UMLChina第十三期专家交流实录
|
UMLChina第十三期专家交流实录 |
北京时间2002年1月10日（星期四）晚上21:30-23:30
嘉宾：Roser S．Pressman博士是软件工程领域国际知名的咨询专家和作者，R.S. Pressman & Associates, Inc的创始人，已有超过30年的产业经验，主要从事工程产品软件和系统软件的开发技术工作和管理工作。 作为产业的实践者和管理者，Pressman博土主要从事的是航空航天应用中高级工程和制造的CAD/CAM系统的开发，他也从事科学及系统程序设计方面的工作。 除了他的产业经验之外，Pressman博士还是BridgePort大学计算机工程系的兼职教授和该大学的计算机辅助设计和制造中心的主任。 Pressman博士是杂志American Programmer（美国程序员）和 IEEE Software（IEEE软件）的编委，是IEEE Software的Manager（管理员）专栏的编辑。他还是ACM、IEEE、Tau Beta Pi、 Phi Kappa Phi、Eta Kappa Nu和Pi Tau Sigma的会员。
Pressman博士著有6本书籍，其中最著名的是已经发行了20多年的“Software Engineering: A Practitioner's Approach”，此书被全世界广泛使用为软件工程教材，最新版为第五版。第五版影印本由清华大学出版社发行，第四版中文译本“软件工程--实践者的研究方法”由机械工业出版社发行。
交流重点：软件工程，软件过程改进...
主持人：钱五哥
网址：焦点网UMLChina小组聊天室（必须登录焦点网方可进入）
以往聊天实录
http://www.umlchina.com/Chat/Newindex1.htm
注：【 】中是qlw5（http://qlw9.myetang.com）在结束后的翻译，比较匆忙，有一些回答没有翻译，见谅
– qlw5
rogerpressman进入聊天室.(2002/01/10
21:21)
csliuyb对大家说: hello roger(2002/01/10 21:22)
seeseax对大家说: not as many persons as I imagine(2002/01/10 21:22)
rogerpressman对大家说: Hello
all. I'm a little early!(2002/01/10 21:22)
vickyday对rogerpressman说: Hello! R U the one we are waiting for?(2002/01/10
21:22)
umlchina对大家说: hello(2002/01/10 21:22)
umlchina对大家说:
hello(2002/01/10 21:23)
umlchina对大家说: 嘉宾到了(2002/01/10
21:23)
brook0344对大家说: I do not get anything to ask,i just want to get some
useful idea(2002/01/10 21:23)
umlchina对大家说: 大家欢迎(2002/01/10
21:23)
siltboy对rogerpressman说: hello!(2002/01/10 21:23)
seeseax对大家说: welcome so much!!(2002/01/10 21:23)
goaha进入聊天室.(2002/01/10 21:23)
luv对大家说: hello Master(2002/01/10 21:23)
jameshwoo对rogerpressman说: How
are you, Dr. Pressman?(2002/01/10 21:23)
mhuang进入聊天室.(2002/01/10 21:23)
brook0344对大家说: anybody the same as me?(2002/01/10 21:23)
rogerpressman进入聊天室.(2002/01/10 21:24)
h_gaofeng对大家说: hello,rogerpressman(2002/01/10 21:24)
javaor进入聊天室.(2002/01/10 21:24)
brook0344对大家说: hello,master(2002/01/10 21:25)
oysterlqd进入聊天室.(2002/01/10 21:25)
rogerpressman对大家说: Sorry, My server is acting strangly. Do you know what
Murphy's law is :-)(2002/01/10 21:25)
siltboy进入聊天室.(2002/01/10 21:25)
rogerpressman对大家说: Does
anyone have any question of software engineering?(2002/01/10 21:25)
mhuang对大家说: hello everybody(2002/01/10 21:25)
csliuyb对大家说: Dr Pressman,
In your opinion, what is the relationship between programing and software
engineering?(2002/01/10 21:26)
seeseax微笑着对rogerpressman说: could u please explain(2002/01/10 21:26)
xuefg进入聊天室.(2002/01/10 21:26)
xuefg对大家说: good night!(2002/01/10 21:26)
h_gaofeng对xuefg说: who are you(2002/01/10 21:26)
mhuang对csliuyb说: maby is is a big topic(2002/01/10 21:27)
xuefg对大家说: i am a programmer(2002/01/10 21:27)
umlchina对rogerpressman说: how do
you think about XP?(2002/01/10 21:27)
csliuyb对rogerpressman说: Dr
Pressman, In your opinion, what is the relationship between programing and
software engineering?(2002/01/10 21:27)
rogerpressman进入聊天室.(2002/01/10 21:27)
yin_feng进入聊天室.(2002/01/10 21:28)
yin_feng进入聊天室.(2002/01/10 21:28)
Charity_Zhou进入聊天室.(2002/01/10 21:28)
vickyday进入聊天室.(2002/01/10 21:28)
derta进入聊天室.(2002/01/10 21:28)
tsino进入聊天室.(2002/01/10 21:29)
h_gaofeng对rogerpressman说: How do you think about RUP and CMM?(2002/01/10
21:29)
umlchina对大家说:
嘉宾链接有问题，稍等(2002/01/10 21:29)
hyuan66进入聊天室.(2002/01/10 21:29)
