Death by Agile Fever

I cover a lot of the same dysfunctions here: postagilist.wordpress.com/2011/04/06/rhetorical...

Despite the title, this article is really not about Agile at all. It is really about hype and incompetence and the misuse of the same.

Try taking every instance if the word Agile and replacing it with "Design Patterns", "Technical Debt", or even "UML" -- the entire article still makes perfect sense.

I'm with you on Design Patterns, since it's run by essentially the same people (Agilists).

Technical debt and UML? Really? Is there a UML manifesto? A certified Technical Debt master?

Few clans and cliques are as closed minded and insular as the Agile movement.

The fact that there is a movement at all (fueled by a manifesto no less) makes it seperatist by nature.... Saying that everything is plagued to the same degree as the Agile movement when it comes to parrotism and separatism, etc, and willful ignorance, does a disservice to these other fields

PA

I just picked three topics at random just to illustrate how the text of the article could apply to anything -- I wasn't saying those examples were necessarily valid.

Yes, you are using a straw argument, to try to shift the blame away from agile.

Agile is guilty of these sins, and it's simply untrue that everything else suffers from them in similar amounts.

Sorry, "everyone does it" is not only not true, it's a very lame attempt to deflect valid criticism from a very guilty party (the Agile movement).

The Agile Apologists pull these stunts constantly -- perhaps that can be added to the list.

Nice try.

If the Agilists would learn, sometime, instead of constantly trying to ignore and shift the blame, perhaps the movement woulnd't be still riddled with all these dysfunctions.

PA

I usually find myself a proponent of questioning Agile practices and examining the real value of Agile. (but I honestly believe that Agile practices are usually the best option) I expected that this article would be aligned to my thinking but I was wrong. This article should be re-named the 'Methodology Fevers'. In my opinion, all of the fevers can be observed by following any methodology too closely and without a critical eye. I didn't see one fever that was only applicable to Agile.

It is unfortunate that the author chose to centre in one one methodology when the same fevers apply to all. That includes the fervent criticism of Agile.

Wasn't that topic already covered in the comments above?

I thought I'd share a list I found on wikipedia recently. en.wikipedia.org/wiki/Agile_software_developmen...

Another criticism is that Agile’s drive to impose itself into business systems is cultish in its nature. Faith-based similarities include:

Agile claims absolute universality, of being applicable to all levels of business, and useful to all members in any given organization.

Agile claims infallibility (that its technique is absolutely correct as-is, and not to be changed or questioned).

Agile requires meetings that involve ritual activity (“planning poker”, for example).

Agile requires system-specific jargon for comprehension (Kanban, story points, scrum, epic, etc.), terms which are not readily understood by the layman or un-initiated.

Agile attempts to create an ideological “monopoly of method”, excluding alternatives, competitive models, and any mixture or dilution of its core principles.

Agile seeks to become closely allied with the “heads of state” (business leaders, in this case) to gain favor, authority, and reinforcement of its claims.

Agile employs professional, full-time clergy (i.e. “Scrum Masters”) who alone possess the appropriate credentials of education and formal ordination (e.g. Agile certifications).

Agile attempts to gain new members through viral reproduction and the socialization of the un-initiated into the ranks.

Agile tries to minimize diversity of opinion by accepting different views within the system only rather than through the formation of new methodologies or alternatives

PA

That link is en.wikipedia.org/wiki/Agile_software_developmen...

I think we may need to agree to disagree. Can you repost the Wikipedia link? I couldn't follow it.

I am DBA/PM that has been on traditional and waterfall projects for over 7 year apiece. (traditional longer...)

I first must state that I also disagree with Scrum and do not follow Scrum in how I prefer to conduct Agile projects.

All of your other comments on Agile are, in my humble opinion, incorrect. Are they your own comments or are they from a specific source?

Again to illustrate my point, I can create a set of opposing statements about waterfall:

- Traditional Waterfall claims absolute universality, of being applicable to all levels of business, and useful to all members in any given organization.

