4 ms·

No best practices: The fallacy of methodologies

25 points by aditya 17y ago

pvg 17y ago
After 4 years at Accenture with a responsibility for 'process' in one's title and enough free time to gain a familiarity with Dilbert cartoons but not enough to consider the possibility that perhaps people have thought deeply about these problems before. Maybe they even wrote papers. http://en.wikipedia.org/wiki/No_silver_bullet http://en.wikipedia.org/wiki/No_silver_bullet And they did, except better. They even came up with that title, except better.
- swombat 17y ago
  There are no such thing as old jokes, my friend. To a newborn, every joke is a world première.
  - pvg 17y ago
    I sort of expect someone who's held a job in tech and self-describes as a CTO of a start-up to actually know some of the good old jokes. My expectations of newborns, are, as you suggest, less stringent.
    - swombat 17y ago
      I tend to write for my readers as opposed to writing for myself. I've rarely found out something that I didn't already know, from reading one of my own blog posts! It seems you got this impression from reading this post that this is a new realisation for me - it's not, I just happened to write about it now. I'll try to make that clearer in future posts.
      - pvg 17y ago
        It seems you got this impression from reading this post that this is a new realisation for me - it's not I didn't know you were the author of the post. The impression I got was that you were blithely unaware of the basic literature on the subject. And really, if you're going to make a post called 'No best practices' and don't reference Brooks at all, it's hard to come up with explanations other than 'plagiarism' or 'ignorance'. In fact, while I have you on the line, let me ask you directly - had you read the Brooks essay?
        swombat 17y ago
        Yes, I have. I'm not quite sure how "someone wrote a related paper 30 years ago" prevents me from writing a blog post on a topic that I want to write about, though.
        pvg 17y ago
        Well the thing is, someone actually wrote a seminal, much quoted paper 30 years ago. You wrote a "related" platitude-laden derivative and borrowed the title 30 years later. Nothing prevents you from doing that at all. Common decency suggests you should reference the original, if you were, as you claim, well aware of it. That's all.
        swombat 17y ago
        Please note that I said I have read that paper. That doesn't mean I was thinking of that paper when I wrote this blog post. Pardon me for not having absolutely everything I have read in mind whenever I write a blog post. Anyway, this conversation is incredibly tedious. I'm done with this thread.
        pvg 17y ago
        You noted it here, not in your post. Citing references and work your work is based on is certainly quite tedious. Sometimes doing the right thing is tedious. Doesn't make it any less right.
- Confusion 17y ago
  I'll respond at this level, although the point was only fleshed out in the subsequent discussion with swombat. Brooks was writing about a different thing. No Silver Bullet is about techniques for writing code. He specifically mentions programming languages, OOP, AI, graphical programming, etc. as examples of 'silver bullet candidates'. His arguments are based on, among other things, the complexity and changeability of the 'software entities', and the main point is that it's hard to get the code right. This all has very little to do with the methodology one uses to organise the writing of code and all the things that need to happen to allow the code to be written. It has become common for people to broaden the meaning of 'No silver bullet' and use it, appropriately or not, in various other contexts. However, in those circumstances, no tribute to Brooks is required if you, for instance, are explaining how there is no single method to make sure your toddler doesn't poke out your cat's eye. Your indignation over the referenceless unoriginality of this blogpost is really unsubstantiated.
derefr 17y ago
> When someone picks out the bits that might work for a specific project but doesn’t succeed, the blame that they receive is "you didn’t follow the whole methodology". I wouldn't say "blame", really; it's more that you shouldn't say you failed while doing X, if you weren't really doing X. That casts a bad light on X, whether it was X's fault or yours that the project failed. Just say you were adopting an organic/syncretic methodology, and leave it at that.
rbanffy 17y ago
> the methodology is rarely to blame for a failed project Truth is, it's always the person in charge who is to blame. That and natural disasters, but those are far less frequent.
edw519 17y ago
