wu :: forums - Print Page


    
      
        wu :: forums
        (http://www.ocf.berkeley.edu/~wwu/cgi-bin/yabb/YaBB.cgi)
      

        riddles >> general problem-solving / chatting / whatever >> Curing perfectionism via intentional bugs.
        
(Message started by: amichail on Apr 18^th, 2009, 9:31am)

Title: Curing perfectionism via intentional bugs.
Post by amichail on Apr 18^th, 2009, 9:31am

What do you think of this?

http://news.ycombinator.com/item?id=567827

Title: Re: Curing perfectionism via intentional bugs.
Post by towr on Apr 18^th, 2009, 2:20pm

I think a shrink might be in order, rather than introducing bugs in an almost certainly already bugfilled program.

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 18^th, 2009, 2:43pm

on 04/18/09 at 14:20:26, towr wrote:

I think a shrink might be in order, rather than introducing bugs in an almost certainly already bugfilled program.

But a shrink might recommend something like this. See:

http://books.google.ca/books?id=w6ESB0Ys4IoC&pg=RA1-PA248&lpg=RA1-PA248&dq=perfectionism+%22intentional+errors%22&source=bl&ots=xr3CG6iS8g&sig=b71qE-dsTc0jxZ8la49Hy0J_tqI&hl=en&ei=U0zpSaXXCMyrtge6jfCXBg&sa=X&oi=book_result&ct=result&resnum=1

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 18^th, 2009, 2:46pm

Also, what about having a tool introduce minor errors into every email? Would that make people worry less about perfect emails?

Title: Re: Curing perfectionism via intentional bugs.
Post by towr on Apr 18^th, 2009, 2:52pm

on 04/18/09 at 14:46:50, amichail wrote:

Also, what about having a tool introduce minor errors into every email? Would that make people worry less about perfect emails?

I doubt it, and it would annoy the hell out of the recipient. Spell checkers were invented for a reason.

Title: Re: Curing perfectionism via intentional bugs.
Post by towr on Apr 18^th, 2009, 2:56pm

on 04/18/09 at 14:43:35, amichail wrote:

But a shrink might recommend something like this. See:

<unnecessarily long url> (http://books.google.ca/books?id=w6ESB0Ys4IoC&pg=RA1-PA248&lpg=RA1-PA248&dq=perfectionism+%22intentional+errors%22&source=bl&ots=xr3CG6iS8g&sig=b71qE-dsTc0jxZ8la49Hy0J_tqI&hl=en&ei=U0zpSaXXCMyrtge6jfCXBg&sa=X&oi=book_result&ct=result&resnum=1)

I think a shrink is about as likely to tell a programmer to write buggy code on purpose as therapy against perfectionism, as he is to tell a surgeon to stab his patient with a kitchen knife to work out his aggression issues.

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 18^th, 2009, 2:58pm

on 04/18/09 at 14:52:44, towr wrote:

I doubt it, and it would annoy the hell out of the recipient. Spell checkers were invented for a reason.

I think the problem is that computers provide the potential for perfection, yet their users are not perfect.

And so you end up with people suffering from perfectionism to such a degree that they get very little work done.

I am beginning to believe that successful people are those who don't care about their errors that much. It's more important to get as much work done as possible.

If you end up doing something that people care about, then you can work out the details and correct the problems later.

Title: Re: Curing perfectionism via intentional bugs.
Post by towr on Apr 19^th, 2009, 8:16am

on 04/18/09 at 14:58:01, amichail wrote:

I think the problem is that computers provide the potential for perfection, yet their users are not perfect.

Neither are computers, and especially not computer programs.

Quote:

And so you end up with people suffering from perfectionism to such a degree that they get very little work done.

Was there any increase in perfectionism since the invention of computers? Are forms of OCD even caused by environmental, rather than merely mental, factors?

Quote:

I am beginning to believe that successful people are those who don't care about their errors that much. It's more important to get as much work done as possible.

Hence the current predicament the world is in. Work, work, work, and throw caution to the wind.
Neither extreme is any good.

Quote:

If you end up doing something that people care about, then you can work out the details and correct the problems later.

Fixing mistakes generally doesn't give as good results as not incorporating them in the first place. Especially in the case of computer programs, patches tend to introduce new bugs and exploits.

Title: Re: Curing perfectionism via intentional bugs.
Post by Obob on Apr 19^th, 2009, 11:41am

As a mathematician, perfectionism is somewhat inherent in what I do. The attitude in mathematics is that if an argument isn't perfect, it isn't worth the paper its printed on. As such, when I write computer programs for use in mathematics research, I need to make absolutely sure that there are as few bugs as possible. To some extent I can test that there aren't bugs by observing that the program has the "expected" behavior on a few cases that I have worked out by hand. If the program has unexpected behavior on even a single case, its a pile of crap and needs to be fixed.

This might be very different from what you are talking about, since at least at points you've focused on bugs in the user interface. The programs I write are designed to be used by nobody but myself, so the interface is stripped down to what I need it to be; the algorithm that I am testing makes up 90+% of the work that goes into the program, and if it isn't flawless then I just wasted my time in programming it.