- Traditional Waterfall claims infallibility (that its technique is absolutely correct as-is, and not to be changed or questioned).

- Traditional Waterfall requires meetings that involve ritual activity (For example, Specification sign-off rituals and project status meetings)

- Traditional Waterfall requires system-specific jargon for comprehension (WBS, Earned Value, Use Case), terms which are not readily understood by the layman or un-initiated.

- Traditional Waterfall attempts to create an ideological “monopoly of method”, excluding alternatives, competitive models, and any mixture or dilution of its core principles.

- Traditional Waterfall seeks to become closely allied with the “heads of state” (IT Leaders, in this case) to gain favor, authority, and reinforcement of its claims.

- Traditional Waterfall employs professional, full-time clergy (i.e. PMI-ceritifed Project Managers) who alone possess the appropriate credentials of education and formal ordination (e.g. Traditional Waterfall certifications).

- Traditional Waterfall attempts to gain new members through viral reproduction and the socialization of the un-initiated into the ranks. Specifically the inclusion of PMI certification as criteria for new hires.

- Traditional Waterfall tries to minimize diversity of opinion by accepting different views within the system only rather than through the formation of new methodologies or alternatives

My main point is that promoting these extreme points of view is misleading, incorrect, and not constructive. Following either the waterfall or agile methodology too closely will cause the same problems.

Neither is totally evil, neither is fully correct.

What you are doing, is another well known antipattern. You are building up a strawman of waterfall and then bravely knocking it down.

As do most agilists -- I cover that in my post here: postagilist.wordpress.com/2012/06/13/the-perenn...

I would grant you that traditional business (so called waterfall) does try to align itself with heads of state but that is the only similarity.

Are there 2 day waterfall master courses? Is there a waterfall movement? Is waterfall inflexible? No, except for the straw versions (see above).

As per the wikipedia link it should appear soon (I pasted it into another message that went into mod Q) in the meantime just google for agile criticism wikipedia and it will take you right to it.

PA

As PostAgilista already stated, your "issue" has already been addressed in his previous post. If what you were saying was true, I would have written this article long ago. With all due respect, my guess is that you are really a Fragilista trying to masquerade as an objective reader.

Thanks for the reply Alex. We will also have to agree to disagree. Let me just say I don't believe that the issue has been dealt to my satisfaction. Just because you state that it has doesn't mean it is true. Also trying to name me a Fragilista is slightly disappointing. I was trying to put forth an opinion for discussion, and you have responded in a similar way to the Agile purists you criticize. Who is now not open to alternate approaches and diversity of opinion?

As PostAgilist mentions in his blog, we should encourage discourse in the following manner:

1) Respecting the opposition

2) Realize that there are multiple paths to success for any given problem

3) Decisions will be made based on reason and practicality, not on academic, philosophical, bullying, emotion, demonizing or any of the above mentioned anti patterns.

4) Argue your case — nicely — supported by facts.

So further to point 4, what are your facts that Agile projects are more susceptible to these issues? What Agile projects have you been involved with? How large were they and what did they deliver? Were they project or product development? What Agile or XP practices did they follow? What traditional practices did they follow?

Terry

Terry, I agree with completely, and I go back to my original point. This is really about hype and misuse.

Are there people in the Agile community that are guilty of the types of hype and misuse described - certainly there are!

Is that problem universal with everyone in the Agile community - absolutely not!

Overall, is the hype problem a serious issue Agile community - this one is a lot more subjective. I tend to think not, but it probably heavily depends on precisely which subgroups you interact with the most.

In any case, it is helpful to be fact-based. It is not helpful to engage in ad hominem attacks.

Yes, you are using a straw argument, to try to shift the blame away from agile.

Agile is guilty of these sins, and it's simply untrue that everything else suffers from them in similar amounts.

Sorry, "everyone does it" is not only not true, it's a very lame attempt to deflect valid criticism from a very guilty party (the Agile movement).

