Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Stuck due to “knowing too much”

programmers.stackexchange.com

Post reply on HN
Print view View original

31–40 of 110 posts

Re: Stuck due to “knowing too much”

#31

Post by gfosco »

pooriaazimi wrote: post #2

I submitted it because it's exactly how I feel lately, and would like to know what others think about it. My current (personal/academic) project involves Node.js, (Iced)CoffeeScript, MongoDB (with Mangoose), PhantomJS, Mocha, Neo4j, Nutch, Solr, Hadoop and Mahout. All of them are new technologies and are changing rapidly and daily, and I find new stuff every day that forces me to change my APIs and the way I do thing…

Try using less things... How about just using Node and Mongo, and building the rest yourself?

Node being in its infancy is a pretty good reason in itself to not tie yourself to the frameworks which are likewise infantile.

I've got several production back-ends and a full-stack e-commerce site running with Node and MySQL only.

Re: Stuck due to “knowing too much”

#32

Post by engtech »

pooriaazimi wrote: post #2

I submitted it because it's exactly how I feel lately, and would like to know what others think about it. My current (personal/academic) project involves Node.js, (Iced)CoffeeScript, MongoDB (with Mangoose), PhantomJS, Mocha, Neo4j, Nutch, Solr, Hadoop and Mahout. All of them are new technologies and are changing rapidly and daily, and I find new stuff every day that forces me to change my APIs and the way I do thing…

Experience has taught me that if you want to get something done, you need to stay off the bleeding edge, or at the very least only drag yourself to the bleeding edge at controlled increments.

The only thing that should be changing often is YOUR code. Everything else should be as stable as possible.

What I find works best is playing leap frog: * your code is unstable / upgrade nothing * your code is stable, stop changing it and upgrade things one at a time

test suites / regressions are a must.

Re: Stuck due to “knowing too much”

#33

Post by antidoh »

angdis wrote: post #24
frankc wrote: post #7

The moderation on programmers.stackexchange.com is completely overbearing. How can a question about analysis paralysis be closed for off-topic? If stackoverflow is for technical question/answer, what is the point the linked stackexchange if not for discussion of a common programmer problem like this? I see stuff like this closed all time time and it's really annoying.

While I agree that very often interesting topics get closed, they DO HAVE a good reason for doing so. The intent of all the stack-exchange sites is to provide a high-quality resource of _answers_ for specific _questions_. If "the answer" to a question ends up being a divergent discussion, then not closing responses invites more of these types of questions. Over time that will subvert the intent of stack-exchange. The…

"Hacker-news, however, is more free-form and definitely much better for "need advice" type questions with highly subjective answers."

Assuming the question doesn't drop past the second page or so before someone who is interested in the question sees it. But yeah, is it is seen and kept alive for a few minutes, the discussion is potentially very useful.

Re: Stuck due to “knowing too much”

#34

Post by antidoh »

benjaminwootton wrote: post #8
frankc wrote: post #7

The moderation on programmers.stackexchange.com is completely overbearing. How can a question about analysis paralysis be closed for off-topic? If stackoverflow is for technical question/answer, what is the point the linked stackexchange if not for discussion of a common programmer problem like this? I see stuff like this closed all time time and it's really annoying.

Agreed. If any StackExchange people read this please do something about this. It's a great site that fills a necessary niche but the moderation is killing it!

At this point it can't be done. StackExchange is the way it is by preference of the older controlling users.

Re: Stuck due to “knowing too much”

#35

Post by wpietri »

Modestly off topic: Can I say how much I have grown to hate StackOverflow and its cousins? Just like this article, half of my favorite threads on there are closed or have been deleted due to violating some sort of rule that is not unreasonable in theory but leads to obvious idiocy in practice.

A great example of people losing track of the forest because of all the damned trees in the way.

Re: Stuck due to “knowing too much”

#36

Post by chmike »

Aqua_Geek wrote: post #16

I also found myself getting stuck and over-engineering. Now, I try to take a phased approach to development: 1. Make it work. 2. Make it good (visually or otherwise). 3. Make it fast.