siltboy对umlchina说: I can't say anthing.(2002/01/10 21:29)
rogerpressman对大家说: SE is a
discipline that requires a process. Programming is a bit less formal. Both work
to accomplish the same thing.(2002/01/10 21:29)
【SE需要过程的约束，写程序相对非正式一些。两种工作的目标是相同的】
siltboy对umlchina说: sorry,it works.(2002/01/10 21:29)
silentsnake进入聊天室.(2002/01/10 21:30)
skyoldx离开了聊天室.(2002/01/10 21:30)
skyoldx进入聊天室.(2002/01/10 21:30)
rogerpressman对大家说: Any
process model (and the related methods to assess maturity are worthwhile.
However, it is important to note that process can be overkill. Every
organization must choos(2002/01/10 21:31)
【任何过程模型和评估成熟度的相关方法都是有意义的。但是必需避免过分强调过程。每个机构必需自己选择…】
umlchina对rogerpressman说: how do
you think about XP?((2002/01/10 21:31)
vickyday进入聊天室.(2002/01/10 21:31)
fcx123进入聊天室.(2002/01/10 21:31)
umlchina对rogerpressman说:
eXtreme Programming(2002/01/10 21:31)
rogerpressman对大家说: XP is an
excellent idea, but it is NOT a replacement for SE discipline. It is a variant
of SE, in my opinion.(2002/01/10 21:31)
【XP是优秀的想法，但它并非SE纪律的替代品。我认为它是SE的一个变种。】
zydedueast进入聊天室.(2002/01/10 21:31)
yin_feng进入聊天室.(2002/01/10 21:32)
h_gaofeng对rogerpressman说: If we don't make request clear,but manager order
we must begin to design and code,what do we do?(2002/01/10 21:32)
Charity_Zhou对rogerpressman说: how to choose an appropriate process
model?(2002/01/10 21:32)
anon进入聊天室.(2002/01/10 21:32)
atlas_tao对rogerpressman说:
what's the position of COM in SE?(2002/01/10 21:32)
tsino离开了聊天室.(2002/01/10 21:32)
brook0344对大家说: yeah,It's not so formal as se(2002/01/10 21:32)
umlchina对大家说:
提问用蓝色(2002/01/10 21:32)
rogerpressman对大家说: You must
understand requirements, but there are many ways to do this. One is to
prototype. that satisfied the manager, but also gets you info(2002/01/10 21:33)
【你必需理解需求，但是有很多方法来完成需求。一种是原型法，此法使经理们满意，但也需要你的信息】
yin_feng进入聊天室.(2002/01/10 21:33)
javaor对rogerpressman说: Dr
Pressman.Do you think J2EE is better than .NET in MIS(2002/01/10 21:33)
vickyday离开了聊天室.(2002/01/10 21:33)
Charity_Zhou对rogerpressman说:
how to choose an appropriate process model for a team or a project?(2002/01/10
21:33)
vickyday进入聊天室.(2002/01/10 21:33)
qlw5进入聊天室.(2002/01/10 21:34)
netd进入聊天室.(2002/01/10 21:34)
vickyday对rogerpressman说: Could you give us some direction about software
size estimate?(2002/01/10 21:34)
wenjiamin进入聊天室.(2002/01/10 21:34)
rogerpressman对大家说: Choosing
the right process model requires a knowledge of the people on the team, the
nature of the project and many other factors. Visit http://www.rspa.com/apm(2002/01/10
21:34)
【选择恰当的过程模型决定于团队成员的知识，项目特征和很多其他因素，参见http://www.rspa.com/apm】
siltboy对rogerpressman说: I have
a fact question. A is a class,B is anther.But A is used in B's funciton.How to
describe the relation?(2002/01/10 21:34)
qlw5对大家说: helloe everyone, sorry for my late(2002/01/10 21:34)
h_gaofeng对rogerpressman说: When we begin to program,we should get rid off
prototype or not?(2002/01/10 21:34)
vickyday对rogerpressman说: Could
you give us some direction about software size estimate?(2002/01/10 21:34)
openeyes进入聊天室.(2002/01/10 21:35)
liuzhiming进入聊天室.(2002/01/10 21:35)
zhangqike进入聊天室.(2002/01/10 21:35)
rogerpressman对大家说: Software
size: There are two approaches. One is LOC counting. Today, this is widely used
bu suspect.(2002/01/10 21:35)
