Oh, you are not alone :)

In fact this is a fairly typical problem of people working alone and with no external feedback. Things like killing a day to decide how to name a temporary variable is a bane of lonely programmer. Next time when you are starting on a new project, make sure you get a supporter or an adopter. The more the merrier, but one should do. This creates a feeling of being accountable to someone and it tremendously helps to make the milestones... which is another thing - having milestones that you cannot afford to miss is golden. This allows breaking out of that stupid loop of obsessing over details that do not matter (while burning out mentally) and keep a larger view of the project in mind at all times. For example, knowing that you are getting someone's blog coverage in two weeks is a sure way to not only get the beta in its best polished state (of the decade :)), but also to redo the website and what not. Productivity jumps by the orders of magnitude.

Between the techniques and what nots I find the most useful thing is writing thoughts down. Say you are working on a project, and currently dealing with a feature A of a module B of a web interface, which is in turn just a smaller part of the whole thing. And then you realize that there is that one other thing that absolutely needs to be done and not forgotten. Write it down. Getting it out of one's head and onto a paper helps freeing up idle cycles that brain would otherwise spend worrying about not forgetting this thought. There is a big, heavily commercialized theory built on top of this simple idea - GTD or Get Things Done - but that's all cruft. Just get Things for iPhone and it should be enough.

So, yeah... I hear you, and (a) it is typical (b) you need firm goals you truly commit to. That's it.

(edit) If you consider chemicals, then no, you don't want them. If you really really really desperate, try modafinil. Very sparingly, like once a month, just to jump-start the productivity and get things rolling.

It's not always a bad thing to go down a path and then to decide it's the wrong path. It's a far cry worse to continue down the wrong path because you're determined to finish what you've started.

If follow through is your issue, maybe you're biting off too much to start or maybe you haven't found a idea worth the follow through. What you've first gotta do is to count the cost of implementation. It might take you a month or so of actual work on your idea to fully grasp the real size of your project, but once you have the picture, you have to ask is this worth seeing it through to completion? Anything is worth it if it's worth it to you, if it's some contribution you want to make that you feel and believe is worth making. Once you find that, then just commit and promise yourself that you'll make one small step at a time until you're done. Form a habit. If you can form a habit you'll more likely succeed than you will if you only work on your "project" here and there as you feel the desire to do so.

First guideline: Be like the ant. http://www.charlesthorntonblog.com/post/The-Ant-Principle.as...

Second guideline: Manage your constraints. Get to version 1 of your deliverable as soon as possible by limiting just how big v1 actually is. (See Eric Ries on Vimeo.)

I bought myself a whiteboard. And on it I write down all the tasks (large and small) and milestones to completing my project. The board obviously goes through revisions as the project grows and morphs. But it does something really important that I wasn't doing before.

I too get a thrill from starting a new project because it's still just a dream but seems acheivable. There's probably some cogsci reason for this, dopamine levels and so on. And when the project starts to drag on you're not getting that dopamine hit anymore and so you get an urge to start a new project to get that rush again.

What you need to do is learn to get a thrill from completing milestones. Being able to cross off one of the tasks on that whiteboard feels awfully damned good. I keep the crossed off tasks on the board for a while because I find that thrill of completing it lasts for a while and keeps me motivated to cross off the next one.

I'd recommend Robert Boice's book "Procrastination and Blocking." It helped me a lot. The basic idea of the book is that the cause of procrastination is impulsiveness (as in the Piers Steel's book mentioned below, except that Boice was there first, I think :).

Spending a year at death's door did wonders for my neurotic crap. Not something I really recommend. But I would recommend you watch the movie "Beyond Rangoon" and then go do some volunteer work at, say, a homeless shelter or working with the chronically ill or something -- something that really punches your buttons and gives you a bit of perspective. Or at least join some email lists for people with chronic, incurable illness.

Also, do some research and consider vitamin therapy. There may be a brain chemistry issue that may be very treatable with aggressive nutrition.

