TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Agile's early evangelists wouldn't mind watching Agile die

616 pointsby prepperpottsabout 5 years ago

85 comments

plughsabout 5 years ago
My first team, some 14 years ago, had the best experience with Agile. What made this work<p>&gt; During planning, everyone understood that the estimates were _estimates_. Tasks could take longer or shorter, that was OK and even expected. The goal was to improve velocity, not to accomplish every story, every time.<p>&gt; Engineers felt comfortable taking stories in areas they knew nothing about. A stated goal was that every engineer understand every piece of the project.<p>&gt; Standups were team only. Engineers felt comfortable saying they got stuck or needed help or got lost in a rathole and didn&#x27;t make the progress they hoped.<p>&gt; Demos were important so that the team and the clients could understand what had changed and what new features were available.<p>But it quickly got lost - managers and PMs and Directors got involved and Jira boards became published for all to see. (In the beginning it was whiteboards and sticky note). Standups and Demos were all about self-promotion. No one ever wanted to take on a task they weren&#x27;t 100% certain they could do. If a task was 3 point it needed to be done in 3 days or there would be questions.<p>When I left the last company it had gotten outrageous. Every task should be completed in three days and in production in five days - and if not that team was Doing Something Wrong.<p>( We were told that this was standard FAANG practice, I have no idea how true that was )<p>What happened was what you&#x27;d expect - shortcuts, tech debt, unit tests cynically design to pass and meet code coverage expectations instead of actually usefully testing.<p>The reality is that any process is going to fail one senior leadership decides it&#x27;s a way to evaluate engineers, not create a good product.
评论 #22971882 未加载
评论 #22971821 未加载
评论 #22974320 未加载
评论 #22974289 未加载
评论 #22972024 未加载
评论 #22974474 未加载
评论 #22976660 未加载
评论 #22973080 未加载
评论 #22972747 未加载
评论 #22974653 未加载
评论 #22974785 未加载
评论 #22976882 未加载
评论 #22974018 未加载
评论 #22977089 未加载
评论 #22973278 未加载
评论 #22977304 未加载
评论 #22974789 未加载
评论 #22975915 未加载
评论 #22974429 未加载
评论 #22971828 未加载
评论 #22985090 未加载
0x262dabout 5 years ago
Agile&#x2F;software engineering practices broadly suffer a lot from the fact that not only is selling software a business model, selling Agile is also a business model, and when you buy Agile consulting it&#x27;s impossible to disentangle whether it&#x27;s being sold to you because it works or because it makes that person money. They don&#x27;t even know because it&#x27;s their job and they convinced themselves of its usefulness.<p>It can still be evaluated on the merits but IMO this greatly pollutes the speed at which software devs as a broadly conceived community can come to consensus understanding of this.<p>Also I think the comparison to lean manufacturing has always been very shallow. I get the metaphor, I just don&#x27;t think that human resources in engineering can be optimized like manufacturing processes. This quote is the best part of the article:<p>&gt; &quot;You’d never hear anyone say, &#x27;We help mechanical engineers be agile. That would be silly. And I mean that in the worst possible sense of the word&quot;.<p>As for the rest of it, I&#x27;m not dying to hear what the person who invented Agile thinks we should do next lol.
评论 #22971574 未加载
评论 #22970619 未加载
评论 #22970330 未加载
评论 #22970544 未加载
评论 #22971686 未加载
评论 #22971259 未加载
评论 #22970866 未加载
评论 #22973815 未加载
评论 #22970302 未加载
james-mcelwainabout 5 years ago
I don&#x27;t know if something is wrong with me, but I simply cannot &quot;decompose&quot; a single feature into 500 different sub-task Jiras.<p>I understand why it would be super cool if software worked like that, but I have to iterate on features holistically and sometimes speculatively.<p>At least for me, software development has never been able to be so cleanly broken down into &quot;first, implement the button&quot; type tasks.<p>But maybe I&#x27;m just dumb.
评论 #22970655 未加载
评论 #22970451 未加载
评论 #22971662 未加载
评论 #22972549 未加载
评论 #22970803 未加载
评论 #22971127 未加载
评论 #22971146 未加载
评论 #22971142 未加载
评论 #22970498 未加载
评论 #22970415 未加载
评论 #22970683 未加载
评论 #22971216 未加载
评论 #22970507 未加载
评论 #22971976 未加载
评论 #22971908 未加载
评论 #22972893 未加载
评论 #22971006 未加载
评论 #22974425 未加载
评论 #22973321 未加载
评论 #22970559 未加载
评论 #22971113 未加载
Someone1234about 5 years ago
Agile is the worst form of Software Development Methodoloy, except for all the others.<p>Which is to say: To all the people jeering for Agile&#x27;s demise, please provide a superior alternative. I came from a world dominated by Waterfall, and I never want to go back to that. A lot of companies get Agile wrong, and it isn&#x27;t a panacea, but it is much closer to what developers are naturally inclined towards (e.g. rapid incremental improvements) than Waterfall ever was.<p>So I challenge anyone who wants to replace Agile, please lean into how developers work rather than trying to mold their work onto your rigid front loaded methodology. Trying to bring in ideas of other industries, like engineering or architecture, that build physical goods and only have one shot at is a folly.
评论 #22972289 未加载
评论 #22970603 未加载
评论 #22970611 未加载
评论 #22971190 未加载
评论 #22973021 未加载
评论 #22976006 未加载
评论 #22970812 未加载
评论 #22971229 未加载
评论 #22972543 未加载
评论 #22973598 未加载
评论 #22970706 未加载
评论 #22970940 未加载
评论 #22973673 未加载
tremoloquiabout 5 years ago
The terms &quot;agile&quot; and &quot;Agile&quot; have two separate meanings.<p>Big &#x27;A&#x27; - &quot;Agile&quot; is waterfall re-branded - an excuse for corporate empire building and business as usual. It involves lot&#x27;s of meetings and not trusting those who build software to do the right thing.<p>Small &#x27;a&#x27; - &quot;agile&quot; is the implementation of the manifesto, which basically comes down to smart people figuring out how to work together towards a goal, often by taking small steps.<p>Until the terminology is sorted out the discussion can&#x27;t help but be confused.
