TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

How to build silos and decrease collaboration on purpose

111 点作者 labwire超过 3 年前

9 条评论

bostonsre超过 3 年前
It sounds reasonable, but I think a key requirement of operating like this would be to have solid leaders that will guide the organization in the correct way.<p>This is similar to the best blog post I&#x27;ve ever read on organizations by coda hale [1]. His perspective of viewing an organization like a distributed computing system is enlightening and highly pragmatic.<p>Scaling organizations is hard. To scale well, you need to avoid contention on shared resources and continually work on force multiplier type projects.<p>[1] <a href="https:&#x2F;&#x2F;codahale.com&#x2F;work-is-work&#x2F;" rel="nofollow">https:&#x2F;&#x2F;codahale.com&#x2F;work-is-work&#x2F;</a>
评论 #28415311 未加载
评论 #28414040 未加载
skmurphy超过 3 年前
In this model communication has to go up to go across the silo. The &quot;white space&quot; in the organization is filled with poison gas (or strong cultural taboos against &quot;breaking the chain of command&quot;). Before you talk to someone you need to talk to their manager (or your manager must talk to their manager).<p>It&#x27;s an interesting thought experiment for organizational design. If you had enough information you could partition the challenges an organization faces into a set of mutually exclusive collectively exhaustive set and assign one team to each challenge. Of course the world does not stand still and your perfect decomposition of teams to problems starts to go obsolete as soon as you announce it.<p>In &quot;organize for Complexity&quot; (<a href="https:&#x2F;&#x2F;www.amazon.com&#x2F;gp&#x2F;product&#x2F;B010CL1X66&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.amazon.com&#x2F;gp&#x2F;product&#x2F;B010CL1X66&#x2F;</a>) Niels Pfaegling suggests every organization has three kinds of networks:<p>1. Formal Reporting Hierarchy: manages formal budget and compliance with laws and regulations.<p>2. Informal Networks: hold social capital and operate in the “white space.”<p>3. Expertise Networks: hold intellectual capital and procedural knowledge that enables value creation.<p>This model remove the second and third category of networks as options. I suspect that an organization with only a formal hierarchy would be neither robust or resilient in the face of changes in the outside world (e.g new competitors, new opportunities, changes in customer needs, etc..). Christopher Alexander wrote about the value of multiple overlapping networks and a messy but lively organization in &quot;A City is not a Tree&quot; (See <a href="https:&#x2F;&#x2F;www.patternlanguage.com&#x2F;archive&#x2F;cityisnotatree.html" rel="nofollow">https:&#x2F;&#x2F;www.patternlanguage.com&#x2F;archive&#x2F;cityisnotatree.html</a> ). I think his insights also apply to organizations.<p>I blogged about Pflaeging&#x27;s insights in <a href="https:&#x2F;&#x2F;www.skmurphy.com&#x2F;blog&#x2F;2018&#x2F;10&#x2F;24&#x2F;7-sets-of-insights-from-organize-for-complexity-by-niels-pflaeging&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.skmurphy.com&#x2F;blog&#x2F;2018&#x2F;10&#x2F;24&#x2F;7-sets-of-insights-...</a>
评论 #28415233 未加载
评论 #28413970 未加载
alecco超过 3 年前
AFAIR, the original idea of having silos came from Sloan&#x2F;GM. Originally the groups were organized by function to reduce costs and streamline production. Say motors, wheels, windshields, etc. But this created problems because the end product lines didn&#x27;t have much power over these groups. Some of these groups even became defiant as they were in high demand so they played the different product lines against each other.<p>The solution was to integrate vertically from the end product lines. So each of Chevrolet&#x2F;Pontiac&#x2F;Cadillac&#x2F;Buick would have its dedicated teams for all it&#x27;s neededs. No more delays and innovation was much faster. This turned GM from a mess to beating Ford.<p>The problem is this was seen as the ultimate organizational structure and Business Schools love magic recipes.