Who you have is much more important than their seniority level. Select the right people for the job based on their skills and ability to work together, not on some set of buckets labeled 'senior' and 'junior'. Make sure you understand what it is you are trying to build, how design improvements/alterations will be ratified by the group, and how disagreements will be unwound and resolved.
You need a boss, usually, unless people are okay with being very open about decisions. The boss's job is also to act as a conduit to the rest of the organization. Go read all of Rands In Repose.
Start the team as small as possible. Lots of great software, and many great sites, are the result of a 1-3 person team who really work well together, understand their market, and understand the architecture. ("Never doubt that a small group of thoughtful, committed people can change the world. Indeed, it is the only thing that ever has." -- Margaret Mead) Be loathe to add more people. Beware the n^2 law of communication: for n people, there are O(n^2) channels of communication and relations to manage. While group meetings help this a bit, there is always a need for 1-1s.
And beware the Mythical Man Month and Brook's Law: adding more people won't accelerate a project go as much as you think it will, and can in fact slow things down. The cost of overhead rapidly overtakes the cost of development. Definitely read that book before doing this sort of planning!
Gender has complex interactions, but again, it is more important to look at the specific people in place rather than seeing them as simply an example of their gender. The Liskov Substitution Principle does NOT apply to humans.
One rule I always keep in mind: The team structure and players WILL change, let's not get to hung up on the current one too much. Figure out what you need NOW, adjust and move forward.
Manage the team structure with skills you need. Some senior people can handle lots of juniors (freshers) to mentor/manage, some can NOT.
In no order or anything:(SOME folks can do all, some only one)
Mentors: teach as well as do (what, how)
Contributers - just do
Architect - know WHAT to do (can also do in Contribute role)
Lead - recognize tasks, can assign and guide (Contribute and Mentor)
Managers - challenge, point, then stay out of the way
Atleast 25% of the team should be Senior software engg. with experience of 4 years and more. 60-70% can be software engg. with experience between 1-3 years and the rest can be freshers.
Assign the role and responsibilities accordingly. Further if there are resource avaliable in the same domain as that the project give them weight-age when deciding the team.
Also I think that freshers should be involved in the development of any project. Though they may be assigned easy tasks, but remember you should always think them as future investment for the project. For once the project is complete you may need resources for its maintenance.
Also sometimes the marital status also effects the decision to take a resource or not. It depends upon the responsibilities and workload involved needed in the project. If the project involves late night shifts or frequent client visits then this is always considered especially when the client is geographically far.
Depends on your own preferences but ask yourself this: do you really want people mentoring others at the start of a project? Or do you want the project written and written well?
Get as many seniors as you can afford and maybe have one or two very bright students if you can find them.
说实话，我同意其他海报的观点。比率没有意义。您想要组建尽可能最好的团队。聘请 N 个最好的受访者并完成项目。唯一的限制是现金流，但总支出肯定低于任何其他团队结构。
The rule used to be 'just enough people so they can jump into the VW convertible and go get a pizza'.
Note the corollaries:
(a) someone has to have a VW convertible (b) they all have like pizza (c) they all have to want to eat together
You can get (a) and (b) fairly easily. (c) is pretty tough these days.
Seriously I agree with the other posters. Ratios are meaningless. You want the best possible team you can put together. Hire the N best interviewees and get the project done. Your only constraint on this is cash flow but the total outgoings are certain to be less than for any other team structure.
发布评论
评论（5）
你拥有的人比他们的资历级别重要得多。根据他们的技能和合作能力来选择适合该工作的人员，而不是根据一些标记为“高级”和“初级”的桶来选择。确保您了解您正在尝试构建什么，团队将如何批准设计改进/变更，以及如何消除和解决分歧。
通常，你需要一个老板，除非人们愿意公开做出决定。老板的工作也是充当与组织其他部门的沟通渠道。去阅读《安息中的兰德》的全部内容吧。
组建尽可能小的团队。许多优秀的软件和许多优秀的网站都是 1-3 人团队的成果，他们真正合作良好，了解他们的市场，并了解架构。 （“永远不要怀疑一小群有思想、有奉献精神的人可以改变世界。事实上，这是唯一曾经拥有过的东西。”——玛格丽特·米德）不愿意增加更多的人。请注意 n^2 沟通法则：对于 n 个人，有 O(n^2) 个沟通渠道和需要管理的关系。虽然小组会议对此有所帮助，但一对一交流始终是必要的。
请注意人月神话和布鲁克定律：增加更多人并不会加速项目会按照你想象的那样进行，而且实际上会减慢速度。管理费用迅速超过了开发成本。 一定在做这种计划之前先读一下那本书！
性别之间存在复杂的相互作用，但同样，更重要的是要关注特定的人，而不是将他们简单地视为其性别的一个例子。里氏替换原理不适用于人类。
祝你好运！
Who you have is much more important than their seniority level. Select the right people for the job based on their skills and ability to work together, not on some set of buckets labeled 'senior' and 'junior'. Make sure you understand what it is you are trying to build, how design improvements/alterations will be ratified by the group, and how disagreements will be unwound and resolved.
