Live data from Hacker News

The 'No True Programmer' Fallacy

jacquesmattheij.com

51–60 of 70 posts

Re: The 'No True Programmer' Fallacy

#51
post #48
post #36

Earlier quoted context omitted.

If you will learn NLP as a side effect, I would avoid the book as a plague. If the author puts such a big lie as NLP, who knows what else is false in the book?

I do not know what experiences you made so that you think NLP is "such a big lie". It is a toolbox of methods which generally do work. Many of them are applied in behavioral therapy, which is proved to have an outstanding effect on depression (as one example). The nice thing is that you do not have to study psychology for that, but have simple methods, which are almost common sense, to deal and/or transform most psyc…

"..generally do work".

If that were so, then could you explain the section at https://en.wikipedia.org/wiki/Neuro-linguistic_programming#S... ? Pulling three quotes from it, "controlled trials failed to show any benefit from NLP", "Numerous literature reviews and meta-analyses have failed to show evidence for NLP's assumptions or effectiveness as a therapeutic method", and "Among the reasons for considering NLP a pseudoscience are that evidence in favor of it is limited to anecdotes and personal testimony".

What evidence was enough for you to conclude that the methods "generally do work"?

Re: The 'No True Programmer' Fallacy

#52
post #22

Do we call people that put together chairs structural engineers? Do we call baristas chemists? Baseball players physicists? Words exist for a reason, and using a spreadsheet does not make you a programmer. However I can already see that programmer is being diluted to mean anyone proficient with a computer. What term should we use for people that can write software? 'Software engineer' feels like overkill for writing…

Software developer is more than good enough in my opinion. However I've encountered two types of programmers, there are those who contribute new ideas, new algorithms, solve new problems, find new, more efficient, safer solutions to existing problems, etc. Then you have the other type who just write glue code between different frameworks and libraries which someone else wrote, they never write anything themselves but download third party packages from Github or whatever to accomplish their tasks. I wouldn't call the latter "software development", coding or scripting would be a better word and yes anybody who is proficient with a computer can code or script but not everybody can be a software developer. It's similar to how assembling your own flashlight, or even assembling your own computer does not make you an electrical engineer.

"Real programmers do not reinvent the wheel" is an overused cliche which has been diluted to mean anyone who does anything other than gluing together third party libraries. The cliche itself does not make sense since the wheel is one of the reinvented technologies in existence, if we hadn't reinvented the wheel we would still be using logs.

Re: The 'No True Programmer' Fallacy

#53

Seems to me the article is fighting some kind of giant strawman. I myself haven't experiences this kind of attitude. I really don't get what the point is, that some people have ego problems and hold pissing contests to see who is the most "awesome programmer"? Well that's a different thing, kind of "who's the best". Well for sure if you ask people in an office what their job is then I don't think that a banker who cr…

Clearly a True Programmer would believe what the original poster says. You're not a True Programmer.

Re: The 'No True Programmer' Fallacy

#54
post #51
post #48

Earlier quoted context omitted.

I do not know what experiences you made so that you think NLP is "such a big lie". It is a toolbox of methods which generally do work. Many of them are applied in behavioral therapy, which is proved to have an outstanding effect on depression (as one example). The nice thing is that you do not have to study psychology for that, but have simple methods, which are almost common sense, to deal and/or transform most psyc…

"..generally do work". If that were so, then could you explain the section at https://en.wikipedia.org/wiki/Neuro-linguistic_programming#S... ? Pulling three quotes from it, "controlled trials failed to show any benefit from NLP", "Numerous literature reviews and meta-analyses have failed to show evidence for NLP's assumptions or effectiveness as a therapeutic method", and "Among the reasons for considering NLP a pse…

I do not know how NLP is taught right know, however I'll explain how I come to that conclusion with a reference to the mentioned book from steve andreas.

One of the basic tools of NLP is what is called "anchoring". An anchor is an association in terms of action->response. One example lots of people can associate with is: You hear a song and have instantly a mood change. You could feel very happy, because you remember meeting your spouse for the first time when this song was played. Or maybe you feel bad because something bad happened when you heared that song and associate the bad feelings with that song.