<p>But I&#x27;ve seen the same problems in IT companies with massive Dev, QA, Production teams who end up antagonizing. Prom passive aggression (e.g. delaying approvals) to the occasional openly sabotaging (e.g. approving faulty releases). From experience, QA drifts into being the most political, Dev drifts into carelessness, and Production&#x2F;Ops ends up over-reacting. This can be solved by integrating vertically.<p>But there is no silver bullet in organizational structures. There are many factors like culture, what&#x27;s the product&#x2F;service, the size of the organization, the geographical distribution of the teams and organization, and what kind of company it is (e.g. lean vs. moonshot).
评论 #28413907 未加载
评论 #28420666 未加载
评论 #28413197 未加载
ineedasername超过 3 年前
<i>Collaboration is when people work together to produce an outcome. When teams are collaborating, it means they’re working with other teams to achieve outcomes together. That’s often a sign the team isn’t set up well. Ideally, it should have what it needs to do what it needs without dependencies on other teams.</i><p>I think you&#x27;d need to be very careful about how you implement that philosophy. It has the potential to be a recipe for every team rolling their own solutions to the same problems over &amp; over again with very little individual knowledge making its way to become <i>institutional knowledge</i>.<p>Making every team self sufficient, if not done carefully, also has the potential to be wasteful when it comes to skillsets that the team doesn&#x27;t need on a full-time basis when a few teams could share the same staffer for whatever that skillset happens to be.
评论 #28412690 未加载
评论 #28418827 未加载
polote超过 3 年前
This is a ckickbait title for an article that fails to communicate what it is really about.<p>OP seems to work in an company that give too much freedom to employees and as a result this is impossible to manage them. This has not a lot to do with collaboration.<p>I feel like what OP is trying to describe is &quot;highly aligned, loosely coupled&quot;
评论 #28413296 未加载
jerome-jh超过 3 年前
Companies are like good software: there should be well defined interfaces. Without them, communication as says the article is &quot;many to many&quot;. This leaves a lot of leeway for those who use information for their own profit. And this includes management: &quot;are you telling me you are not aware [you should have done this | we work like that | of this email from customer]?&quot;
评论 #28416181 未加载
ilyash超过 3 年前
&gt; A little collaboration is fine, .... Bezos structured Amazon so that teams were as independent as possible.<p>Maybe too much independent? It looks like AWS had roughly zero collaboration between teams. I wonder how they even get IAM to work with all the services.<p>* Basic things like how you call pagination fields is f%cked up. <a href="https:&#x2F;&#x2F;gist.github.com&#x2F;ilyash&#x2F;f845d050df5c7c0805183790b28ac3f3" rel="nofollow">https:&#x2F;&#x2F;gist.github.com&#x2F;ilyash&#x2F;f845d050df5c7c0805183790b28ac...</a> . That precludes straightforward implementation of wrappers and what not.<p>* Code being published of quality which says (99% sure) it was never reviewed (at least not by someone proficient in programming and the language).<p>* CloudFormation lags supporting new features of other services.
评论 #28415520 未加载
mpweiher超过 3 年前
&quot;The expected benefits of modular programming fall into three classes: (1) managerial -- development time could be shortened because separate groups would work on each module with little need for communication (and little regret afterward that there had not been more communication);&quot;<p>-- D. L. Parnas, <i>ON THE CRITERIA TO BE USED IN DECOMPOSING SYSTEMS INTO MODULES</i>, 1971, <a href="https:&#x2F;&#x2F;prl.ccs.neu.edu&#x2F;img&#x2F;p-tr-1971.pdf" rel="nofollow">https:&#x2F;&#x2F;prl.ccs.neu.edu&#x2F;img&#x2F;p-tr-1971.pdf</a>
xsquared2超过 3 年前
“Otherwise, the communication burden on teams will grow at an exponential rate” Actually just quadratic.
评论 #28413026 未加载