The Agile Apologists pull these stunts constantly -- perhaps that can be added to the list.

Nice try.

If the Agilists would learn, sometime, instead of constantly trying to ignore and shift the blame, perhaps the movement woulnd't be still riddled with all these dysfunctions.

PA

I am confused. Is it the Agilist (people) that is causing the Fevers and not Agile in the original post?

Agile claims absolute universality...
Agile claims infallibility...
Agile requires meetings that involve ritual activity...
Agile requires system-specific jargon for comprehension...
etc.
PA

That list was from Wikipedia.

As per jargon, I do think there is a problem with jargon the way it is frequently used in agile...

I cover that in this post postagilist.wordpress.com/2012/06/12/orwell-on-...

I do believe that everyone should be open to ideas no matter their origin so agree with you on that

PA

That's all true and I look forward to a good discussion.

However I also list as an antipattern (#7) asking for proof of others without supplying your own first.

If you supply your own evidence in that area, then it's fair to ask the same of others.

Agilists constantly demand proof of this or that, from others, but supply no proof that agile itself works. The recent Voke report surverying 200 companies strongly suggests that it doesn't.

PA

Thanks for the reply PA.

I would be happy to share my own personal experiences with Agile and Traditional. I find surveys for both the pro and anti Agile perspective less valuable as they don't allow people to provide context or examples with their answers.

As to what is fair to ask in the way of proof, I would like to point towards the book 'Candle in the Dark' by late, great Carl Sagan. Carl Sagan pointed out correctly that it is the reponsibility of the person promoting the idea to provide the facts. It is not the responsibility of the opponent to provide facts to disprove the idea. Typically people in pseudo-science would state that if their proposed idea is untrue, then science should be able to provide facts to clearly disprove the proposed idea. This principle is core to the objective scientific method. (It is no surprise that people propose these positions, one can look at Euler's last theorem as an example of how much harder it is to prove negative postulation than positive postulation)

This article proposed that these fevers are more predominant in the Agile methodology than other methodologies. The primary responsibility is not mine to prove that this statement is untrue.

I also agree that Agile purists also propose statements which are untrue of waterfall without any facts as well. I would be equally critical of a pro-Agile article.

Terry

First of all, I think the issues that Bell raises are self evident.

It doesn't take a lot to find the parrotism, etc, found in agile.

But more importantly, Bell is rejecting Agile, in many ways, because there is NO PROOF that agile is effective.

Since you feel that, someone proposing something should prove it, then, the agile field should have been proving things all along.

They haven't.

So if Bell comes out and says, oh, there is no proof that Agile works, there is no proof that Scientology works, then, it would be unrealistic to say that he should then have to prove that Scientology does not work.

In any case Bell is listing fevers, not listing the amounts that they occur in the population.

Clearly the fevers exist, to what extent is not a topic of this article, and that is the data you are asking him to provide.

It seems nonsensical.

I think there is agreement, from Curt, at least, that these fevers DO exist in the agile community, so let's get on subject here and talk about how to eradicate the fevers from the agile community, not get into red herring distractions as to whether other communities also have a few of these fevers, or what the exact population of agilists that suffer from them are.

PA

Agilists constantly demand proof of this or that, from others, but supply no proof that agile itself works. The recent Voke report surverying 200 companies strongly suggests that it doesn't.

No, I'm not aware of how seriously flawed that survey was... I would post links but I keep getting mod Q'd for posting links.

I am aware that the blog of the person you posted to, is heavily biased in favor of agile, and went on a rant against Voke's conclusions.

Voke's sampling of 200 companies is the largest independent analysis that I'm aware of, and they report that only 2% of the companies could identify identifiable success with agile.

Naturally the Fragilistas were up in arms about it and wrote articles like you linked to, but that doesn't change the conclusions of Voke in that regard.

Since you are making an (unsupported) assertion, that, the Voke report, is somehow stastically flawed, can you support your assertion in that area ?