This good. But how do you handle it when designing a protocol and a data encoding ? You can't iterate.

Re: Stuck due to “knowing too much”

#37

Post by wpietri »

My solution to this is a simple kanban board. Trello.com is a good on-line one, but when possible I'll make them out of index cards or post-it notes on the wall.

People think that a kanban board is a to-do list, but there's a subtle difference. The real heart of is limiting the number of work-in-progress items. That means that the backlog is less a to-do list and more a things-I-am-definitely-not-doing-now list.

For me, every time I have one of those "It would be nice" thoughts, I write it down on a card for the kanban, throw it in the backlog, and forget it until the next time I make a prioritization pass. Most of the cards end up in what I think of as "the pit of nice ideas", the bottom segment of the board that never moves up because more important things always come up.

For the curious, you can see a photo and a little more explanation here:

http://www.quora.com/What-are-some-uncommon-ways-to-work-sma...

Re: Stuck due to “knowing too much”

#38

Post by mbenjaminsmith »

geophile wrote: post #27

Some lessons in software engineering take a long, long time to sink in. (For me, at least): - Don't generalize too early, which seems to address the original question pretty directly. - Don't optimize too early. - Maintainable > Correct > Fast. Maintainable = comprehensible. If it isn't maintainable, you won't know if it's correct. If it isn't correct, there's no point in making it fast. - Minimize state. Deriving so…

I'm starting to see this more clearly as time passes. I've spent years developing the ability to "engineer" things and only now am I realizing that I've learned to "over-engineer" things. Writing software has taken on a life of its own that is actually counter to its purpose.

I don't think in terms of simplification though, I see it as "write what you need, when you need it." But I absolutely agree that you should avoid generalization aggressively and optimization should rarely be on the table. I also agree about minimizing state. Do it all "on the fly" unless there's known performance hit.

Re: Stuck due to “knowing too much”

#39

Post by vibrunazo »

joshklein wrote: post #30

This is one of the most pernicious forms of procrastination, because it's hard for smart people to accept that perfectionism is a negative trait. The more self-worth you derive from your previous successful outcomes, the more susceptible your ego becomes to fear of future failures. Your amygdala is telling you that you are successful already, so don't shake the boat. Tread carefully and you'll always have an excuse f…

Arrogance is probably the trait that most limits our intelligence. I've seen many smart people do extremely stupid stuff just out of fear of admitting they are prone to mistakes and failure. Humbleness is hard to master, but it's very important for us to try and practice it as much as possible.

Potential of success = (Intelligence / arrogance) + luck

Re: Stuck due to “knowing too much”

#40

Post by wpietri »

angdis wrote: post #24
frankc wrote: post #7

The moderation on programmers.stackexchange.com is completely overbearing. How can a question about analysis paralysis be closed for off-topic? If stackoverflow is for technical question/answer, what is the point the linked stackexchange if not for discussion of a common programmer problem like this? I see stuff like this closed all time time and it's really annoying.

While I agree that very often interesting topics get closed, they DO HAVE a good reason for doing so. The intent of all the stack-exchange sites is to provide a high-quality resource of _answers_ for specific _questions_. If "the answer" to a question ends up being a divergent discussion, then not closing responses invites more of these types of questions. Over time that will subvert the intent of stack-exchange. The…

Honestly, the original intent of a startup almost doesn't mater.

What a business does is always an intersection of what the producers and the consumers want. If users want something divergent but reasonable, the answer shouldn't be, "Well, fuck you then." The right response is, "Boy, that's a a great opportunity."

As proof, look at Friendster. They owned the social networking space before Facebook even existed. But it had started out as a dating site, and the founder stubbornly insisted that it stay a dating site no matter what the users wanted. At this point they're not even a punchline; MySpace has now taken over the spot of best-known failure.

I've stopped contributing to StackOverflow because I'm tired of them killing good discussions. Why take the risk that yet another good answer will get eaten?

Post reply on HN