rogerpressman对大家说: The other
is FP. This is not easily applied in engineering-type apps, but can be useful in
IT apps(2002/01/10 21:36)
【软件规模：有两种方法，一种是LOC。目前广为使用。另外一种是FP，它不便应用于工程性项目，但完全适用于IT应用】
qlw5对大家说: dear rogerpressman, what does suspect mean?(2002/01/10 21:36)
zik进入聊天室.(2002/01/10 21:36)
xuefg进入聊天室.(2002/01/10 21:36)
xuefg微笑着对qlw5说:
请问你的主页没有了吗？我怎么上不去？(2002/01/10 21:36)
vickyday对rogerpressman说: But
how to estimate more effectively?(2002/01/10 21:36)
xchyj进入聊天室.(2002/01/10 21:37)
qlw5对大家说:
主页依然存在,不过服务器限制了IP(2002/01/10 21:37)
rogerpressman对大家说: ISuspect
means that it's difficult to determine what a LOC is in many modern languages
(e.g., visula app development). However, it can be used effectively, if proper
conv(2002/01/10 21:37)
【我怀疑很多新型语言，如可视化编程，中很难使用LOC，但是…】
seeseax对rogerpressman说:
What’s the next generation UML, what's new it will bring to us?(2002/01/10
21:37)
rogerpressman对大家说: To
estimate effectively, you must develop estimates in more than one way. You then
reconcile the different estimates.(2002/01/10 21:37)
【为了更有效地估计，你必需用多种方法来估计。然后整合各个估计值】
csliuyb进入聊天室.(2002/01/10 21:38)
vickyday对rogerpressman说: We are
using DELphi method to make the results more creditable(2002/01/10 21:38)
sun221进入聊天室.(2002/01/10 21:38)
xuefg微笑着对qlw5说:
怎么上？(2002/01/10 21:38)
qlw5对大家说: [to jameshwoo]我不是今天的guest,不过希望你课程和实践结合紧密(2002/01/10
21:38)
vickyday对rogerpressman说: But FP
sounds like too delicate and complex(2002/01/10 21:38)
rogerpressman对大家说: UML is an
important analysis and design approach. Altho it's used in may placed, there are
many more SEs who do not use it at all. Before we worry about the next
generati(2002/01/10 21:38)
【UML是很重要地分析设计方法，虽然很多项目使用它，仍有很多SE从来不用UML，因为他们担心其前景，我希望看到现有方法的广泛应用】
goldarcher进入聊天室.(2002/01/10 21:39)
rogerpressman对大家说: Before we
worry about the next generation, I'd like to see widerusage of the current
model(2002/01/10 21:39)
Charity_Zhou对rogerpressman说: I
am reading your book <<Software Engineering>> 5th,any advice about
your book?(2002/01/10 21:39)
rogerpressman对大家说: Re my
book: it's very long :-). Pick and choose the info you need. Also visit the web
site.(2002/01/10 21:40)
【关于我的书：很长:-】，选择你想看的东西，留心相关网站)
csliuyb进入聊天室.(2002/01/10 21:40)
xuefg进入聊天室.(2002/01/10 21:40)
xuefg微笑着对rogerpressman说:
how are you thank about cmm?(2002/01/10 21:40)
jameshwoo对rogerpressman说: Do
you think a person mainly having mechanical engineering background may become a
good software engineering professional?(2002/01/10 21:40)
rogerpressman对大家说: The CMM
is a worthwhile process model, but it is NOT a religion. You must apply it with
common sense.(2002/01/10 21:41)
【CMM是很好的模型，但决非宗教信仰。必需以平常心看待它】
zik对rogerpressman说: How about
Rational Unified Process?(2002/01/10 21:41)
vickyday对rogerpressman说: Do u
think the tools developed by Rational will do great help to the engineering of
the process?(2002/01/10 21:41)
qlw5对大家说: Dear jameshwoo,
do yo uthink CMM is applicable for most project, which project type is the most
suitable?(2002/01/10 21:42)
rogerpressman对大家说: Many year
ago, as a young man, my first engineering degree was in ME. What a
coincidence!!(2002/01/10 21:42)
【很多年前，偶还年轻的时候，我第一个工程学位是ME。真是巧合！！】
y_z进入聊天室.(2002/01/10 21:42)
siltboy对rogerpressman说: how do
you think test in software engineering? and now ,what do the American Companies