How specifically was it flawed?

Agilists always claim that it's an "empirical" process yet when the empiracal data is not in their favor they automatically say that the empiracal data is not valid.

No surprise there -- I think that's part of Cook the Books Fever right?


PA

Since I couldn't include this much text in my previous response:


In any case here is a summary of the Voke report (google adtmag voke to find it)

---

The survey results include the following (verbatim from the report):

Out of over 200 survey participants, we received only four detailed comments describing success with Agile.

40% of participants that use Agile did not identify a benefit.


Some participants (7%) also noted that Agile developers were happy due to less future planning and less documentation.

Survey participants report that developers use the guise of Agile to avoid planning and to avoid creating documentation required for future maintenance.

Be aware that the Agile movement might very well just be either a developer rebellion against unwanted tasks and schedules or just an opportunity to sell Agile services including certification and training.

---
PA

The fact that there are a number of defined Agile "processes" (Scrum, Kanban, XP, etc.) pretty much shows that there's no "one way to do it" by those who think Agile is worthwhile.

The manifesto sets out a few simple statements that the people who created it believe lead to a better software engineering experience and believe will lead to better crafted, more maintainable software products that actually deliver value to their customer in a timely manner. It's hardly a cult.

Really? XP was the original Cult, and the one that laid out the path of taking things on faith -- though it is mostly dead, it's roots live on in the marketing efforts of Scrum.

Scrum is the most cult like with the rituals, the scrum masters, the certifications, the do it by the book mentality and all the rest.

Kanban is the least populated of the agile space, however, David Anderson has made some pretty outlandish statements, and, Kanban was taken from manufacturing and has little to do with software. Whether it's a cult or not is too early to say for Kanban, but certainly the other two have proven their cult nature.

Merely saying that something is not a cult doesn't make it not a cult.

I know that Agilists give the same credence to mere statements and opinions as they do facts, but that is a dysfunction that we don't all share.

I'm sure Scientologists believe that it isn't a cult. Most everyone else thinks that it is.

PA

I am aware that the blog of the person you posted to, is heavily biased in favor of agile, and went on a rant against Voke's conclusions.

Voke's sampling of 200 companies is the largest independent analysis that I'm aware of, and they report that only 2% of the companies could identify identifiable success with agile.

Thanks for the reply PA.

There is nothing in the world self-evident except for taxes and death. Self-evidence is a prop when proof cannot be provided.

I agree that the Agile group should have been providing proof all along and they are equally if not more in wrong as far as providing proof. But that unprofessionalism does not excuse others from being professional. The 'he started it' argument between sibling adolescents should remain there. Each proposal must stand on its own with its own facts.

How did we get to the question of whether Agile works? (this is another chapter in Sagan's book that once the proposal's facts are under question, the proponents change the question)

And no it isn't non-sensical and Bell is not just listing fevers that occur in the population. If he called it Software Development or methodlogy fevers I would have no issue.

He called it Agile fevers without any facts to back up that the fevers afflict Agile projects more than other projects. The title of the article implies the distribution he feels they occur in the population.

I agree that the fevers absolutely exist and that they absolutely exist in the Agile community. No question.

What is unproven if whether they exist to a greater extent in Agile than in other methodologies. That is the current question that bears supporting evidence.

So far I hear crickets.

Agree 100% Curt. I also classify myself of a hybrid-methodologist. I take part of all different methodologies that support each specific project and team I currently have.

I like to describe my Agile style as just finding better ways...

This is just the kind of Fragilista response that Alex anticipated when he wrote the article.

Your arguments, such as they are, are illuminating and illustrative.

You don't care about eradicating dysfunction in the agile community, you only care if it's greater somewhere else, so you can conveniently sweep it under the rug.

Hey forget about the crime rate in Boston, it's worse in Zimbabwe! Whew, I'm relieved.

On a side note I was discussing with a friend how people who are engaged in the practice of monetizing agile should state so in their responses on the internet, just like stock advisers who talk about stocks disclose whether they have an open position on it.