OP's post says more about the consulting industry than the best practices they promote. Let me explain... I was brought in by a Big 5 firm to rescue some of their clusterfucks. One project was 18 months old and one year behind. Another was "90% complete" for the past year. They gave me a laptop with their own systems analysis / project management app on it. Everything had to go through it. After 2 days it was clear that this "system" was cause of most of their problems. Why? Because no one thought any more. They just plugged everything into the app and submitted the reports every Friday whether anything made sense or not. I showed the app to a friend of mine who was an expert consultant (not Big 5!). It was clear to him why they did this. He said, "The sole purpose of 'best practices' and tools like this are to make people who don't know what they're doing appear as if they did. They pay people $50 per hour and bill them at $300 per hour. The difference, after expenses, goes to the partners. That's their business model. It's that simple. So, forget about best practices and just do your fucking job. You'll eat them alive." Thanks, Tom. I've been doing that ever since. You were right.
- notauser 17y ago
  What he actually said was that his consulting company told him not to use all of the process all of the time but rather to pick and chose the bits that were appropriate. I'm not saying that you are lying about your experience, or that I have much respect for the big consulting companies or their business models, but your comments misleadingly characterize what he posted. (In my experience the big companies I have worked for had better processes, and better guidance on how to apply them. But they had to have that to cater for some of their other issues.)
j_baker 17y ago
I agree with this post 100%. However, there is one thing I'd like to note. While I agree that one should not be blindly obedient to methodologies, one should make an informed decision about what parts to pick and choose. Unless you've actually tried the whole thing, chances are you won't be informed enough to make a good decision. Otherwise, it's way too easy to look at some parts of the methodology and say "well that just sounds stupid" without really seeing the benefit. I'm not saying everyone needs to drink the Koolaid at first, but I do feel that it's better to err on the Koolaid side until you have a better understanding of what works and what doesn't.
halffull 17y ago
It's method, not methodology. The former refers to a specific process or procedure employed to accomplish a task while the latter describes the study of such processes and procedures. The discussion we are engaging in here is a form of methodology; the things being discussed are methods. I don't tend to get hung-up on usage, but it really is shocking to see how many articles, blog posts, and comments written on this subject make this mistake.
- nollidge 17y ago
  Merriam-Webter [1]: 1 : a body of methods, rules, and postulates employed by a discipline : a particular procedure or set of procedures 2 : the analysis of the principles or procedures of inquiry in a particular field You're speaking of definition (2), but OP is using (1) properly, in my estimation. [1] http://www.merriam-webster.com/dictionary/methodology http://www.merriam-webster.com/dictionary/methodology
  - halffull 17y ago
    USAGE NOTE Methodology can properly refer to the theoretical analysis of the methods appropriate to a field of study or to the body of methods and principles particular to a branch of knowledge. In this sense, one may speak of objections to the methodology of a geographic survey (that is, objections dealing with the appropriateness of the methods used) or of the methodology of modern cognitive psychology (that is, the principles and practices that underlie research in the field). In recent years, however, methodology has been increasingly used as a pretentious substitute for method in scientific and technical contexts, as in The oil company has not yet decided on a methodology for restoring the beaches. People may have taken to this practice by influence of the adjective methodological to mean "pertaining to methods." Methodological may have acquired this meaning because people had already been using the more ordinary adjective methodical to mean "orderly, systematic." But the misuse of methodology obscures an important conceptual distinction between the tools of scientific investigation (properly methods) and the principles that determine how such tools are deployed and interpreted. http://www.answers.com/main/ntquery?s=methodology http://www.answers.com/main/ntquery?s=methodology
Mz 17y ago
It's always good to distinguish between rules and judgment. Rules should assist in good judgment, not supplant it and push it out. This is essentially the root of much criticism of government and other large bureaucracies: That you often can't do things which make sense because it is against some silly rule. Start-ups have a chance to put good judgment to use without being hampered by entrenched, illogical, outdated or counterproductive rules. Hopefully, they will take that opportunity and run with it and then resist temptation to implement a bunch of bad rules (which is basically the topic of discussion of several recent threads here along the lines of "Don't go punishing all of your customers over the actions of one/a few").