评论 #22970610 未加载
评论 #22970566 未加载
评论 #22970760 未加载
评论 #22970466 未加载
luniasabout 5 years ago
IMO Agile has become regulatory capture. It&#x27;s a means by which non-engineers can extract value from a booming market which doesn&#x27;t directly benefit from their skills.<p>That being said. I think there is a lot of wisdom in the original agile manifesto. The core principles are solid, but the methodology has clearly been co-opted by consultants and supported by management looking to increase the headcount under themselves.<p>I&#x27;ve often struggled to understand why my team is made up of only 20% engineers with the other 80% pretending to create value by holding meetings to tell engineers what to build next when I feel like it&#x27;s your clients that should be doing that.<p>Ultimately it&#x27;s engineering that becomes the constrained resource which leads to technical debt in favor of pushing out product&#x27;s features.<p>I would venture a guess that most engineers have used (critically) more software in their lives than any non-technical person driving the development of the product. Why then are engineers not the most consulted people on the efficacy and value of new features? I think there is a big myth out there that engineers are incapable of directly handling client feedback.
评论 #22982699 未加载
mywittynameabout 5 years ago
I&#x27;ve been through three full-fledged &quot;agile transformations&quot; in my career. I hate agile so much. I hate it because it encourages dogmatic behavior instead of rational behavior.<p>I think it has survived so long because it provides its practitioners with two ace excuses: when things fail, they say, &quot;what we did wasn&#x27;t really agile,&quot; and when somebody proposes and improvement they don&#x27;t like, it&#x27;s always, &quot;that sounds like waterfall.&quot; A team can&#x27;t recover from their agile transformation until both of those phrases disappear from the team lexicon.<p>&gt; “Way too much of Agile has been not about technology, but about people and about managing things and about getting stuff done — not necessarily getting the right stuff done, but getting stuff done — rather than what engineering is about,”<p>I&#x27;m not sure agile was ever designed to fix this. The whole process is really handy-wavy about the actual engineering: &quot;we&#x27;ll just make this a spike.&quot; The process also demonizes documentation and design work, which is antithetical to most engineering.<p>That being said, I love iterative development, and I like the concept of stories as descriptions of functionality. But I do think agile, as taught by most consultants, is really designed for consultants who build CRUD web-based applications. Sprints are timeboxed so consultants can charge by the spring, and stories lack implementation details because companies hiring consultants don&#x27;t want or need to know how their sausage is made. The further away your product is from that area, the less effective agile is; some companies really need to design their sausage well, up front, because they are going to be eating it for years.<p>A process that&#x27;s built up over years has become the way it is typically for good reason. And in my experience, teams forced to transition to agile will end up shoehorning their existing process into &quot;agile&quot; and calling it done, so most executives don&#x27;t really see how their &quot;transformation&quot; failed.
评论 #22992392 未加载
baneabout 5 years ago
It&#x27;s gotten to the point where it seems that the projects that do more of the Agile &quot;stuff&quot; move much slower because they&#x27;re spending so much of the working time doing that stuff than slinging code. Where I&#x27;m at, we&#x27;re running something around 4 dozen projects at once and the teams that really focus on doing all the ceremonies are simply left in the dust by the teams that don&#x27;t.<p>Here&#x27;s another anecdote, a couple big teams that went the full SAFe route spent 3 years building something that ended up being thrown away and replaced in 4 weeks by 3 staff working part-time who were just given some instructions, some deadlines, and told to go code.<p>It&#x27;s become repeatable, big teams toiling for years getting outpaced by smaller, more &quot;agile&quot;, teams.<p>The balance has gotten out of whack again and it&#x27;s really time to start over again. When I go back and read the original manifesto, what I see today in modern practice looks nothing like what was intended. The point in too many shops has been to accomplish the Agile things, not to deliver.
评论 #22977098 未加载
RcouF1uZ4gsCabout 5 years ago
&gt; At the time, software development was suffering from a mistaken belief: that building things fast and building things well were fundamentally opposed.<p>I thing “good, fast, cheap; pick 2” is still generally valid.
评论 #22971161 未加载
评论 #22971126 未加载
评论 #22970785 未加载
vpeters25about 5 years ago
Every time I see an &quot;agile sucks&quot; post, I take the time to read it and every time (so far) I have found they blame the process for some key part of agile they missed. Quote from the article:<p>“Way too much of Agile has been not about technology, but about people and about managing things and about getting stuff done — not necessarily getting the right stuff done.”<p>This is the whole point of agile: progress on iterations, inspect and adapt at the end of each iteration.<p>Your team might build the &quot;wrong stuff&quot; for an iteration, realize it (inspect), then make a course correction (adapt). If you end up delivering the &quot;wrong stuff&quot; is because you didn&#x27;t follow this very core principle of agile.<p>I find it hard to believe these so called &quot;Agile Early Evangelists&quot; can make such a statement. Their background in lean development should have made the familiar with empirical process controls, from where lean and agile come from.<p>My guess the author quoted them selectively to fit the &quot;agile sucks&quot; narrative of the article.<p>Edit: expanded last 2 paragraphs.
评论 #22970700 未加载
评论 #22973770 未加载
评论 #22976721 未加载
评论 #22970567 未加载
legitsterabout 5 years ago
There are many fingers to point here, but I point mine at bigger consulting firms like Accenture. I have first hand experience with their miserable &quot;agile&quot; process at Microsoft.<p>If you want a stapler, you have to schedule a meeting 3 weeks in advance with 2 team leads and 3 contractors who decide what sprint &quot;getting a stapler&quot; will be assigned to. Then &quot;getting a stapler&quot; gets pushed back two sprints because some priority came along (a vice president somewhere just learned that his binder was on hold for four months). An Accenture contractor from India informs you that a hole punch is on the way. Apparently you checked the wrong box on the stapler requisition form, so you have to start the process over again.<p>Then the Accenture quarterly deck shows they met 97% of all stapler demands for the quarter! Isn&#x27;t life so much more easier and more measurable now? But they will need to hire more contractors to keep up with capacity.
评论 #22972146 未加载
vbtempabout 5 years ago
For the past several years now, each time I join a new team or project, my first objective is to detach and destroy the &quot;Agile&quot; process workflow.<p>When you finally liberate a team from Agile, it&#x27;s just breathtaking how much you can focus on delivering working software that gets deployed with quick iterations that&#x27;s closely aligned with the business and customer&#x27;s needs. When free from the tyranny of Agile, teams can be effectively self-organize, remove micro-managers, and quickly adapt to changing needs and requirements. My experience is that staff are usually much happier, more productive, and less stressed once agile is gone.<p>I mean contemporary Agile as pushed by corporations and those awful &quot;coaches&quot; (who never seem to be actual developers) --- if you were you design a system whose end goal was making great developers unhappy, unproductive and locked into a dysfunctional system, Agile would be it.