One way to break away from a specific "action->response" pair is by using a pattern interrupt and overwriting the usual response with another response. You do that regularly yourself, for example when you learned a false fact and correct it afterwards.

Of course you can also create new associations. You probably do not have any strong feelings associated with a song when hearing it for the very first time, but you can associate something with it.

Anchoring, pattern interupts and overwriting old anchors is heavily used in behavioral therapy, and is also the basic concept for a lot methods in NLP. For me the parallels to behavioural therapy are why I conclude "generally do work".

I think the bad reputation NLP is receiving is because of "hype-riders", people who want to rip the help-seeking people off their money. And as you do not have to study psychology to practice NLP (or say you can do NLP), charlatants have an easy entry.

In Steve Andreas book, you learn how to rewrite old anchors effectiveley, specifically the ones that make up your identity, the "how you know you are you".

Example Scenario:

Current situation: Suppose you think you are bad at maths and that is simply your personality, that's why you will never solve any math problem in reasonable time. Effect: You will not even try to solve math problems or will not have fun doing it, thus taking more time or limiting yourself in other areas as well as panicing as soon as you see a math problem.

How you change that? ->

1) Think about how you know you are bad at maths.

2) Do you imagine one or multiple situations where you were bad at maths? Are these pictures or video? How are they organized in your head? In color or black/white? Big/small?

3) Now take every example in your head that support that belief and change the modalities. Already at that point you will notice with each transformed example, that you feel less bad about maths.

4) Next you build positive images around maths in the same way as before, thinking about situations where you were good at maths and transform your images to the same modalities the bad images were before.

5) Pro step: Build values for yourself such as: intelligent, resilient, motivated and associate these values with situations were they helped you in solving math problems or other problems. This would decouple your identity from "maths" and help you not fall into panic as soon as you encounter a math problem or any other problem you think you are not good at (remember, you are intelligent, resilient and motivated?)

I do not know if specifically this technique with exactly these steps is used in todays scientific therapy. However, it is based on anchoring/overwriting old anchors and I do not see much novelty there.

Re: The 'No True Programmer' Fallacy

#55
post #7

Earlier quoted context omitted.

Read some of Norvigs code and tell me he's not a programmer, his stuff is some of the most elegant and efficient that I've ever seen.

Putting programmers on a pedestal fits with the YC culture of marketing via lionization of technology personalties, but it's counter-productive if your aim is to actually grow in the field. Hero worship blinds you to two simple facts, in almost all cases: - They're not doing anything you couldn't also do if you worked at it as hard as they have. - Like everyone, they're fallible, making mistakes, and may even not des…

I'm not putting him on a pedestal at all. He's by objective standards simply one of the best programmers that I know of, just like Usain Bolt is an extremely fast runner.

Re: The 'No True Programmer' Fallacy

#56

Earlier quoted context omitted.

Putting programmers on a pedestal fits with the YC culture of marketing via lionization of technology personalties, but it's counter-productive if your aim is to actually grow in the field. Hero worship blinds you to two simple facts, in almost all cases: - They're not doing anything you couldn't also do if you worked at it as hard as they have. - Like everyone, they're fallible, making mistakes, and may even not des…

I'm sorry to disagree but you're wrong. There are better programmers than I, that no matter how hard I work I will never be as good as. Not everyone can be a Mozart or Mozart wouldn't be special.

It's not as black and white as you put. Linus may be great at writing operating systems kernels, he may be better than you or I are at it but he wouldn't be as good at designing machine learning artificial intelligence algorithms, if we had to put him in the shoes of the developers at Google which run the world's largest search engine he wouldn't now be objectively the best, neither would the Google developers do to well if they had to maintain the Linux kernel. Same thing with Mozart, he is one of the best composers in his genre of music but there are other genres as well.

Re: The 'No True Programmer' Fallacy

#58
post #54
post #51

Earlier quoted context omitted.