test their product?(2002/01/10 21:43)
rogerpressman对大家说: The CMM
is applicable to all projects. However, it may be overkill for some rapid
response projects and smaller projects(2002/01/10 21:43)
【CMM可以应用在各种项目中，但是对于快速响应性和小型项目未必适用】
siltboy对rogerpressman说: I think
testing is very important in future.(2002/01/10 21:43)
qlw5对rogerpressman说: Dear
rogerpressman, do you think new project can use CMM Model?(2002/01/10 21:43)
rogerpressman对大家说: Testing
is the weakest of all SE disciplines, yet is is the activity that absorbs the
most time. We need good strategies and very good tactics.(2002/01/10 21:44)
【测试是SE中最弱的一环，也是最耗时间的一环。我们需要耗的策略和战术】
netd对rogerpressman说: Do you
think software engineering really can solve the problem of software?(2002/01/10
21:44)
longzhifang进入聊天室.(2002/01/10 21:44)
qlw5对rogerpressman说: sorry, I
do not know who is jameshwoo?(2002/01/10 21:44)
rogerpressman对大家说: A new
project can use the CMM model.(2002/01/10 21:44)
【新型项目可以使用CMM模型】
vickyday对rogerpressman说: Do
u think the tools developed by Rational will do great help to the engineering of
the process?(2002/01/10 21:44)
siltboy对rogerpressman说: Now,I
don't know how to do a good thing in OOT.(2002/01/10 21:45)
xuefg微笑着对rogerpressman说:
now can i thank that CMM is not suit smaller projects?(2002/01/10 21:45)
siltboy进入聊天室.(2002/01/10 21:45)
detective进入聊天室.(2002/01/10 21:45)
xuefg进入聊天室.(2002/01/10 21:45)
qlw5对rogerpressman说: however I
do not think pure new project can use CMM, the reason it that it is not
repeatable?(2002/01/10 21:45)
rogerpressman对大家说: Rational
tools are extremely useful. The add to the SEs insight. But it is more important
the the SE himself/herself understand good practices.(2002/01/10 21:45)
【Rational工具极为有用，他们加深人们对SE的了解。但更重要的是SE自己要掌握好的实践】
siltboy对rogerpressman说: Can you
give some advices?(2002/01/10 21:45)
siltboy对rogerpressman说: Can you
give some advices in OO Testing?(2002/01/10 21:46)
mhuang对大家说: how do you
think about the importance regarding a architecture or infrastructure to an
application development?(2002/01/10 21:46)
csliuyb进入聊天室.(2002/01/10 21:47)
Charity_Zhou对rogerpressman说:
Accoding to Jacobson's talking,in the future the analysis will take the larest
part of the whole process ,how about ur think?(2002/01/10 21:47)
rogerpressman对大家说: OO
testing is different than conventional testing. See my book, chapter 23 for a
complete discussion.(2002/01/10 21:47)
【OO测试与传统测试不同，参见拙著的第23章】
sun221进入聊天室.(2002/01/10 21:47)
brook0344对大家说: hi,what does
the key "smaller projects" in your opinion?(2002/01/10 21:47)
brook0344对大家说: mean?(2002/01/10 21:47)
atlas_tao进入聊天室.(2002/01/10 21:47)
atlas_tao对rogerpressman说: What
do you think about MS COM?(2002/01/10 21:47)
siltboy对rogerpressman说: And I
can't get difference of OO Intergration Test and Unit Test. How do you
think?(2002/01/10 21:47)
rogerpressman对大家说:
Architecture has become one of the most important design ideas. Infrastruture
depends on good architecture.(2002/01/10 21:48)
【体系结构已经成为一种最重要的设计观念。底层结构依赖于好的体系结构】
Charity_Zhou对rogerpressman说:
the same as RUP says?(2002/01/10 21:48)
csliuyb对大家说: Could you tell
me what kind of aspects does the SE research efforts interested in
now?(2002/01/10 21:48)
jeffray进入聊天室.(2002/01/10 21:49)