评论 #22970180 未加载
评论 #22977080 未加载
raiyuabout 5 years ago
Can we just replace agile, and every modern management practice with some mix of common sense, talking to the customer, talking to the team, and being able to voice concerns to upper management where people actually listen instead of running their own agenda? =]
评论 #22969972 未加载
评论 #22970019 未加载
评论 #22970085 未加载
评论 #22970423 未加载
评论 #22970561 未加载
评论 #22970303 未加载
评论 #22970324 未加载
评论 #22970055 未加载
评论 #22970667 未加载
评论 #22970853 未加载
adwnabout 5 years ago
&gt; <i>Women dominated computing professions from their inception in the 1940s all the way through the 1960s. As the scope of computers’ usefulness became clearer, however, those jobs started going to men.</i><p>Bullshit. Back then, the term &quot;programmer&quot; didn&#x27;t mean what it means today. A &quot;programmer&quot; was someone who entered a program into a computer, or even earlier, plugged in wires according to a schematic. The actual development of the program was almost exclusively done by men, even more so than today (this is not supposed to be a value judgement, nor do I want to justify the situation back then or today).<p>It&#x27;s really disappointing to see this tired myth of a supposed &quot;Golden Age of women in computer science&quot; repeated in the article.
评论 #22973717 未加载
at_a_removeabout 5 years ago
I was exposed when it was still eXtreme Programming (back when &quot;extreme&quot; was the most nineties adjective you could have) and didn&#x27;t much care for it then.<p>I keep hearing it touted as this kind of universal cure-all but I just do not think it is applicable everywhere. On a personal level, the more any given thing is touted to me as the solution for any and all problems, the more suspicious I grow. The more fervor, the more I draw away. The more cultish it seems (convert, adhere, rituals, special names, and then finally the narcissism of small differences in practice as exemplified by that old Emo Williams routine), the less I want to participate.<p>The more I have seen it shoehorned into places where it seemed ill-suited, the more this was revealed to me, yet the evangelism continues. Nothing like watching the tape backup guy, suffering from a flare of gout, hobble over to a room to perform his &quot;stand-up&quot; routine to hammer that home. Of course, we &quot;probably weren&#x27;t doing it right,&quot; but the more I have seen the slavish adherence to the strictures of Agile the more it seemed like this was an exercise in polishing a boss&#x27; resume so that he could say he had done The Things when he went elsewhere.<p>Reader, he did.<p>Perhaps the Agile that can be critiqued is not the eternal Agile.
评论 #22972212 未加载
dragonwriterabout 5 years ago
The weird thing is that all the criticisms of misplaced focus in agile made in the article were criticisms commonly made <i>by</i> early Agile evangelists (including Poppendieck!) of pre-Agile approaches, and the things pointed to that are needed which make Agile irrelevant are the things that Agile methods were specifically advocated as superior for by those evangelists.<p>The real problem doesn&#x27;t seem to be that Agile is irrelevant, it&#x27;s that Agile (and this is actually fairly quickly explicit in the Manifesto and accompanying Principles) cannot be limited to a siloed tech organization but must fully incorporate the business sponsor of the project, whether that&#x27;s the firms business management for a software&#x2F;services firm, whether it&#x27;s an external customer for a firm doing bespoke software development, or whether it&#x27;s an internal customer for an IT organization doing enterprise-internal development. And despite that being clear in the Manifesto and Principles, it very often is the case that “Agile” is seen as something the tech team is (or worse, <i>does</i>) that stops at their boundaries, and that fundamentally doesn&#x27;t work.<p>There are other equally serious problems, like ritualistic adoption of fixed processes being described as “Agile”, which it clearly is exactly what Agile is most opposed to; the use of the phrase “Agile (or Scrum) ceremonies” is a clear sign that an organization has been infected with this toxic mindset.
_jalabout 5 years ago
Having been through &quot;agile conversion therapy&quot; at two different companies and then having seen what the companies actually did several months later, I think most of the value of &#x27;agile&#x27; has been realized. To get much more out of it, you need some combination of a more capable culture and better disciplined people.<p>And I&#x27;m not talking about just engineering - I&#x27;m mainly talking about customers and management.<p>Most of the gains come from tighter feedback loops between engineers and other stakeholders than tend to happen without an intentional effort. This works because it is simple - tell teams to talk amongst themselves every day, get feedback from customers more incrementally than you otherwise would, etc.<p>Most of the rest of the promised gains need better disciplined customers and managers - ill-defined specs, dropping projects to pick something else up, only to switch back a couple months later, managers who don&#x27;t pay enough attention to what&#x27;s being done until it is too late, etc.<p>Perhaps the problem is that it is a management fad aimed at engineers. It needs a v2 aimed at nontechnical people, &quot;become competent at asking for what you need&quot;.
评论 #22973475 未加载
flohofwoeabout 5 years ago
Call me jaded, but that&#x27;s probably because the cow stopped giving milk, and they need to come up with a new cult&#x2F;religion&#x2F;ideology to continue selling books and conference tickets (and consulting gigs of course).
评论 #22970778 未加载
评论 #22970320 未加载
Yhippaabout 5 years ago
Nearly everywhere I&#x27;ve worked the quarterly earnings report is the problem. Due dates mean that time has to be managed so any attempt to use a methodology seems to regress to some not fun way of developing software.<p>In places I&#x27;ve worked without the pressure of the quarterly earnings report I&#x27;ve seen that developers tended to use whatever methodology worked best. It could be Agile, pair programming, or just walking over to each other&#x27;s desks and talking on IM.
commandlinefanabout 5 years ago
I don&#x27;t want it to die, I want it to become what it was originally supposed to be.
评论 #22970843 未加载
评论 #22970370 未加载
genezetaabout 5 years ago
Speed. Sometimes I wonder if people mean speed as in velocity, or if they mean the drug, seeing as they seem to obsess so much about it.<p>&gt; Engineering is about seeing a problem, understanding it and using the technology you’re good at to solve it as quickly as possible.<p>Everywhere you turn to there is someone thinking that speed is the most important thing ever. Everything must be done <i>quickly</i>. The best way to do something is <i>quickly</i>. If something takes longer, it&#x27;s clearly worse.<p>Even in an article about &quot;the problems with Agile&quot;, people still fall into the trap of speed.<p>Engineering is clearly <i>not</i> about solving problems &quot;as quickly as possible&quot;. It&#x27;s about solving them in the <i>best</i> way possible, where you first define &quot;good&quot; as a balance between all the advantages and all the disadvantages of each possible solution. <i>Speed</i> is but one of the factors you evaluate.
评论 #22976497 未加载
alangibsonabout 5 years ago
I&#x27;m not here to defend Agile, but one of the things people get perpetually wrong is thinking about Agile as methods and tools. Agile is actually a set of values and principles. You were supposed to fill in your own methods and tools to support Agile values and principles, but everyone basically just said &#x27;too hard; do Scrum&#x27;
评论 #22972815 未加载
评论 #22974157 未加载
xnxabout 5 years ago
One of the all-time most cursed diagrams I&#x27;ve seen: <a href="https:&#x2F;&#x2F;www.scaledagileframework.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.scaledagileframework.com&#x2F;</a> . I could not believe it was not parody.