Regards,
PA

It is unfortunate that we are back to the name-calling stage. Really? After all we have discussed on the importance of professional discourse without demonizing the opposition. (Your words, not mine)

I care about eradicating dysfunctions in all communities. I have already stated that I acknowledge these Methodology fevers do exist in the Agile communities as well. I'm not sure how you interpret this as sweeping it under the rug? I'm not stating that they exist more elsewhere and I'm not trying to sweep them under the rug.

We in the Agile community will continue to work to address these and eradicate them. My comments were intended to ensure you don't have a false sense of security and sweep them under the Waterfall rug. They exist in that methodology too, I suggest you work on eradicating them on your projects as well.

Not sure what to make about your comment on monetizing Agile. This comment can be made on any speciality by consultants. I monetize Data Architecture as well. You and Alex monetize traditional project management approaches.

So in the absence of addtional facts or context, I'll continue to take this excellent advice from Alex and address methodology fevers on all my projects, I encourage you to do the same.

Thanks for this article. It's a breath of fresh air.
I have both a Masters degree and a PhD - both studying aspects of agile. I personally think it is a good set of tools and ideas for developing software, however I have never been an evangelist (perhaps because I have too much objective science in my blood). I have nothing against agile practices, tools, techniques, practices, principles, theories, or philosophies. I too find some of the pro-brigade to be rather tiresome. I also felt this during the UML fever phase. Keep up the good work on providing some balance!

First....and foremost, I would like to publically apologize to Terry for not being nice. He is right on the money in his criticism of me. Those who know me will attest to this being uncharacteristic of me, and I was at the end of a not so good day when I replied. But, this is no excuse. As far as debating this issue, I would prefer discussing it on the phone or privately via email if you are interested.

I was just going to let this die, but a few anonymous emails came my way suggesting that I owed Terry a response. So, here is what I hope to be a very structured, logical, and nice response to what I interpret his question/comment/point to be.

1) First, I do not know the PostAgilist, but he has already provided this forum with all of the information necessary to address the issues and concerns as I have understood them to be. He understood the deeper message of “Death by Agile Fever” far beyond the tongue in cheek Fevers which are only indicative of a larger problem.

2) I am not anti-Agile, I am anti-waste.

3) I have seen all of the documented Fevers with my own eyes and had to trim out about five others for the sake of article brevity.

4) The article is not anti-Agile as stated in the fourth paragraph:

Before any Fragilistas who might be reading this article get their knickers in a twist due to a belief that it is an attack on Agile and must be avenged with flaming commentary against its author, fear not.

5) I do not blame Agile itself for any of the madness that is the subject of the article as stated in the first sentence of “The Andidote”:

Agile doesn’t cause the Fevers previously described, people do. Whether these people are well intended, have studied at the finest schools, or have high IQs, they are typically ignorant of Agile in many dimensions.

6) I would like to think that we are all on the same page at this point and that the issue to debate is whether it is legitimate to replace the word “Agile” with Waterfall, for example, in “Death by Agile Fever”. Yes, one could do a global substitution in the article and some of the fevers might ring true, but that is not really the point of all this.

7) As the PostAgilist clearly understands, there is much more to this than word substitution in an article and claiming that Agile is just like the crazes or fads that might have preceded it. Agile has become religious, cultish, and has spread throughout software organizations without precedent as I have observed in my 30+ year career. Since Terry used Waterfall as a comparison, let me do the same in real life situations where I have replaced Agile with Waterfall:

- How many times have any of you ever seen or heard of senior management’s performance criteria being measured by the number of their programs that adopt Waterfall (without considering that the benefits of Agile are variable and based on specific context)?

- Let’s say we have a multi-million dollar program that has been under development for over 5 years. This program is fixed price, the requirements are static, the customer is on the other side of the world, and is not integrated with the development team. Let’s say that this program is working on its last software delivery. How many times have you heard of a program like this adopting Waterfall on the last build to “get done faster”?