rogerpressman对大家说: In
general "Smaller projects" involve 2 - 3 people for no more than 1 to
2 months and are NOT business critical.(2002/01/10 21:49)
【一般来说，小型项目是2～3个人、不超过2个月、也不是关键项目】
oldblue进入聊天室.(2002/01/10 21:49)
netd对rogerpressman说: What dose
Infrastructure means? Can you give me an example to explain the difference of
infrastructure and architecture?(2002/01/10 21:49)
brook0344对大家说: 好的(2002/01/10
21:49)
rogerpressman对大家说: You
question on OO testing are good one, but are beyond the scope of a chat. See my
book, where your question is answered.(2002/01/10 21:50)
【你关于OO的问题很好，但这超出乐本次讨论的范围，参见我的书，里面回答了你的问题】
qlw5对rogerpressman说: Dear
rogerpressman, By my experience, I found out that although CMM is a general
approach, it is actually based on waterfall lifecycle model, what is your
opinion?(2002/01/10 21:50)
qlw5进入聊天室.(2002/01/10 21:50)
rogerpressman对大家说: Stated
very simply, architecture describes the relationship among processing and data
components in a system, how they communicate and how they interact.(2002/01/10
21:50)
【简单的说，体系结构描述了系统中处理和数据部件间的关系，关于它们的通信方式和交互方式】
jeffray对大家说: hi I am
coming(2002/01/10 21:51)
mhuang进入聊天室.(2002/01/10 21:51)
mhuang对大家说: will you use a
free and open source infrastructure in your project?(2002/01/10 21:51)
rogerpressman对大家说:
Infrastructure defines the support systems that enable the software to work
(e.g., operation systems, external components, etc.(2002/01/10 21:51)
【底层结构定义了支持软件工作的系统（例如，操作系统，外部部件等等）】
siltboy对rogerpressman说: Thank
you very much!(2002/01/10 21:51)
liyajia129进入聊天室.(2002/01/10 21:52)
jameshwoo对rogerpressman说: Dear
Pressman, do you have any opinion on educating of software engineering,
especially in china?(2002/01/10 21:52)
rogerpressman对大家说: I don't
agree that the CMMis based on the waterfall. It defines those aspects of process
that should exist in every software organization.(2002/01/10 21:52)
【我不同意CMM是基于瀑布模型。它定义了所有软件机构中过程的特征】
csliuyb进入聊天室.(2002/01/10 21:52)
jeffray对rogerpressman说: hi ,I
am one of your older readers :)(2002/01/10 21:52)
csliuyb对大家说: Could you tell
me what kind of aspects does the SE research efforts interested in
now?(2002/01/10 21:53)
seeseax对rogerpressman说: How do
you think about PSP (personal software process)?(2002/01/10 21:53)
rogerpressman对大家说: SE
education is always a challengfe. It is important to get a good background in
computer science, to get pactical experience, and to learn how to work as a
team.(2002/01/10 21:53)
【SE教育具有挑战性。最好有计算机科学背景、实际工作经验并且了解如何协作】
Charity_Zhou对rogerpressman说: a
question about LINUX,how about your think of LINUX to used in important
server,such as DB server(2002/01/10 21:54)
y_z对rogerpressman说: Is Rational
tools good at real time area?(2002/01/10 21:54)
rogerpressman对大家说:
Humphrey's work on PSP and TSP are excellent. I do believe however, that a great
amount of effort (by the SE) are required to apply PSP properly.(2002/01/10
21:54)
【Humphrey关于PSP和TSP的工作非常杰出。但我坚信（SE）需要大量时间来正确实施PSP】
javaor对rogerpressman说: Dr
Pressman.How can a Domain professioner to be helpful to SE?Use UML?any advise is
expected.(2002/01/10 21:54)
rogerpressman对大家说: If the
efort is applied, the results are very good.(2002/01/10 21:54)
【如果投入大，效果应该很好】
jeffray对rogerpressman说: How do
you think about extreme programming(XP)?(2002/01/10 21:55)