评论 #22970893 未加载
评论 #22974141 未加载
评论 #22970899 未加载
评论 #22973318 未加载
_bxg1about 5 years ago
Agile is one of those words that&#x27;s come to mean nothing at all. It can be anything from &quot;iterate fast and do weekly stand-ups&quot; to &quot;spend all of your time allocating point values to planned sprint tasks so that management feels like they&#x27;re in total control&quot;. It&#x27;s been co-opted by the very bureaucracy it was designed to reign in.
empath75about 5 years ago
I was saying in a team meeting the other day that using jira stories as any kind of performance metric is wrong-headed and counterproductive and metrics should focus on the business value provided by the team and people looked at me like I grew a second head.
评论 #22970132 未加载
评论 #22970066 未加载
illuminatedabout 5 years ago
The main issue my teams have been facing with Agile methodologies is the number of the workflow interruptions built in. The only team members happy with those interruptions were Scrum Masters and PO&#x27;s but that&#x27;s because those were parts of their flows, so they had to have them in order to have their job done.<p>The only long term solution for us was to have days dedicated to Agile ceremonies and days where no team member can be interrupted based on an agile ceremony need (there were few exceptions, like critical bug discovery, etc.). In weekly sprints, we had at least 3.5 days of uninterrupted time and in two-week sprints we had 7-7.5.<p>That changed a lot the pace of those sprints as people had a lot of time to do their job and everyone was happy.
评论 #22970666 未加载
评论 #22970604 未加载
cesiumabout 5 years ago
When something develops to an extreme, it becomes its opposite. This is true for &quot;agile&quot;. Agile has become synonymous with &quot;Jira&quot;, which is a direct contradiction of its original premise. The original &quot;Agile Manifesto&quot; states: Individuals and interactions over processes and tools, Working software over comprehensive documentation, Customer collaboration over contract negotiation, Responding to change over following a plan. Ask somebody these days what agile means and they&#x27;ll tell you &quot;Jira&quot; or some other tool combined with heavy process. Another example of people confusing the means for the end itself.
评论 #22973259 未加载
davnicwilabout 5 years ago
The process is invented to help to do your job.<p>Problems begin when that gets forgotten, and following the process <i>becomes</i> your job -- not following the process is doing your job wrong.<p>Always remember that your job is to build stuff, any way that works.
评论 #22971990 未加载
peterwwillisabout 5 years ago
The core of Agile, Scrum, DevOps, XP, TDD, etc, is just to get a developer to truly understand what it is they&#x27;re <i>doing</i> and make a meaningful connection to the real results of it. But before they can ever begin to do that (which is a huge challenge in and of itself), the organization needs to <i>get the hell out of their way</i>, but at the same time, <i>invest tons of money and effort to help them do what they should be doing</i>.<p>If you just &quot;step back&quot; from trying to control the average bunch of engineers, they will miserably fail, because they have never gone to a school that taught them how to rapidly drive business value through software. If you hire the right people, they already learned all of that, and so they&#x27;ll be successful.<p>That&#x27;s how the Netflixes of the world survive. Their organization and process would seem <i>totally insane</i> to any &quot;average&quot; tech engineering organization. But it works for them because they learned how to work together the correct way, the business got the hell out of the way, and it also invested a ton in helping them do what they need to do.<p>Agile does not give you any of that. Agile is just a promise of what <i>should</i> be happening, but it doesn&#x27;t tell you how to get there at all. It&#x27;s a pipe dream.<p>Agile is not the problem, though. It&#x27;s businesses thinking the engineers are simple tools to be used to produce widgets, and if they just shove enough extra business roles and processes around the engineers, that will end up creating value. But the solution is actually to <i>remove</i> all that extra crap, and get the engineers to do everything in a way that the PMs and DMs and QEs etc are no longer necessary.
评论 #22972621 未加载
pragmaticabout 5 years ago
Speaking of Lean, I remember my mother telling me about their hospital adopting lean and that meant they could keep only so many supplies on hand in the lab. I remember thinking at the time &quot;wow, isn&#x27;t it a bad idea to try to limit the amount of supplies on hand at a hospital&quot;<p>I guess we&#x27;ll have to see if we ever have a major health care crises. Wait...<p>And we wonder why the hospitals didn&#x27;t have enough PPE on hand. Did the MBA virus spread to hospitals and did they misapply methodologies like Lean where it had no place, except as short sighted cost cutting?<p>One of the major problems in industry I&#x27;ve seen again and again is management, etc pretending they know more about the job and it&#x27;s requirements than the people doing it.<p>As Lean originated in manufacturing with the WORKERS making the improvements and the WORKERS being empowered (anyone could stop the line if there was a problem) it&#x27;s bit ironic that a corrupted version was used later to beat workers into submission and achieve petty cost cutting versus process improvement as in the health care anecdote above.
aeturnumabout 5 years ago
If you should meet Agile along the road, destroy it!
wodenokotoabout 5 years ago
What is the alternative to agile?<p>Calling it a to-do list instead of user stories? Not aligning teams on where we are in the to-do list regularly and removing and adding things?<p>Agile proponents often say &quot;waterfall&quot;, but I have no idea how you would go about working waterfall day to day.<p>Would you just make all your user stories&#x2F;to-do&#x27;s on day one and never update them and now you have waterfall?<p>Having entered the industry long after agile became the new normal, it seems like the whole discussion is either &quot;stand-up sucks&quot; or people discussing variations of what is essentially &quot;Regularly report where you are on the to-do list, often update the teams to-do list&quot;, but never &quot;Why not this different paradigm instead&quot;.<p>Maybe I am lacking imagination, but I am not really sure what a serious alternative is here.