This need for perfectionism in my programming doesn't cripple my ability to program; it just makes me think harder about the problem to make sure I really understand what is going on before trying to "guess" a computer program solution.

Granted, the programs I write generally fit within a couple to few hundred lines. But understanding them in depth is absolutely essential to make sure that the program is actually computing what I want it to compute. Adding extra bugs would just add a huge extra headache, because I can't afford to let them live.

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 19^th, 2009, 12:42pm

on 04/19/09 at 11:41:57, Obob wrote:

As a mathematician, perfectionism is somewhat inherent in what I do.

Do you ever check anything more than once? Does it get worse over time?

Title: Re: Curing perfectionism via intentional bugs.
Post by Obob on Apr 19^th, 2009, 1:00pm

Of course I check things more than once (when it comes to research). It takes so long to develop a page of original mathematics that by the end of it I have thought about the material on the page for probably 10+ hours. The issue is not that perfectionism slows me down; rather, you can't go forward building a tower of ideas if you don't understand the previous ones nearly perfectly.

I wouldn't say that I am a perfectionist to a fault. Little things like misplaced punctuation or whatnot, while it should definitely be corrected, don't keep me awake at night. However, little details in a big argument certainly can. There is a sense in which it can be a valuable skill to suppress worrying about the little details and focus on the big picture. However, this is more useful for developing new ideas than it is for actually proving or publishing anything. When it comes to the end of the day, you DO have to worry about the little details if you want to actually prove anything. It can often turn out that the "little details" are actually where all the difficulty, and perhaps even interest, in an argument lies. Without actually worrying about the little details, you won't know if something is actually just a little detail or in reality the crux of the whole matter.

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 19^th, 2009, 1:11pm

on 04/19/09 at 13:00:44, Obob wrote:

Of course I check things more than once (when it comes to research).

If this number is not too large and doesn't increase with time, then there's probably no problem.

But I suspect in some people, it does get worse over time to the point of near paralysis.

In the case of a mathematician, it may be because he/she has built a reputation that he/she would like to keep and so there is more pressure to get things right.

Another reason is that the normal process of aging makes it more difficult to be correct, hence the need for more checking.

Finally, you don't "control" your brain. You may want to focus your attention on details that matter and worry less about such things as punctuation, etc. Your brain may not agree with you in that regard.

Title: Re: Curing perfectionism via intentional bugs.
Post by Grimbal on Apr 20^th, 2009, 4:42am

What about intentionally introduce failures in airplanes? Will people worry less about crashes?

Title: Re: Curing perfectionism via intentional bugs.
Post by Eigenray on Apr 20^th, 2009, 5:00am

on 04/20/09 at 04:42:43, Grimbal wrote:

What about intentionally introduce failures in airplanes? Will people worry less about crashes?

I don't know about that, but I do know that you should always bring a bomb with you when you fly. After all, the probability of there being a bomb on your plane is pretty small, so the probability of there being two...

Title: Re: Curing perfectionism via intentional bugs.
Post by rmsgrey on Apr 20^th, 2009, 6:07am

on 04/20/09 at 05:00:01, Eigenray wrote:

That's why I engraved my name on this bullet :P

Title: Re: Curing perfectionism via intentional bugs.
Post by Grimbal on Apr 20^th, 2009, 6:34am

I think the cure against perfectionism is a tight deadline. The program might not be perfect, but you can still congratulate yourself that you wrote the best program possible giving the time you had.

Title: Re: Curing perfectionism via intentional hugs.
Post by towr on Apr 20^th, 2009, 6:39am

You get a better topic title when you misread it, I think. But would it work?

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 20^th, 2009, 10:04am

on 04/20/09 at 06:34:17, Grimbal wrote:

I think the cure against perfectionism is a tight deadline. The program might not be perfect, but you can still congratulate yourself that you wrote the best program possible giving the time you had.

Conferences are better than journals in that regard as the latter have no deadlines.

But even with a deadline, you would be more productive if you don't check things too many times.

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 20^th, 2009, 10:07am

on 04/20/09 at 04:42:43, Grimbal wrote:

What about intentionally introduce failures in airplanes? Will people worry less about crashes?

Anything you would introduce -- manually or in an automated way -- must be minor. There are many imperfections that you can introduce in an airplane that have no impact on safety.

Title: Re: Curing perfectionism via intentional bugs.
Post by towr on Apr 20^th, 2009, 10:11am

on 04/20/09 at 10:04:48, amichail wrote:

But even with a deadline, you would be more productive if you don't check things too many times.

You might be more productive, but you'd be less successful in getting your paper accepted.

Title: Re: Curing perfectionism via intentional bugs.
Post by amichail on Apr 20^th, 2009, 10:14am

on 04/20/09 at 10:11:52, towr wrote:

You might be more productive, but you'd be less successful in getting your paper accepted.

Not clear. There's a tradeoff between checking things that you've checked already and making better use of this time to write a better paper.