siltboy对rogerpressman说: Do you
think the software reuse is related with software complex ?(2002/01/10 21:55)
rogerpressman对大家说: SE
research today focuses on componet-based development, formal methods and many
other areas.(2002/01/10 21:55)
【目前SE研究集中在基于部件的开发、形式化方法和其他领域】
liyajia129进入聊天室.(2002/01/10 21:55)
qlw5对rogerpressman说: I agree
with you, but CMM need plans, but plans is only applicable while the project do
not change often(2002/01/10 21:55)
liyajia129对rogerpressman说: a
question about RUP ?(2002/01/10 21:55)
csliuyb进入聊天室.(2002/01/10 21:56)
csliuyb对大家说: but as post
graduate students just finished 4 years of studying, how can they gether the
experience?(2002/01/10 21:56)
sun221进入聊天室.(2002/01/10 21:56)
rogerpressman对大家说: CMM
implies that planning is necessary, but that doesn't mean that we can't change
the plans as we learn more.(2002/01/10 21:56)
【CMM意味着策划是必需的，但是但它并不意味着我们不能改进计划】
forestinnorway进入聊天室.(2002/01/10 21:57)
jameshwoo对csliuyb说: I agree
with you.(2002/01/10 21:57)
wenjiamin进入聊天室.(2002/01/10 21:57)
wenjiamin对rogerpressman说:
how do you think about XP and RUP?(2002/01/10 21:57)
rogerpressman对大家说: Software
resue should REDUCE software complexity. However, this assume that the resuable
compoents are well designed, well tested and appropriate for the
project.(2002/01/10 21:57)
【软件复用应该可以降低软件复杂性，但是这假定部件具有良好设计、良好测试并适用于项目】
vickyday进入聊天室.(2002/01/10 21:57)
vickyday对rogerpressman说: I
think SE is quite a practical subject.If I want to pusue my study in USA for SE,
could u give me some advice?(2002/01/10 21:57)
y_z对rogerpressman说: Is Rational
tools good at real time area? Or Rhapsody is better.(2002/01/10 21:57)
qlw5对rogerpressman说: Too many
changes means that we need aonther plan, which implies a new development model,
CMM is developmented 10 years ago, why we must use it?(2002/01/10 21:58)
Charity_Zhou对rogerpressman说:
all of us know LINUX is maintanced by people all around the world, how about
this model of software development(2002/01/10 21:58)
rogerpressman对大家说: There are
many universities in the USA that have SE curricula. Use the internet to find
those that have programs of interest.(2002/01/10 21:58)
【美国很多大学都有SE课程，到网上查去吧】
wenjiamin对rogerpressman说: I
think XP is equal to confusion ,and you ?(2002/01/10 21:59)
qlw5对rogerpressman说:
Charity_Zhou, you can find many development model in the internet, the SCM is
strong inside the open source development(2002/01/10 21:59)
jameshwoo对大家说: also care
about MIT's action.(2002/01/10 21:59)
rogerpressman进入聊天室.(2002/01/10 22:00)
rogerpressman对大家说: I never
said that you MUST use the CMM. What you should do is to apply the spirit of the
CMM and adapt it to build a process that is right for your company.(2002/01/10
22:00)
【我从来没有说必需用CMM。你要做的是采用CMM的精髓并根据实际情况来使用】
vickyday进入聊天室.(2002/01/10 22:00)
vickyday对rogerpressman说: If u r
the professor, what would u expect for me to be ur student? If I have CS
background and some working experience for SE, is that OK?(2002/01/10 22:00)
rogerpressman对大家说: If XP is
used to avoid discipline, I agree that it can cause confustion. However, if XP
is used in a disciplined manner it can be quite effective.(2002/01/10 22:00)
【如果说XP是为了避免规矩，我同意这会造成迷惑。但是如果XP被规范地使用，它还是很有效率地】
wenjiamin进入聊天室.(2002/01/10 22:00)