You need a boss, usually, unless people are okay with being very open about decisions. The boss's job is also to act as a conduit to the rest of the organization. Go read all of Rands In Repose.
Start the team as small as possible. Lots of great software, and many great sites, are the result of a 1-3 person team who really work well together, understand their market, and understand the architecture. ("Never doubt that a small group of thoughtful, committed people can change the world. Indeed, it is the only thing that ever has." -- Margaret Mead) Be loathe to add more people. Beware the n^2 law of communication: for n people, there are O(n^2) channels of communication and relations to manage. While group meetings help this a bit, there is always a need for 1-1s.
And beware the Mythical Man Month and Brook's Law: adding more people won't accelerate a project go as much as you think it will, and can in fact slow things down. The cost of overhead rapidly overtakes the cost of development. Definitely read that book before doing this sort of planning!
Gender has complex interactions, but again, it is more important to look at the specific people in place rather than seeing them as simply an example of their gender. The Liskov Substitution Principle does NOT apply to humans.
Good luck!
我始终牢记一条规则：球队结构和球员都会改变，我们不要太执着于当前的结构和球员。弄清楚你现在需要什么，调整并继续前进。
用您需要的技能管理团队结构。一些高级人员可以指导/管理大量初级人员（新生），有些则不能。
没有顺序或任何事情：（有些人可以做所有的事情，有些人只能做一个）
担任贡献角色）
One rule I always keep in mind: The team structure and players WILL change, let's not get to hung up on the current one too much. Figure out what you need NOW, adjust and move forward.
Manage the team structure with skills you need. Some senior people can handle lots of juniors (freshers) to mentor/manage, some can NOT.
In no order or anything:(SOME folks can do all, some only one)
do in Contribute role)
团队中至少 25% 的人员应为高级软件工程师。拥有4年以上经验。
60-70% 可以是软件工程师。具有1-3年经验，其余的可以是新人。
相应地分配角色和职责。此外，如果项目所在的同一领域有可用的资源，那么在决定团队时会给他们权重年龄。
我还认为新生应该参与任何项目的开发。尽管他们可能被分配了简单的任务，但请记住，您应该始终将它们视为项目的未来投资。一旦项目完成，您可能需要资源来维护它。
有时，婚姻状况也会影响是否获取资源的决定。这取决于项目所需的职责和工作量。如果项目涉及深夜轮班或频繁拜访客户，那么总是会考虑这一点，特别是当客户地理位置较远时。
Atleast 25% of the team should be Senior software engg. with experience of 4 years and more.
60-70% can be software engg. with experience between 1-3 years and the rest can be freshers.
Assign the role and responsibilities accordingly. Further if there are resource avaliable in the same domain as that the project give them weight-age when deciding the team.
Also I think that freshers should be involved in the development of any project. Though they may be assigned easy tasks, but remember you should always think them as future investment for the project. For once the project is complete you may need resources for its maintenance.
Also sometimes the marital status also effects the decision to take a resource or not. It depends upon the responsibilities and workload involved needed in the project. If the project involves late night shifts or frequent client visits then this is always considered especially when the client is geographically far.
取决于您自己的喜好，但问问自己：您真的希望人们在项目开始时指导其他人吗？还是希望项目写得好、写得好？
尽可能多地招募高年级学生，如果你能找到的话，也许还有一两个非常聪明的学生。
Depends on your own preferences but ask yourself this: do you really want people mentoring others at the start of a project? Or do you want the project written and written well?
Get as many seniors as you can afford and maybe have one or two very bright students if you can find them.
过去的规则是“只要有足够的人，他们就可以跳进大众敞篷车去买披萨”。
注意推论：
(a) 某人必须拥有一辆大众敞篷车
(b) 他们都喜欢披萨
(c) 他们都必须想一起吃饭
你可以很容易地得到 (a) 和 (b)。 (c) 这些天相当艰难。
说实话，我同意其他海报的观点。比率没有意义。您想要组建尽可能最好的团队。聘请 N 个最好的受访者并完成项目。唯一的限制是现金流，但总支出肯定低于任何其他团队结构。
The rule used to be 'just enough people so they can jump into the VW convertible and go get a pizza'.
Note the corollaries:
(a) someone has to have a VW convertible
(b) they all have like pizza
(c) they all have to want to eat together
You can get (a) and (b) fairly easily. (c) is pretty tough these days.
Seriously I agree with the other posters. Ratios are meaningless. You want the best possible team you can put together. Hire the N best interviewees and get the project done. Your only constraint on this is cash flow but the total outgoings are certain to be less than for any other team structure.