25 points by aditya 17y ago

pvg 17y ago

After 4 years at Accenture with a responsibility for 'process' in one's title and enough free time to gain a familiarity with Dilbert cartoons but not enough to consider the possibility that perhaps people have thought deeply about these problems before. Maybe they even wrote papers. http://en.wikipedia.org/wiki/No_silver_bullet http://en.wikipedia.org/wiki/No_silver_bullet And they did, except better. They even came up with that title, except better.

swombat 17y ago
There are no such thing as old jokes, my friend. To a newborn, every joke is a world première.
- pvg 17y ago
  I sort of expect someone who's held a job in tech and self-describes as a CTO of a start-up to actually know some of the good old jokes. My expectations of newborns, are, as you suggest, less stringent.
  - swombat 17y ago
    I tend to write for my readers as opposed to writing for myself. I've rarely found out something that I didn't already know, from reading one of my own blog posts! It seems you got this impression from reading this post that this is a new realisation for me - it's not, I just happened to write about it now. I'll try to make that clearer in future posts.
    - pvg 17y ago
      It seems you got this impression from reading this post that this is a new realisation for me - it's not I didn't know you were the author of the post. The impression I got was that you were blithely unaware of the basic literature on the subject. And really, if you're going to make a post called 'No best practices' and don't reference Brooks at all, it's hard to come up with explanations other than 'plagiarism' or 'ignorance'. In fact, while I have you on the line, let me ask you directly - had you read the Brooks essay?
      - swombat 17y ago
        Yes, I have. I'm not quite sure how "someone wrote a related paper 30 years ago" prevents me from writing a blog post on a topic that I want to write about, though.
        pvg 17y ago
        Well the thing is, someone actually wrote a seminal, much quoted paper 30 years ago. You wrote a "related" platitude-laden derivative and borrowed the title 30 years later. Nothing prevents you from doing that at all. Common decency suggests you should reference the original, if you were, as you claim, well aware of it. That's all.
        swombat 17y ago
        Please note that I said I have read that paper. That doesn't mean I was thinking of that paper when I wrote this blog post. Pardon me for not having absolutely everything I have read in mind whenever I write a blog post. Anyway, this conversation is incredibly tedious. I'm done with this thread.
        pvg 17y ago
        You noted it here, not in your post. Citing references and work your work is based on is certainly quite tedious. Sometimes doing the right thing is tedious. Doesn't make it any less right.
Confusion 17y ago
I'll respond at this level, although the point was only fleshed out in the subsequent discussion with swombat. Brooks was writing about a different thing. No Silver Bullet is about techniques for writing code. He specifically mentions programming languages, OOP, AI, graphical programming, etc. as examples of 'silver bullet candidates'. His arguments are based on, among other things, the complexity and changeability of the 'software entities', and the main point is that it's hard to get the code right. This all has very little to do with the methodology one uses to organise the writing of code and all the things that need to happen to allow the code to be written. It has become common for people to broaden the meaning of 'No silver bullet' and use it, appropriately or not, in various other contexts. However, in those circumstances, no tribute to Brooks is required if you, for instance, are explaining how there is no single method to make sure your toddler doesn't poke out your cat's eye. Your indignation over the referenceless unoriginality of this blogpost is really unsubstantiated.

derefr 17y ago

> When someone picks out the bits that might work for a specific project but doesn’t succeed, the blame that they receive is "you didn’t follow the whole methodology". I wouldn't say "blame", really; it's more that you shouldn't say you failed while doing X, if you weren't really doing X. That casts a bad light on X, whether it was X's fault or yours that the project failed. Just say you were adopting an organic/syncretic methodology, and leave it at that.

rbanffy 17y ago

> the methodology is rarely to blame for a failed project Truth is, it's always the person in charge who is to blame. That and natural disasters, but those are far less frequent.