rogerpressman对大家说: I
particularly like the pair programming concept in XP.(2002/01/10 22:01)
【偶尤其喜欢XP中的配对编程】
vickyday对jameshwoo说: What do u
mean as MIT's action?(2002/01/10 22:01)
overflight进入聊天室.(2002/01/10 22:01)
jameshwoo对大家说: Open Course
Library Plan.(2002/01/10 22:01)
rogerpressman对大家说: I'm no
long a professor, but I was one for a long time. I expected hard work and good
thinking from all my students.(2002/01/10 22:01)
【偶不再是教授，但偶曾干过很长时间。我期望从学生身上看到勤奋工作和好主意】
liyajia129进入聊天室.(2002/01/10 22:01)
docsa进入聊天室.(2002/01/10 22:02)
csliuyb进入聊天室.(2002/01/10 22:02)
fcx123进入聊天室.(2002/01/10 22:02)
fcx123对rogerpressman说: the
smallest one only have 12 developers. Do you think it is departure of CMM's
creator original intention?(2002/01/10 22:02)
fcx123对rogerpressman说: We know,
CMM theory are very important to large software enterprises! But as I know, in
the software enterprises which ware certified CMM level 2 or above,(2002/01/10
22:02)
jameshwoo对大家说: hard work
and good thinking, I agree.(2002/01/10 22:02)
derta对rogerpressman说: Hi,What
do u think about the software development?Are there some good or mixed patterns
for SE?(2002/01/10 22:02)
derta对rogerpressman说: Will it
go some extremes? One hand,big companies leading,such as IBM;another hand, open
source code prggramming,such as something for Linux.(2002/01/10 22:03)
caozq进入聊天室.(2002/01/10 22:03)
qlw5对rogerpressman说: Dear
rogerpressman, I know clearly that CMM is based upon practices, however SW-CMM
v1.1 is about 10 years old, although the spirit is still there, the practice has
been(2002/01/10 22:03)
qlw5对rogerpressman说:
changed(2002/01/10 22:03)
silentsnake进入聊天室.(2002/01/10 22:04)
vickyday对qlw5说: But the
general principles wont change.(2002/01/10 22:04)
qlw5对rogerpressman说: Dear
jameshwoo, what do you mean the pattern?(2002/01/10 22:04)
rogerpressman对大家说: Lots of
questions :-) I get the impression that the CMM is a very "hot" topic
in China. Remeber that the CMM is not a religion to be applied
dogmatically.(2002/01/10 22:05)
【这么多问题，我意识到CMM再中国很“热”，记住CMM不是盲目崇拜的宗教】
qlw5对rogerpressman说: vickyday,
no I do not think you know about CMM(2002/01/10 22:05)
jameshwoo对vickyday说: They Press
their main Course Material to internet. (www.MIT.edu)(2002/01/10 22:05)
rogerpressman对大家说: It is a
phiosophy that can be adapted to meet your needs.(2002/01/10 22:05)
【适用于你的需求才是哲学】
Charity_Zhou离开了聊天室.(2002/01/10 22:05)
qlw5进入聊天室.(2002/01/10 22:05)
greatfox进入聊天室.(2002/01/10 22:05)
rogerpressman进入聊天室.(2002/01/10 22:06)
rogerpressman对大家说: The fact
that the CMM is 10+ years old does not invalidate it as a worthwhile
philosophy.(2002/01/10 22:06)
【事实上CMM虽已经超过10岁，并不妨碍它成为哲学】
csliuyb进入聊天室.(2002/01/10 22:06)
csliuyb对大家说: Dr, pressman,
If u r the professor, what knowledge would u expect your student have? or rather
what kind of courses they should learn first? as a post graduate
student(2002/01/10 22:06)
qlw5对rogerpressman说: General
questions on CMM, you can refer to me, I have some experience on CMM(2002/01/10
22:06)
fsbat进入聊天室.(2002/01/10 22:06)
seeseax对rogerpressman说: I am
reading Dr.Humphrey's book on PSP, but seems few SE's around us do it, do you
have any advice on such situation?any advice on applying PSP and TSP?(2002/01/10
22:06)
xujnfeng进入聊天室.(2002/01/10 22:06)