"..generally do work". If that were so, then could you explain the section at https://en.wikipedia.org/wiki/Neuro-linguistic_programming#S... ? Pulling three quotes from it, "controlled trials failed to show any benefit from NLP", "Numerous literature reviews and meta-analyses have failed to show evidence for NLP's assumptions or effectiveness as a therapeutic method", and "Among the reasons for considering NLP a pse…

I do not know how NLP is taught right know, however I'll explain how I come to that conclusion with a reference to the mentioned book from steve andreas. One of the basic tools of NLP is what is called "anchoring". An anchor is an association in terms of action->response. One example lots of people can associate with is: You hear a song and have instantly a mood change. You could feel very happy, because you remember…

Once upon a time we (that is, general medical consensus in the US) thought that milk would help treat ulcers, and that ulcers were mostly due to diet or lifestyle (overwork, smoking, etc). There was a model of how milk would be a good treatment, and doctor would recommend milk to patients with ulcers.

Starting a few decades ago, we found that "the vast majority of ulcers are caused by H pylori infection" (quote from http://www.ulcerresourcecenter.com/ulcers-and-milk ), and can be treated with drug therapy.

I point this out as an example of how even if there is a reasonable mechanistic model of how a therapy might work, which is widely believed to be true, that doesn't mean it is true. The Wikipedia link I posted gives reference to attempts to verify if NLP is a useful treatment, and concluded that it isn't.

As for "pattern interrupts and overwriting old anchors is heavily used in behavioral therapy", if you search for '"pattern interrupts" "behavioral therapy"' on Google there are very few hits. Duck Duck Go has even fewer. This suggests that "pattern interrupts" is not a term used in behavioral therapy. If you further look at http://www.ncbi.nlm.nih.gov/gquery/?term=%22pattern+interrup... you'll see that "pattern interrupt" is not commonly used, with only 214 matches in PubMed. A spot check shows only a handful used in the sense you mean.

Therefore, the term "pattern interrupts" is likely not the term used in behavioral therapy. You are, I believe, making a statement about a homologous treatment, perhaps operant conditioning? But operant conditioning, and its application, predate NLP, so I don't think it's reasonable to attribute its effectiveness to NLP, nor would it be novel in NLP if Skinner discussed it in the 1950s.

("Operant conditioning" has 18356 matches in PubMed, and the first 7 papers all use it in the cognitive psychology sense.)

Re: The 'No True Programmer' Fallacy

#59

Seems to me the article is fighting some kind of giant strawman. I myself haven't experiences this kind of attitude. I really don't get what the point is, that some people have ego problems and hold pissing contests to see who is the most "awesome programmer"? Well that's a different thing, kind of "who's the best". Well for sure if you ask people in an office what their job is then I don't think that a banker who cr…

>Seems to me the article is fighting some kind of giant strawman. I myself haven't experiences this kind of attitude.

For real? Because it shows up in HN all the time.

E.g. for what he writes:

>The IT world is rife with this ‘no true programmer’ nonsense, the ‘real’ programmers are the ones that have mastered ‘x’ (insert name of arcane and difficult to use programming language here), the rest is still stuck on ‘y’ (insert accessible and easy to use language here). The latter of course aren’t real programmers, if they were then they would get it.

How's that not an example of "the Blub" reasoning? http://www.paulgraham.com/avg.html

Re: The 'No True Programmer' Fallacy

#60
post #58
post #54

Earlier quoted context omitted.

I do not know how NLP is taught right know, however I'll explain how I come to that conclusion with a reference to the mentioned book from steve andreas. One of the basic tools of NLP is what is called "anchoring". An anchor is an association in terms of action->response. One example lots of people can associate with is: You hear a song and have instantly a mood change. You could feel very happy, because you remember…

Once upon a time we (that is, general medical consensus in the US) thought that milk would help treat ulcers, and that ulcers were mostly due to diet or lifestyle (overwork, smoking, etc). There was a model of how milk would be a good treatment, and doctor would recommend milk to patients with ulcers. Starting a few decades ago, we found that "the vast majority of ulcers are caused by H pylori infection" (quote from…

Sorry about the bad terminilogy, I have my input from a psychology student. Maybe someone can clarify about the correct terminology.

I researched more, and at least the wikipedia arguments against NLP are critisized heavily, too. http://realpeoplepress.com/blog/research-in-nlp-neurolinguis...

There is an organisation that aims for NLP to be recognized scientifically http://www.researchandrecognition.org/

At this point, you are right that there is not much scientific evidence. I hope for the research group that it produces some proper results soon

Post reply on HN