temacabout 5 years ago
&gt; At the time, software development was suffering from a mistaken belief: that building things fast and building things well were fundamentally opposed. Mary knew this to be untrue from her work in product development and manufacturing, where the ‘lean’ practices that sprung up in Japanese car factories and subsequently spread across the globe often ruled the day.<p>If that&#x27;s part of the original reasoning, then it was very random. R&amp;D has not much to do with manufacturing. So I would greatly prefer an actual example from &quot;product development&quot; rather than car factories...<p>I mean what build our software is e.g. gcc. I don&#x27;t know if gcc is feeling lean, and actually I don&#x27;t care much :)
arminiusreturnsabout 5 years ago
I recently listened to a lecture about the future of programming, and there was a sentence that really stuck out to me. The speaker said something along the lines of &quot;If you go to an agile conference, there aren&#x27;t that many devs there and it&#x27;s mostly management.*<p>As an ops type usually only on the peripheral of sprints etc I feel I don&#x27;t fully grasp the nuances of the different methods enough to really comment, but that just really got my attention for some reason.<p>Found the video, linking at the timestamp that includes the manifesto for agile (which I think has some obvious flaws)<p><a href="https:&#x2F;&#x2F;youtu.be&#x2F;ecIWPzGEbFc?t=3678" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;ecIWPzGEbFc?t=3678</a>
brainlessabout 5 years ago
So much frustration over something that in theory is so good for building stuff.<p>If you go over to <a href="http:&#x2F;&#x2F;agilemanifesto.org&#x2F;" rel="nofollow">http:&#x2F;&#x2F;agilemanifesto.org&#x2F;</a> and read these simple 4 sentences, it makes sense:<p><pre><code> * Individuals and interactions over processes and tools * Working software over comprehensive documentation * Customer collaboration over contract negotiation * Responding to change over following a plan </code></pre> What goes horribly wrong is managers, and their incapability. They do not work with teams, instead they use agile as a means to simply get updates at the end of week and set orders down the ranks every Monday morning.
评论 #22975804 未加载
winridabout 5 years ago
When I was a manager I&#x27;d get the team together every quarter and we&#x27;d agree on &quot;this is what we&#x27;re going to do&quot; - and then we&#x27;d do it, usually. At the end of the quarter we&#x27;d have to explain why certain pressures pushed things out, but usually the business cared more that we kept churn down.<p>For over a year I pushed back on story pointing and things that felt micro manage-y to me.<p>It was very hard on me, but it was worth seeing the productivity.<p>Eventually I was forced to give in to Agile with the big A and left soon after.<p>I can&#x27;t get another management job because everyone wants to do Agile in a way I hate to manage - so now I&#x27;m a developer again... :)
ghevshooabout 5 years ago
In my experience the main reason that agile mostly sux in large enterprises is because it is agile in name only. Not a single one of the people who are responsible from managing the process have even heard of the “agile manifesto” let alone read it, nor have any of them ever worked as a software engineer. So in most cases it could easily be better than what it is.<p>But I object to the authors complaint “and it’s pushing women engineers into non-technical roles.” and elevating the primacy of engineering above all else.<p>People come in all levels of competencies and have all manner of skills. I don’t care if you are a man or a woman. I care about how good you are at doing the tasks that need to be done to get the job done. And in large companies there are big jobs that require a lot of people and a lot of coordination.<p>I work with a woman who is not a very good software engineer, but she is good at administering and coordinating the work among other software engineers and she quite enjoys doing those tasks.<p>While she can improve on some of her skills as a software engineer and I try to help her with those, the author’s recommendation seems to be that because she’s a woman, I should push her away from work that she likes doing, that she’s good at doing, and which is important work to do.<p>This is objectively terrible advice, but some might accuse me now of male chauvinism. If so, fine. But I will forward this article to my colleague to ask her what she thinks and I’ll let her opinion be the ruling decision.
kitotikabout 5 years ago
A lot of the bastardization and commercialization of the Agile core tenets could have been mitigated by the early pioneers broadcasting the dirty secret:<p>To deliver any real business value following these philosophies requires an extremely mature, experienced team. Not just great engineers, but also great communicators that actually care at least a little bit about the mission&#x2F;product&#x2F;goal.<p>This simply <i>does not scale</i> and is a complete antithesis of 99% of organizations culture and hierarchical structure.
KingOfCodersabout 5 years ago
I was a coder in the 80s and did XP in 1999 and became an early Scrum Master as a developer. Agile has been stolen from developers. The rise of Scrum came at the same time as the rise of the product manager, and the Scrum PO became a PM. From then on in many companies it was not about doing the right thing but getting the most features out of developers, to implement the PO vision, with little say. We became assembly line workers, put to comfort with high salaries.
brightballabout 5 years ago
Ultimately, people need agreed upon terms for communicating planning and expectations.<p>That&#x27;s it. The more people are involved, the more complicated the process gets and all of these approaches evolve out of trying to find an agreeable and effective way of doing it so that everyone doesn&#x27;t have to figure out a new solution every time.<p>As much of a buzzword hell as it is, I really believe that Scaled Agile Framework (SAFe) is the closest thing to the right balance of trade offs.
joejerryronnieabout 5 years ago
When I&#x27;m hiring scrum masters, seeing &quot;Agile Coach&quot; and &quot;Agile Transformation&quot; on a resume is actually a bit of a red flag for me. I don&#x27;t need someone who can teach my organization how to theoretically run Scrum, I need someone who can engage with our dev teams, understand the expectations they&#x27;re being held accountable for, the challenges in meeting those expectations, and tailor a methodology (probably Scrum-based) which puts the dev teams in the best position to succeed and be happy while doing it.<p>This may be a bit different depending on the make-up and maturity of the team. Sometimes, we need daily engagement with team members and &quot;structured&quot; collaboration, other times we just need to make sure everyone understands the sprint goals and then get out of their way. The challenge has always been when upper management wants to see a single dashboard which boils all teams down into a set of pre-defined metrics - and then positively reinforces teams (or their managers) for hitting metrics instead of delivering solutions. Training your VP&#x27;s can be exhausting.
viburnumabout 5 years ago
Corporate agile is a mixed bag but Extreme Programming is still good though.
评论 #22970015 未加载
larkinrichardsabout 5 years ago
I still think agile has a place as a focus for discussion when you&#x27;re taking a large organization from a waterfall model (or any model that tries to commit to delivering a set of uncertain features by a specific deadline) to a more reasonable and effective process for software development.<p>In this situation, you&#x27;d get shot down for following rigid Agile, because the process gets in the way of delivering value. What you end up doing is using the concept of &quot;agility&quot; to sell some agile-lean-hybrid-involve-the-stakeholders-think-small-and-demonstrate-the-value-of-delivering-value. It&#x27;s not about the Method, it&#x27;s about the outcome.<p>No one needs to go black-and-white rigid Agile, that&#x27;s where it went wrong. But agility is a good way to describe the concept of efficiently changing (and handling change).<p>I just feel that every article about capital &quot;a&quot; Agile immediately sinks to level of pedantry that is a complete waste of time, so... I didn&#x27;t read the OP.
joejerryronnieabout 5 years ago
The problem isn’t “Agile”. The problem is organizations which are not good at software development. Teams which fail with Agile will also fail with Waterfall or any other development methodology. Teams which generally succeed with one methodology may find a different methodology a bit more efficient, but they’re gonna be successful either way.
jquastabout 5 years ago
And just like writing software, when specifications of the agile process are not clear, almost founded on a “chose your own implementation” scheme, the result is a big-ridden over-designed and terribly unclear (and endlessly debated) process. implemented differently at each and every company. Just grand.<p>Just like the very process of software design this was meant to aid, they failed to understand the importance of clarity, brevity, and providing exacting specifications or instructions.<p>Fuck you, agile. This is not how engineering should be done in <i>any</i> profession. Thousands of hours of my career has been wasted clarifying this unclear process to swarms of people who have no single resource to learn from. And each new company brings a new “we do agile, but....” exception to adapt to, and damned be if anyone ever writes a single fucking rule of this process down, so we get to endlessly debate what is this “agile” thing at each and every opportunity.