Best of luck.

i hear you man. i find myself in nearly the same predicament, though i just lost my job…happy days indeed. i'm with abcd_f—write it down and tell someone about it. focus on doing one thing really well but still giving yourself time to decompress, to think. don't let other ideas distract you—save them for later. finish one thing. one thing that you love and can be proud of. and share it with others—especially the design aspects. find someone who has an eye for detail to help you—someone who believes in what you are doing—even if you are yourself a good designer. have them help you create a grand plan for your product, pare it down to a manageable size, and iterate from there. the thing i'm learning about software, job-hunting, child rearing is that it is never done. as a perfectionist myself, it's a hard pill to swallow—there is ALWAYS disappointment, but i don't think that anything meaningful ever gets made without disappointment. and people who don't get disappointed probably never change things (though the suicide rate is lower). life is hard. my three years since graduation have gone from bad to worse. don't feel bad about not finishing school—i thought that i would be able to get something useful that at least put food on the table and pay the mortgage, and i find myself scraping by doing the same stuff or worse than before i decided to go to school. educational inflation sucks, especially out here in the rust belt.

so now i'm doing exactly the advice i'm giving you, except with about half resolve and double the complexity—advice is easier to give than to follow. i'm asking for help from the few who believe in me. here's an episode of a podcast that i've found cathartic— http://5by5.tv/b2w/7 —it doesn't get anything done, but it has gotten me thinking and laughing at least. listen to it while doing dishes or yardwork.

as for medication—i was diagnosed with depression and it ended up being vitamin d deficiency. also, yoga helps.

here on HN is an entertaining presentation by mike lee (http://www.infoq.com/presentations/Making-Apps-That-Dont-Suc...) in which he talks about selling everything to work for will shipley for one year without pay. so if none of my advice suits you—do something crazy that propels you in the direction you want to go.

As psychologist Piers Steel points out in his book The Procrastination Equation, the problem is not perfectionism but impulse control.

Even though you have no problem getting started initially, the label of procrastination applies. Here is my summary of the book:

Book: The Procrastination Equation, Piers Steel (psychologist), 2011

  www.procrastinus.com

Perfectionism does not lead to procrastination - this is well studied. It may be that they are thought to be linked because of the cases where there is this discrepancy in behavior. Procrastination is a result of impulsiveness. Self-control and delaying gratification are key to controlling procrastination.

Procrastinators suffer from

  * weak impulse control
  * lack of persistence
  * lack of work discipline
  * lack of time management skill
  * inability to work methodically

Motivation can be modeled by

  (expectancy * value) / (impulsiveness * delay)

  * The numerator is Expected Utility Theory in economics
  * Expectancy is the perceived likelihood of reward or success
  * Value is the perceived value of the reward
  * Delay is the perceived delay in receiving the reward
  * Impulsiveness is the tendency to (irrationally) pursue immediate reward instead

Impulsiveness is moderated by the prefrontal cortex

  * the prefrontal cortex is late to develop in humans,   
  impulse control develops slowly in children
  * adults with damage to the prefrontal cortex may be 
  markedly more impulsive

Temptation - defeating impulse control

Important factors:

  * Proximity to temptation is a major factor in impulsiveness
  (low barriers to gratification)
  * Variable schedule of reinforcement causes a robust response

Modern society offers many more sources of temptation

Expectancy - optimism, expectation of success

  * too much pessimism causes procrastination
  low expectation of success keeps us from starting
  * too much optimism causes procrastination
  unrealistic ease of success may delay starting until the last moment

Techniques for improving optimism:

  * success spirals - progressive series of successes build confidence (e.g. earning scout badges). regularly stretching  your limits is important to teach yourself confidence in your ability to tackle something difficult
  * vicarious victory - relating to someone’s success story, finding inspiration in books, movies, inspirational speakers, joining a group of inspirational people
  * wish fulfillment - visualization of success and contrasting with where you are now. Visualization that only focuses on the goal may drain motivation to complete the necessary steps. As you visualize attaining the goal and then contrasting the current situation, maintain your optimism so that you can translate this visualization into a plan of action.
  * Plan for the worst, hope for the best - develop strategies to recover from falling back into old habits.  Anticipate temptations and find ways to counter them.