- How many times have you heard of people making concocted claims that they have saved programs tens of millions of dollars by adopting Waterfall in favor of methodologies or tools that were previously being used, and people actually believing it?

- How many times have you been working a fixed price software development proposal with static requirements, an inflexible schedule, the cost of software is deemed to be too high, and a non-technical person proposes that usage of Waterfall could help reduce the cost (without understanding anything about the development context)?

I could go on and on with Agile misconceptions, ignorance, and hype examples that are the fault of people, their self-interests, their ignorance, and not Agile itself. Anyway, if you believe that the examples above have never been heard of in a Waterfall context because only an Agile context is worthy of them, well, I guess we are done.

So, for many of the reasons already stated by the PostAgilist, and based on the few examples of Agile brainwashing I have provided above, I do not believe that it is legitimate to replace Agile with any other word in my article. But as I already suggested, this position is hardly worth debate or expenditure of energy, and is just a distraction from the real problem.

From my knothole, the Agile movement has brainwashed people like nothing else like we have seen in the history of the software industry. That brainwashing does not confine itself to the technical ranks like UML Fever largely does, but transcends into the management ranks with negative consequences as I have never seen before. The result of this brainwashing causes bad decisions, wastes money, and makes my job much more difficult…all the while that the people fueling the brainwashing fail to acknowledge it or take responsibility for it (Yes, I gave that small child scissors, but I did not tell him to run with them). So, there you have it…I hope I answered the comment/question/issue.

Your explanation is clear and helpful. May be, I should start examining if any of our colleagues has caught such a fever.

You referred to PA’s postings but on 13/09/2012 06:11, PA said that you are “rejecting Agile, in many ways, because there is NO PROOF that agile is effective” and works.

If it is not effective and not working at all, we should abandon it altogether instead of just eradicating the fever. Which path should I suppose to take?

One key fever that I think Alex has only just started to characterize in this last comment is what I'd call "System Blindness" fever. The afflicted believe that "Agile" is just another "method" to "deploy" (in the "pants down, bend over, here comes the new method" kind of way ;), without much consideration at all, if any, of the systemic context in which it might be applied. They care far more about the "label" being applied to Projects (which is fairly easy) rather than perfecting an Agile mindset and practicing continuous improvement throughout the organization and across divisional boundaries (which is fiendishly hard).

We need a system for transforming mediocre teams into outstanding teams. Agile concepts and habits show great promise in helping organizations to achieve this nearly magical feat - provided people consider the objective of the exercise to be creating more delight for customers, with more joy in work for people. We need to figure out how to inspire more people to develop a more balanced and nuanced understanding of agility, its values, principles and habits - rather than just ending up with lots of religious fervor and zeal.

Leung,


The answer to your question is in "The Antidote" section:

The objective of defining one’s software development process should not be to make it Agile-based, but defining it to be the most efficient and productive it can be given a program’s unique context. The Frog and the Octopus Model [3] proposes nine universal concepts that are relevant in some form to all software development efforts (The Frog) and a set of program unique variability factors (The Octopus) influencing those universal concepts. The Frog and the Octopus can be used to help guide the definition of a program’s software development process without the clutter of whichever silver bullet might currently be in fashion.

"Death by Agile Fever" is not about Agile being good or bad, valuable or not valuable. It is all about Agile based development processes being misapplied by the ignorant, the misinformed, and those with self interests. So, you need to do an inventory of your specific context and decide what tactics are the best for developing software to meet your customer's needs, your customer's schedule, your budget, etc, etc.

Hi Alex, thanks for the response. I only checked back today into this discussion after a while away.

I do agree that these situations/fevers exist perhaps more in the Agile arena than others. (Although all methodologies can suffer to them to a degree)

In many ways I think we all manage projects in the same manner, but are disagreeing on language and semantics.

Terry

I think you just demonstrated Simon Said perfectly