AzzieElbababout 5 years ago
Here is the talk by Dave Thomas <a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Dave_Thomas_%28programmer%29?wprov=sfla1" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Dave_Thomas_%28programmer%29...</a><p><a href="https:&#x2F;&#x2F;youtu.be&#x2F;a-BOSpxYJ9M" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;a-BOSpxYJ9M</a>
netjiroabout 5 years ago
Agile, as commonly encountered nowadays, is mostly just cargo cult. People chasing dogma and buzzwords with zero understanding of why and when it has great application and benefits. Just like so many other trends in the past.<p>Way too easy to &quot;seem&quot; productive while actually being net negative for the project.<p>That said, I&#x27;ve had excellent agile projects that have delivered huge value to the customer at cost and time well below initial estimates.<p>These days I ask people who shout for agile to actually explain the fundamentals to me, as though I have no clue. Very rarely do I find anyone who can hit even one of the four original core ideas and values of &quot;agile&quot;:<p><a href="https:&#x2F;&#x2F;agilemanifesto.org&#x2F;" rel="nofollow">https:&#x2F;&#x2F;agilemanifesto.org&#x2F;</a><p>And when I bring up the core ideas, and why they <i>can</i> work really well, I often see that the people who shout agile actually don&#x27;t agree with the fundamentals.
5cott0about 5 years ago
Agile is what happens when stakeholders want a car but really need a wheelbarrow &amp; end up with a skateboard that only has 3 wheels. Product re-brands it as a scooter and takes a victory lap &amp; now your company culture is an extremely toxic work environment of implicit &amp; unaccountable power structures.
fourseventyabout 5 years ago
I like the development methodology laid out my David Cancel in his book Hypergrowth. I like it because he has founded several large successful companies and actually used this methodology successfully. Unlike most of these &quot;agile gurus&quot; who have never sold a product to a customer in their lives.
redismanabout 5 years ago
In my experience &quot;Agile&quot; whatever that means is still the best of all the bad options. Is there really a better methodology? I feel like it works pretty well as long as management doesn&#x27;t pick and choose the easy parts of it and then add their own layer of BS on top, which is very common.
rafaelvascoabout 5 years ago
Well, every company I worked with, &quot;Agile&quot; was just a means for management to keep feeling they&#x27;re in control, when they&#x27;re not, there&#x27;s no such thing as in control, or to promote themselves with the higher ups.<p>My best team is me alone, but other than that, the best team I ever had was one in which there was no meetings, no dailies etc. It was just me and a few others, receiving the outlines of what the final product had to do, what inputs and outputs it should process, and then filling this outline with code and hard work until it was whole. It was far from perfect of course, but we weren&#x27;t bothered by unnecessary processes. There was no agile back then yet....<p>That said, the methodologies are not the problem. The problem is peoples attitude and mentality and how they use them.
MattyRadabout 5 years ago
I think the spirit of the article is summarized by Putt&#x27;s Law <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Putt%27s_Law_and_the_Successful_Technocrat" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Putt%27s_Law_and_the_Successfu...</a>
helen___kellerabout 5 years ago
I&#x27;ve come to best appreciate agile by calling &quot;agile&quot; whatever practices work best for my team.<p>Sometimes we move things in and out of a sprint mid-sprint because new information came up and it&#x27;s high priority? It&#x27;s not ideal (and we&#x27;ll discuss it in retro) but that&#x27;s agile.<p>We&#x27;ve sat a fat 5-point ticket in our backlog for 3 sprints straight because it needs a decent sum of meeting time and dedicated attention from multiple engineers who keep facing higher priority work and not quite having time for it? That&#x27;s fine, we&#x27;re being agile.<p>Our backlog grooming meeting ended after 8 minutes because we had nothing to discuss and we&#x27;re all focused on delivering? Very agile.
评论 #22970710 未加载
评论 #22976023 未加载
d--babout 5 years ago
One thing that&#x27;s for sure is that Agile completely shifted the culture of software engineering, and generally in a good way. I am very thankful for that, and I think that this is where Agile has done its job.<p>Then, the problem with frameworks like Agile is that smart people will use them wisely, and less smart people will try to apply the concepts without really grasping the ideas and not really gain anything out of it.<p>But you can&#x27;t really prevent people from reading the book, and do stand-up scrum meetings and sprints and shit...<p>Unless someone comes up with something else (which most likely will be equally twisted) Agile is here to stay. I just wouldn&#x27;t mind watching conversations about Agile die.
lifeisstillgoodabout 5 years ago
Some thoughts I wanted to get out:<p>I think we need Outcome Driven Development- we define a metric measured in say graphite, we link a ticket to that asserting the ticket will make the metric change this way or that, and we commit code to make the metric change<p>so we start a process of defining what we want, implementing how to measure it (there could be a null metric and the ticket is to make it not null) and then measuring our success of changing the metric<p>this will have organisational impacts as well - no one should take a ticket without authority span to impact that metric<p>this will start in the direction of software as literacy or the developer anarchy
zabilabout 5 years ago
&gt; software developers need to become what they truly are: engineers.<p>I couldn&#x27;t agree more
评论 #22972228 未加载
zhenchaoliabout 5 years ago
Distracting snake oil. That’s what I think about when I hear agile.
uDontKnowMeabout 5 years ago
Tangentially related, has anyone tried to implement the ideas described in Juval Lowy&#x27;s new book, &quot;Righting Software&quot; (<a href="https:&#x2F;&#x2F;www.amazon.com&#x2F;Righting-Software-Juval-L%C3%B6wy-ebook&#x2F;dp&#x2F;B0822XJZ48" rel="nofollow">https:&#x2F;&#x2F;www.amazon.com&#x2F;Righting-Software-Juval-L%C3%B6wy-ebo...</a>)? His methodology of project management seems much more systematic than week-by-week sprint planning.
S_A_Pabout 5 years ago
I’ve worked for 2 big oil and gas firms. I don’t care much about checking agile boxes. I went from being damn near employee of the month to short list for being let go when those companies started to live and die by “agile”. I’m about 50% to blame as it’s a great demotivational tool. The other half is managers forgetting why they are running those agile projects. I left both places within 6 months of agile being gospel. Not sad at all.
projektfuabout 5 years ago
I agree with the title but find the reasoning hard to follow. It seems like the author keeps changing playing fields and tries to stitch a single narrative.
shaunxcodeabout 5 years ago
Csp should replace it. Dog food and turtles all the way down. If you can’t model your workflow in terms of processes and typed channels with buffers what business do you have implementing anything more complex (conways law etc.)<p>THAT said agile was a breath of fresh air at the time and it got us to question a lot of assumptions related to how we specify and develop software.
