing: even "one in ten million" means that there are quite a few genius programmers out there among us. And we owe a lot to them: not only have we many great tools and projects which would never have existed without them, but they also let us get a glimpse inside their beautiful minds through the messages / blog they post and through the video they make.

Why downvote?

The down vote comes from the idea that this is not the only way to create good, useful, widely used and appreciated code. Of course there are geniuses that have enough raw talent, dedication and focus to do amazing things with ease. But there are plenty of others that while not considered a 'genius', will have to use pure grit to accomplish feats of greatness.

Fabrice Bellard seems to me an outlier to the author's premise. He's like the very definition of a hermit genius-programmer.

Well of course there will always be outliers =]. I prefer the Satoshi Nakamoto outlier myself.

Donald Knuth.

Fitz actually jokes about METAFONT in the video at 11:29: http://youtu.be/0SARbwvhupQ?t=11m27s

I'm pretty sure "Satoshi Nakamoto" is 2-3 people.

I can think of a ton of programmers that have single handled changed our profession, Linus, ESR, DHH, Doug Cutting, Brendan Eich, James Gosling.. I dunno, I don't think it's really that rare at all.

Out of how many programmers in the world, ever? They're basically statistical noise against the backdrop of all coders everywhere.

As a sidenote, I'd put Grace Hopper up there instead of DHH, and Dennis Ritchie and Ken Thomopson seem to be missing. ; )

John Carmack.

He's not a hermit, but definitely a genius programmer who has built things by himself without putting himself "out there."

Yeah definitely I was thinking about him while driving home today. In the book Masters of Doom you get a good sense for how much time he spent thinking and programming out ideas he came up in what seems like total-isolation. There wasn't Internet or books on the subject available at the time because it didn't exist yet. In the Graphics Programming Black Book there are also anecdotes by Abrash that give you a sense for what kind of solitary genius this guy has.

There's two ways of being known as a hot-shit programmer. The first has been described by Joel ("Smart. Gets things done.") and Rands ("Be productive, be fantastically clever when necessary, speak truth to power, hit your dates, and don’t ship crap."). These are people who deliver the goods. Deliver them often enough and folks will know your name.

The second way is harder, particularly for the introverted. This involves self-marketing, being obviously and repeatedly correct. It also requires a bit of political adeptness and a bit of visionary helps as well. Both Joel and Rands have done very well here and there's a number of others who have made a name for themselves via the web. If you have a choice, choose this method as it yields better visibility. Although it also means that screw ups are magnified if and when they occur.

Am I the best programmer going? No. If you're reading this, I suspect you're not either... But I work hard at delivering the goods, and I'm good at what I do. People like this get remembered when new projects start. No matter how big or small your niche is or how many you fill, impress others. And don't needlessly piss them off - people remember assholes far longer than good 'uns.

Over the past few years, I've put myself out there as well, not because I'm extroverted but because there is no better laboratory test of your ideas/convictions.

I'd only add one thing to this approach: make sure that you are at least solving one of your own problems and that the solution is real -- i.e. it really benefits you in some measurable way.

I did this a few years ago with a relatively obscure gem that solved a real problem I actually had with refactoring business logic. I put it out there to get feedback on it and was rather disappointed with the initial reactions ("You don't understand MVC!" "You've done it all wrong!"). But I put myself aside and tried to understand the feedback from their point of view.

One thing that became obvious is that most people didn't see the solution as valid because they had completely different context and approach. Fair enough. There were also several that simply 'cargo-culted' mvc responses without realizing that my gem was a thin wrapper around methods that DHH had already put in Rails (so if I didn't understand MVC, neither did he! ;) And no one commented on the biggest problems I legitimately saw with my approach: thread-safety and closures on controller instances in a prod env.

But none of that stopped this gem from being incredibly useful as I tried to break apart a giant monolithic legacy app that challenged traditional refactoring approaches. Maybe it's not perfect, but at least it's a way forward.

So the moral of the story is: put your idea out there, expect feedback and criticism (maybe only criticism since the people who like things tend to keep quiet (e.g. the rvm story last year)) -- but so long as this thing is useful to one person (only you?), use it to make your life easier!

If it's useful to others, that's a bonus learning opportunity, but that shouldn't be your sole motivation.

off topic: I missed this kind of talk at the last two google io's

Those who can't code, blog.

Reply on news.ycombinator.com