[ Pobierz całość w formacie PDF ]
.In thisarticle, How to Manage Geeks, Schmidt describes the notion of a technical career ladder that would put people on track to becomedistinguished engineers:If you don t want to lose your geeks, you have to find away to give them promotions without turning them intomanagers.Most of them are not going to make very goodexecutives and, in fact, most of them would probably turnout to be terrible managers.But you need to give them aforward career path, you need to give them recognition,and you need to give them more money.Twenty years ago, we developed the notion of a dual careerladder, with an executive career track on one side and a tech-nical career track on the other.Creating a technical ladder isa big first step.But it s also important to have other kinds ofincentives, such as awards, pools of stock, and non-financialtypes of compensation.At Novell, we just added a new title:distinguished engineer.To become a distinguished engineer,you have to get elected by your peers.That requirement isa much tougher standard than being chosen by a group ofexecutives.It s also a standard that encourages tech people tobe good members of the tech community.It acts to reinforcegood behavior on everyone s part.2In this way, Google appeals to the hacker mindset these are thetechnology fanatics the company wants to recruit.The assumptionis that geeks are not rewarded by giving them managerial titles.Yes,monetary rewards are important, but nonfinancial types of com-pensation are at least as important.Peer review and respect countfor an awful lot.This peer review approach harks back to the principle of com-petition for honor, which was popular in the literature of the 18th-century Enlightenment; competition for honor was a central theme inthe analyses of merit by Helvétius, Diderot, and the Encyclopédistes.They described merit as being derived from relationships of mutualesteem rather than tokens of honor handed down from above: TrueCoworkers Are the Best Judges 71 glory consists in the regard of people who are themselves worthy ofregard, and only this regard equates to merit, wrote the author ofthe article Esteem in Diderot s Encyclopédie.* Or as Montesquieuexplained in Book III of his Spirit of Laws, Honour sets all the partsof the body politic in motion, and by its very action connects them;thus each individual advances the public good, while he only thinksof promoting his own interest. 3This competition for honor provides an elegant solution to aproblem common to all companies that employ skilled specialists:How do you increase organizational bureaucracy while simultaneouslyproviding opportunities for high-level engineers without shuttlingthem into management positions that will only prevent them fromdoing what they do best?A Tool for Quality ControlPeer review is also a formidable force for ensuring quality.It sup-ports the most important principles of programming standardizeddevelopment and quality control because accepted projects mustmeet accepted standards.Repeated, detailed discussions among colleagues regarding com-pany programs encourage the natural development of a commonvernacular.Any urge toward nonstandard development will be nippedin the bud, as nobody wants to risk having his or her project fail thepeer review test.Everyone knows that their peers are sure to rejecta project that doesn t fit the vernacular.This process solves one of the biggest problems software com-panies face when they innovate: building a Tower of Babel, withlines of products and modules that can t communicate, which canbe expensive both in terms of actual cost and inflexibility.Futuredevelopment will be not only more costly but also limited by the factthat knowledge remains locked in the memories of the developerswho created the code; in effect, development becomes proprietary.* A group of 18th-century intellectuals led by Diderot collaborated to compile the first encyclopediaof science.The project took 26 years to complete (1751 1777), with more than 140 authors writing70,000 articles.Their Encyclopédie contained 26 volumes of text and 11 volumes of illustrative engravings.72 Chapter 6 As owners of the program, these programmers become indispens-able.No one else can upgrade or maintain the software.This createsproblems whenever management wants to reassign team members,terminate employees, or simply negotiate salary increases.Peer reviews also encourage thorough source code documenta-tion.Traditionally, programmers have resisted documenting theircode because the process is time consuming and it interrupts theirworkflow.Instead, they put documenting aside until later, whichoften means never.Programmers subject to peer review, however, must show theircolleagues the algorithm or program modules they are writing.They are required to document their code as they write it, with thedocumentation itself becoming a means of quality control.Qualityis no longer the domain of inspectors, as is standard practice withina traditional company.Instead the quality control is enacted by thelabor force itself, along the lines of the Toyota model.Using peer reviews has other subtle benefits, too.For one, itmodifies management practices and the organizational hierarchy,simplifying large projects by dividing them into smaller pieces.Peersmay be unwilling to examine a very long program because doing somight require too great an investment of their time.As in scientific publishing, where peer review is used to vetarticles before they are assembled into books, Google s peer reviewershave less work to do when projects are subdivided.(This mode ofevaluation also enables Google to track development more closelyand to cancel a project sooner if it isn t viable [ Pobierz caÅ‚ość w formacie PDF ]