评论 #22973333 未加载
koonsoloabout 5 years ago
There are only 2 things your agile team needs: a retrospective and a manager that wants to improve the process.<p>Everything else can flow from there.
jackcosgroveabout 5 years ago
I never understood story points. Is it just psychological to use a different unit of time, so that managers&#x27; expectations aren&#x27;t so hard and fast?<p>Story points remind me of those weird units of work the cloud providers bill you for, or the company scrip that a coal mine would have issued 100 years ago. I call story points AgileBucks.<p>Why not just use standard units?
评论 #22975994 未加载
mindcrimeabout 5 years ago
<i>Agile&#x27;s early evangelists wouldn&#x27;t mind watching Agile die</i><p>It&#x27;s easy to say that, but the obvious follow-on question is &quot;and replace it with what, exactly?&quot; Maybe the answer is in TFA, but I didn&#x27;t read it yet. I hope they have <i>something</i> to propose, otherwise this discussion is pretty vacuous.
mjfisherabout 5 years ago
This seems like as good an excuse as any to share a link to &quot;Modern Agile&quot; again:<p><a href="http:&#x2F;&#x2F;modernagile.org&#x2F;" rel="nofollow">http:&#x2F;&#x2F;modernagile.org&#x2F;</a><p>I think it&#x27;s probably a good reinterpretation of the original manifesto that is a bit more fundamental and a bit more direct. I like the concepts a lot.
ChicagoDaveabout 5 years ago
The Agile paradigm has been corporatized and all of its capabilities turned into heavy process and oversight bullshit. But worst of all, management rarely sees the need to get involved, especially from the financial management perspective. It&#x27;s very often a zombie project management apocalypse in the making.
PeterStuerabout 5 years ago
There is a surplus of &quot;organizer&quot; roles in IT, call it PM&#x27;s, call it Agile Coaches, call it PO&#x27;s ... for every software engineering methodology. This is indeed because these roles are far more accesible than the &#x27;hard&#x27; engineering skills in IT.
say_it_as_it_isabout 5 years ago
How many people here refused advice suggesting to go with one&#x27;s strengths and instead took a different path in life, developed new capabilities, and acquired new strengths? Progress was slow and difficult. Yet, you persevered. You weren&#x27;t agile, were you?
jayd16about 5 years ago
Did anyone actually read as far as the sub-header? &quot;Agile&quot; is used to scapegoat a lot of issues but gender equality is a new on me. Pigeonholing women into non-technical roles is a problem but I don&#x27;t see how dumping agile would solve it.
评论 #22971376 未加载
kchoudhuabout 5 years ago
They&#x27;ve all moved on to newer grifts, and don&#x27;t need competition from legacy grifts.
zwiebackabout 5 years ago
I thought it was dead already. The people that pushed agile where I worked already moved on the next hot thing twice. Now programmers may have a chance to go back to the roots of agile and do it right, without micromanagement from evangelists.
achenatxabout 5 years ago
the essence of the article is that agile is mainly a project management methodology. It is for decomposing tasks into chunks that allow you to deliver something valuable in the given time frame. One of the best parts is that if you run out of time or money, you have something that is usable with what should be the most important features.<p>What it doesnt give any guidance to is how to organize information about how to build the right thing. If a system is very large, without some methods to understand and communicate what you are building, it just becomes ad hoc. Agile is not a product management&#x2F;requirements identification methodology.
postfactoabout 5 years ago
Agile In Their Own Words<p><a href="https:&#x2F;&#x2F;github.com&#x2F;rayfrankenstein&#x2F;AITOW&#x2F;blob&#x2F;master&#x2F;README.md" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;rayfrankenstein&#x2F;AITOW&#x2F;blob&#x2F;master&#x2F;README....</a>
robjanabout 5 years ago
The site has blocked Hong Kong (and possibly other) IP addresses for some reason.<p>Here&#x27;s an archive link <a href="https:&#x2F;&#x2F;archive.is&#x2F;wip&#x2F;OFF0F" rel="nofollow">https:&#x2F;&#x2F;archive.is&#x2F;wip&#x2F;OFF0F</a>
zabilabout 5 years ago
IMHO the Agile manifesto still holds good. It applies to arguments in the article around processes (and tools) to ironically fix itself.
csoursabout 5 years ago
If you consider that science advances, what did Agile improve on, and where has Agile been been improved upon?
a3nabout 5 years ago
Ugh. I sometimes miss software development, but I don&#x27;t miss any of this.