Value - the perceived value of completing a task

  * create a chain of goals that helps connect less pleasurable tasks to the ultimate desired goal
  * frame goals positively, rather than goals of avoiding something negative
  * make games out of tasks - avoid boredom
  * justify tasks by connecting them to your goals
  * recognize your available energy, and plan around it - schedule difficult tasks for your morning and mid-day peak performance (likely between 10 and 2)
  * commit to a regular schedule of exercise and sleep
  * snack as needed - avoid hunger
  * make sleep predictable, with a regular wind-down routine
  * respect your limitations
  * as an antidote to task avoidance, identify and do related tasks that are less intimidating - whittle down the main task until it is less intimidating
  * reward yourself for accomplishments

Impulsiveness

  * identify and put temptations out of reach
  * satisfy your needs first before they become a distraction.
  * schedule your leisure time ahead - work harder knowing your leisure is ensured
  * add disincentives to your temptations, a penalty - e.g. a personal tax for infractions
  * mentally contaminate temptations, making them seem less attractive in your mind
  * eliminate cues that trigger temptation - e.g. keep your workspace clean

Criticizes S.M.A.R.T. goals (specific, measurable, attainable, realistic and time-anchored)

  * specific is redundant with both measurable and time-anchored
  * attainable is redundant with realistic
  * missing key concepts that are important to effective goals

Instead, goals must be

  * challenging (expectancy not too high - too easy)
  * meaningful (high value)
  * framed in specific terms so that you know when you have to achieve them - what you have to do and when you have to be done
  * if long-term, then broken down into a series of short-term objectives. particularly daunting goals should start with a small goal to kick off.
  * organized into routines that occur regularly at the same time and place. A predictable work schedule is important.

BTW, your topic was translated to Russian and posted to popular Russian IT resource Habrahabr. Here you can read comments: http://translate.google.com/translate?js=n&prev=_t&h...

James Altucher's blog. http://www.altucherconfidential.com

also, read Thoreau and Emerson. And remember this...

THE WORLD NEEDS YOU. Get away from paranoid self-pity, and you'll be fine. You're great. The world waits to delight in YOU. All this time...you've been BECOMING YOURSELF> knock it out of the park this time.

Focus is one of the most important leadership skill, we need to learn to take responsibility to finish our job, and have same level of focus for any task and its sub task's at all times.

If a child cries we can easily distract him by giving a new toy, but if we learn to keep focus on one goal I guess we can perform little better then the child.

You don't have to do everything alone. If you like writing and refactoring code - awesome! There are people out there who like to document stuff, there are people who like to design or polish. My point is - if you don't do everything yourself - find someone to do it with, so that the "Project" is complete.

If it's not too personal, how do you pay bills? It seems like you are working independently. Maybe you should work within a group so that there is some external support combined with some external pressure to get past your perfectionism (much like @abcd_f mentioned).

Maybe not the answer you're looking for, and I didn't see it in your list of remedies that you've tried, but have you considered medical help? As in, medication for ADD or similar?

What do you really want to do? School? A good job? A cool project? Partner up with someone to keep you accountable to at least one goal at a time!

Ritalin?

I feel you, I feel the same way.

I also like starting projects more than finishing them.

I think writing down your ideas will help solidify them. I find from experience that the more I write about my ideas, the more likely they are to come to fruition.

see: http://www.paulgraham.com/discover.html

This is not exactly a new idea, but I think it doesn't apply equally to all people.

You're the type of person who likes starting projects more than finishing them. This has implications about how your brain works or generally on how you think. Not everyone is like that, some people prefer to finish projects they already have, and it's not a matter of some acquired discipline either.

If you're like me, your ideas might seem scattered, you jump from one idea to the next, one day you think "man wouldn't it be awesome to build an app that does X!" then two days later, you feel constrained, and want to explore other ideas. So you jump to some other project or idea. Exploring various ideas seems very exciting, that's why you feel a bit drained if you spend a lot of time on just one idea.

When your ideas are scattered like that, you need to work on polishing them and flushing them out. The best way is either writing about them, or talking to other people about them. If you're introverted, writing about them is probably easier, and even better perhaps.

This next bit might be controversial, but if you like starting projects more than finishing them, your personality might be xNTP on the Myers-Briggs indicator, which (the next bit is even more controversial) means you have "extraverted intuition"[0]. This basically means the same thing I said above: your ideas are all over the place, only when you express them do you get to solidify them and get a better picture.

[0] http://greenlightwiki.com/lenore-exegesis/Extraverted_Intuit...