edw519 17y ago

OP's post says more about the consulting industry than the best practices they promote. Let me explain... I was brought in by a Big 5 firm to rescue some of their clusterfucks. One project was 18 months old and one year behind. Another was "90% complete" for the past year. They gave me a laptop with their own systems analysis / project management app on it. Everything had to go through it. After 2 days it was clear that this "system" was cause of most of their problems. Why? Because no one thought any more. They just plugged everything into the app and submitted the reports every Friday whether anything made sense or not. I showed the app to a friend of mine who was an expert consultant (not Big 5!). It was clear to him why they did this. He said, "The sole purpose of 'best practices' and tools like this are to make people who don't know what they're doing appear as if they did. They pay people $50 per hour and bill them at $300 per hour. The difference, after expenses, goes to the partners. That's their business model. It's that simple. So, forget about best practices and just do your fucking job. You'll eat them alive." Thanks, Tom. I've been doing that ever since. You were right.

notauser 17y ago
What he actually said was that his consulting company told him not to use all of the process all of the time but rather to pick and chose the bits that were appropriate. I'm not saying that you are lying about your experience, or that I have much respect for the big consulting companies or their business models, but your comments misleadingly characterize what he posted. (In my experience the big companies I have worked for had better processes, and better guidance on how to apply them. But they had to have that to cater for some of their other issues.)

j_baker 17y ago

I agree with this post 100%. However, there is one thing I'd like to note. While I agree that one should not be blindly obedient to methodologies, one should make an informed decision about what parts to pick and choose. Unless you've actually tried the whole thing, chances are you won't be informed enough to make a good decision. Otherwise, it's way too easy to look at some parts of the methodology and say "well that just sounds stupid" without really seeing the benefit. I'm not saying everyone needs to drink the Koolaid at first, but I do feel that it's better to err on the Koolaid side until you have a better understanding of what works and what doesn't.

halffull 17y ago

It's method, not methodology. The former refers to a specific process or procedure employed to accomplish a task while the latter describes the study of such processes and procedures. The discussion we are engaging in here is a form of methodology; the things being discussed are methods. I don't tend to get hung-up on usage, but it really is shocking to see how many articles, blog posts, and comments written on this subject make this mistake.

nollidge 17y ago
Merriam-Webter [1]: 1 : a body of methods, rules, and postulates employed by a discipline : a particular procedure or set of procedures 2 : the analysis of the principles or procedures of inquiry in a particular field You're speaking of definition (2), but OP is using (1) properly, in my estimation. [1] http://www.merriam-webster.com/dictionary/methodology http://www.merriam-webster.com/dictionary/methodology
- halffull 17y ago
  USAGE NOTE Methodology can properly refer to the theoretical analysis of the methods appropriate to a field of study or to the body of methods and principles particular to a branch of knowledge. In this sense, one may speak of objections to the methodology of a geographic survey (that is, objections dealing with the appropriateness of the methods used) or of the methodology of modern cognitive psychology (that is, the principles and practices that underlie research in the field). In recent years, however, methodology has been increasingly used as a pretentious substitute for method in scientific and technical contexts, as in The oil company has not yet decided on a methodology for restoring the beaches. People may have taken to this practice by influence of the adjective methodological to mean "pertaining to methods." Methodological may have acquired this meaning because people had already been using the more ordinary adjective methodical to mean "orderly, systematic." But the misuse of methodology obscures an important conceptual distinction between the tools of scientific investigation (properly methods) and the principles that determine how such tools are deployed and interpreted. http://www.answers.com/main/ntquery?s=methodology http://www.answers.com/main/ntquery?s=methodology

Mz 17y ago

It's always good to distinguish between rules and judgment. Rules should assist in good judgment, not supplant it and push it out. This is essentially the root of much criticism of government and other large bureaucracies: That you often can't do things which make sense because it is against some silly rule. Start-ups have a chance to put good judgment to use without being hampered by entrenched, illogical, outdated or counterproductive rules. Hopefully, they will take that opportunity and run with it and then resist temptation to implement a bunch of bad rules (which is basically the topic of discussion of several recent threads here along the lines of "Don't go punishing all of your customers over the actions of one/a few").