softwarejoshabout 5 years ago
proper agiles a way of life, cramming the unproductive types of people into it was never going to be as good as making a team of agile people
atulatulabout 5 years ago
All movements go too far.- Bertrand Russel
charles_fabout 5 years ago
Agile is not a methodology or a process, it is literally[1] a set of values and principles which can be summarized by &quot;stop being stupid and trying to reproduce what they do in construction engineering, and start discussing about the why and collaborate, doing things that matter, and being flexible, since after all, we&#x27;re doing _soft_ware.<p>The set of methodologies and processes that ensued, as well as the entire industry of coaches, consultants, evangelists and &quot;practitioners&quot; was probably coming from a good place - trying to propose an implementation for how to put in place those abstract values and principles, or to help with that. The commercialization, envy and corruption that followed is what the problem is. &quot;Agile&quot; has become a synonym for &quot;cargo-cult brainless implementation of sprints and stand-ups to copy something called scrum but effectively doing things in a very much waterfall and old fashion that isn&#x27;t even remotely close to the original intent&quot;.<p>The problem is of semantics order. What does need to &quot;die&quot; is the &quot;agile&quot; industry, with cargo-cultish coaches whose only value is to repeat Scrum says X, XP says Y, Agile says Z, my other customers do Ω, without trying to convey the reason behind those implements ; &quot;agile&quot; methodos like &quot;safe&quot; which have very little agility backed in and whose &quot;Enterprise Architects&quot; proponents usually don&#x27;t get any of the intent.<p>But take a look at the manifesto[1] and tell me that you want it to die, I feel like the arguments will be very complex. Re: the that we should become &quot;engineers&quot; (semantics and cargo-cult again), there&#x27;s a very good reason why agile applies well to software, but not to mechanical engineers. The economic equations between those two practices are completely different[2][3], the &quot;engineering&quot; phase of mechanical engineering is cheap compared to the production costs and so iterations have effectively an extremely high overhead cost, whereas the &quot;engineering&quot; phase of software engineering represents most of the cost, and so iteration and adaptability have a very low overhead.<p>But saying &quot;agile needs to die&quot; because of how the &quot;agile industry&quot; corrupted it is similar to saying &quot;email needs to die&quot; because what you see most is spam.<p>[1]: <a href="https:&#x2F;&#x2F;agilemanifesto.org&#x2F;" rel="nofollow">https:&#x2F;&#x2F;agilemanifesto.org&#x2F;</a><p>[2]: <a href="https:&#x2F;&#x2F;www.developerdotstar.com&#x2F;printable&#x2F;mag&#x2F;articles&#x2F;reeves_design.html" rel="nofollow">https:&#x2F;&#x2F;www.developerdotstar.com&#x2F;printable&#x2F;mag&#x2F;articles&#x2F;reev...</a><p>[3]: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=RhdlBHHimeM" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=RhdlBHHimeM</a>
parasubvertabout 5 years ago
I’ve been doing XP for 22 years, and was on Ward’s wiki as the C2 project was being executed, watching the ideas grow in real time. I worked for Pivotal for the last 5 years. Here are my thoughts.<p>Agile alone is insufficient to successful products, which the original proponents would tell you. It’s a body of practices. There are more practices than just those. People seem to want “12 steps to an easy rich life” and want to scream when someone offers only 6 of them.<p>Most of the Agile practices the industry has adopted and evolved without calling them agile. So in that sense, it’s a “success”.<p>But agile coaches often are so focused on the original 20 year old practices and they miss the bigger picture of business and products and supporting technology, and the evolution that’s taken place in each during that time.<p>Any software method that is delivering software for users to consume has to be complemented by outcome-focused approaches like user-centered design, product development, customer development, and budgetary funding &#x2F; management practices that don’t weaponize the use of openness, sharing, and metrics for political purposes that hurts people’s self worth. If you’re missing those, agile isn’t going to help you much.<p>Ultimately just letting engineers be engineers is partly how we got into this mess in the first place. People don’t know how to organize themselves without some kind of constraints: design constraints, time, physics, customer constraints, quality, etc. Talented teams turned loose that ultimately deliver little of value is a cliche at this point. Agile tries to use time as a constraint to limit active work in progress to force the delivery of customer visible &#x2F; valuable results that can be tweaked. It’s not the only way to constrain the problem space, but “cost of delay” does have a fairly universal explanatory value in showing what really matters to a business, as lean product development and manufacturing has shown. But that might not be the outcome you’re looking for.<p>whatever process or practices you do, Agile or not, it has to be organized end-to-end to be tied to some kind of tangible mission or outcome or else it’s just a form of self-deception. For example, if we think products need to be usable and solve actual problems, and that design can evolve, then Product owners need to speak for the Customers and have the power to make decisions. Engineers need to be empowered to make the technical decisions. These seem obvious but they’re rare: committees and political strings attached are the norm. If you get both an empowered balanced team of product, design, and engineering you’ve got potential for an engine of collaboration and learning, which is the whole point. Don’t build projects, build human&#x2F;techno systems that grow and improve and evolve.<p>Agile is also not universally applicable and much of the resentment stems from a religious conversion therapy or Developer Rehab approach to marketing (even though some really could use a stint in rehab). Agile is applicable to some kinds of software (user-Centered software in particular), which happens to be a big chunk of industry. It isn’t applicable to everything, such as deeply technical components, pure or applied scientific R&amp;D, or certain kinds of exploratory work. Nor is it applicable to jobs with set and unchangeable requirements.<p>People are best to understand there are a spectrum of practices and processes suited to different environments and stages of organizational evolution. Or they could reject such complexity and nuance and just adopt SAFe, I guess.
caseymarquisabout 5 years ago
The article mentions Agile as an attempt to bring lean manufacturing methodology to software development. I make software which is used as part of lean manufacturing initiatives (among other initiatives), and see a lot of different manufacturing environments. I want to point out that lean manufacturing is often the wrong tool for producing significantly more of something complex or ramping up production speed of a complex product. In its defense, you could say that lean should be this tool, but is usually so misinterpreted or done so incorrectly that it fails to be this tool. (funny that agile reportedly has the same problem)<p>Generally speaking, lean is a tool for reducing various kinds of &#x27;waste&#x27; and &#x27;doing more with less&#x27;. That&#x27;s the part people seem to focus on anyway. The problem is that people become myopic. Lean becomes the justification for premature optimization and focusing on details that don&#x27;t matter.<p>Sometimes, waste is good. Or, stated differently, not all waste is equal. If you have 5 machines that each produce $1000 of value per hour run, and you&#x27;re attempting to go lean by rearranging them so you can lay off a $20&#x2F;hour employee, then you are focusing entirely on the wrong thing.<p>Lean is often pushed with the idea that employees shouldn&#x27;t be standing around doing nothing, or that you shouldn&#x27;t produce excess material which will sit in a pile. The problem with this is it often fails to account for needed excess capacity, and when something goes wrong, everything falls apart.<p>The correct move (depending on margins) is usually to hire more inexpensive humans to stand around and make sure the expensive machines never stop generating value because you overburdened one of the humans. When you want to make things faster, you don&#x27;t stop producing when the next machine in the process can&#x27;t keep up (producing a pile of unused parts). You buy another machine that performs the next process or figure out what it needs to run faster.<p>Lean is the methodology most people use when they can&#x27;t make big changes that will have a major impact on production. It&#x27;s a tool for making the existing (possibly broken) system function better. It&#x27;s culturally good to focus on reducing waste, but often it&#x27;s the waste you&#x27;re not measuring or thinking about that&#x27;s killing you.<p>Lean isn&#x27;t bad any more than optimization is bad, it&#x27;s just not the tool you typically want to be starting with. I work directly with customers, have no deadlines that aren&#x27;t self imposed, and couldn&#x27;t tell a story point from a scrum master, so I can&#x27;t say if the failures of lean manufacturing implementation extend to agile, but I&#x27;d be curious to hear if they do.<p>A tool I find more useful than lean: <a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Theory_of_constraints" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Theory_of_constraints</a>
jahajaabout 5 years ago
I guess it&#x27;s hard to keep methodologies from being co-opted by management for their own ends, since they do after all have the &quot;power&quot;.
评论 #22970078 未加载
评论 #22970947 未加载
andarleenabout 5 years ago
“and it’s pushing women engineers into non-technical roles” - wtf
评论 #22972339 未加载
tabelleabout 5 years ago
why are people still talking about agile?
评论 #22969966 未加载
评论 #22970247 未加载
评论 #22970133 未加载
评论 #22970170 未加载