zanotowane.pl doc.pisz.pl pdf.pisz.pl trzylatki.xlx.pl
.In thisarticle, How to Manage Geeks, Schmidt describes the notion of a technical career ladder that would put people on track to becomedistinguished engineers:If you don t want to lose your geeks, you have to find away to give them promotions without turning them intomanagers.Most of them are not going to make very goodexecutives and, in fact, most of them would probably turnout to be terrible managers.But you need to give them aforward career path, you need to give them recognition,and you need to give them more money.Twenty years ago, we developed the notion of a dual careerladder, with an executive career track on one side and a tech-nical career track on the other.Creating a technical ladder isa big first step.But it s also important to have other kinds ofincentives, such as awards, pools of stock, and non-financialtypes of compensation.At Novell, we just added a new title:distinguished engineer.To become a distinguished engineer,you have to get elected by your peers.That requirement isa much tougher standard than being chosen by a group ofexecutives.It s also a standard that encourages tech people tobe good members of the tech community.It acts to reinforcegood behavior on everyone s part.2In this way, Google appeals to the hacker mindset these are thetechnology fanatics the company wants to recruit.The assumptionis that geeks are not rewarded by giving them managerial titles.Yes,monetary rewards are important, but nonfinancial types of com-pensation are at least as important.Peer review and respect countfor an awful lot.This peer review approach harks back to the principle of com-petition for honor, which was popular in the literature of the 18th-century Enlightenment; competition for honor was a central theme inthe analyses of merit by Helvétius, Diderot, and the Encyclopédistes.They described merit as being derived from relationships of mutualesteem rather than tokens of honor handed down from above: TrueCoworkers Are the Best Judges 71 glory consists in the regard of people who are themselves worthy ofregard, and only this regard equates to merit, wrote the author ofthe article Esteem in Diderot s Encyclopédie.* Or as Montesquieuexplained in Book III of his Spirit of Laws, Honour sets all the partsof the body politic in motion, and by its very action connects them;thus each individual advances the public good, while he only thinksof promoting his own interest. 3This competition for honor provides an elegant solution to aproblem common to all companies that employ skilled specialists:How do you increase organizational bureaucracy while simultaneouslyproviding opportunities for high-level engineers without shuttlingthem into management positions that will only prevent them fromdoing what they do best?A Tool for Quality ControlPeer review is also a formidable force for ensuring quality.It sup-ports the most important principles of programming standardizeddevelopment and quality control because accepted projects mustmeet accepted standards.Repeated, detailed discussions among colleagues regarding com-pany programs encourage the natural development of a commonvernacular.Any urge toward nonstandard development will be nippedin the bud, as nobody wants to risk having his or her project fail thepeer review test.Everyone knows that their peers are sure to rejecta project that doesn t fit the vernacular.This process solves one of the biggest problems software com-panies face when they innovate: building a Tower of Babel, withlines of products and modules that can t communicate, which canbe expensive both in terms of actual cost and inflexibility.Futuredevelopment will be not only more costly but also limited by the factthat knowledge remains locked in the memories of the developerswho created the code; in effect, development becomes proprietary.* A group of 18th-century intellectuals led by Diderot collaborated to compile the first encyclopediaof science.The project took 26 years to complete (1751 1777), with more than 140 authors writing70,000 articles.Their Encyclopédie contained 26 volumes of text and 11 volumes of illustrative engravings.72 Chapter 6 As owners of the program, these programmers become indispens-able.No one else can upgrade or maintain the software.This createsproblems whenever management wants to reassign team members,terminate employees, or simply negotiate salary increases.Peer reviews also encourage thorough source code documenta-tion.Traditionally, programmers have resisted documenting theircode because the process is time consuming and it interrupts theirworkflow.Instead, they put documenting aside until later, whichoften means never.Programmers subject to peer review, however, must show theircolleagues the algorithm or program modules they are writing.They are required to document their code as they write it, with thedocumentation itself becoming a means of quality control.Qualityis no longer the domain of inspectors, as is standard practice withina traditional company.Instead the quality control is enacted by thelabor force itself, along the lines of the Toyota model.Using peer reviews has other subtle benefits, too.For one, itmodifies management practices and the organizational hierarchy,simplifying large projects by dividing them into smaller pieces.Peersmay be unwilling to examine a very long program because doing somight require too great an investment of their time.As in scientific publishing, where peer review is used to vetarticles before they are assembled into books, Google s peer reviewershave less work to do when projects are subdivided.(This mode ofevaluation also enables Google to track development more closelyand to cancel a project sooner if it isn t viable [ Pobierz caÅ‚ość w formacie PDF ]