vickyday进入聊天室.(2002/01/10 22:06)
vickyday对qlw5说: Yes I dont know
all of them. But as far as the KPAs we are applying, I think they are useful and
wont be outdated.(2002/01/10 22:06)
stoic进入聊天室.(2002/01/10 22:07)
rogerpressman对大家说: As I
mentions PSP and TSP require formal training and then the discipline to apply
the methods. It is also not being used widley in the USA. laziness, I
think.(2002/01/10 22:07)
【我已经说过PSP和TSP需要正式培训，并规范实施。在美国也并没有被广泛使用，我想】
vickyday对qlw5说: Because CMM
doesnt tell how to do, but set the aim U must achieve.(2002/01/10 22:07)
qlw5对rogerpressman说: Dear,
rogerpressman, but someone, maybe DeMacro or High SmithIII do not agree with
CMM, they think CMM is burden to progress, what;s your oppion?(2002/01/10 22:07)
xujnfeng离开了聊天室.(2002/01/10 22:07)
javaor对rogerpressman说: Hi,Dr
Pressman.What team or team leader is your expected.(2002/01/10 22:09)
rogerpressman对大家说: Also
DeMarco and Highsmith are both friends, I disagree. The CMM is a burden only if
it is applied thoughtlessly. If its spirit is applied with commn sense, it can
assist(2002/01/10 22:09)
rogerpressman对大家说: It can
assist an organization in doing good SE work.(2002/01/10 22:09)
【虽然DeMarco and Highsmith都是我的朋友，我也不同意他们的观点。CMM成为负担的原因是没有经过思考。如果其精髓被领会，它能够很好地帮助机构地SE工作】
yin_feng进入聊天室.(2002/01/10 22:09)
rogerpressman进入聊天室.(2002/01/10 22:10)
rogerpressman对大家说: There is
an excellent book on Team Leadership by Jerry Weinberg. Title On Becoming a
Technical Leader. I recommend it.(2002/01/10 22:10)
【Jerry Weinberg有一本关于团队领导的好书。Becoming a Technical
Leader，推荐你看看】
atlas_tao进入聊天室.(2002/01/10 22:11)
atlas_tao对rogerpressman说:
what's the position of MS COM in SE?(2002/01/10 22:11)
mhuang进入聊天室.(2002/01/10 22:12)
mhuang对大家说: where can we
get it?(2002/01/10 22:12)
jameshwoo对rogerpressman说: Dr.
Pressman, you are in a software design team nowadays?(2002/01/10 22:12)
rogerpressman对大家说: MS COM is
one of a number of component-based technologies for SE. All attempts to achieve
reuse will benefit an organization, as long as good architectural design is
done(2002/01/10 22:12)
【MS COM是一种部件技术。如果有好的体系结构，所有实现复用的技术都是有意义的。】
greatfox对rogerpressman说: how about you think of msf?(2002/01/10 22:13)
qlw5对rogerpressman说: Dear
rogerpressman, I think things are changing, for example, Criticle Computer
resource are NEVER a KP in CMMI 1.02, while is a KP in SW CMM v 1.1 L2 wht's
your opinion(2002/01/10 22:13)
rogerpressman对大家说: Thing are
changing. They always do. It's reasonable to assume that as the CMMI evolved,
there will be changes from the oder CMM. That does not invalidate older ideas,
it(2002/01/10 22:15)
【事情在变化，一直如此。假定随着CMMI的发展，这些东西会和CMM不同，但是其精神是一致的】
seeseax对rogerpressman说: what
kind of method is suitable for a small~middle project, no business critical,
mainly for research and interests ?(2002/01/10 22:15)
rogerpressman进入聊天室.(2002/01/10 22:15)
rogerpressman对大家说: It simply
means that we learned from past experience and have modified our
approach.(2002/01/10 22:15)
【它的意思是我们从经验中学习并修正我们的方案】
atlas_tao对大家说: does the architecture design + component-based
technologies can achieve success?(2002/01/10 22:15)
brook0344对大家说: sa China
has been a country in WTO,What position will American software be on in
China,tommorrow?(2002/01/10 22:15)
vickyday进入聊天室.(2002/01/10 22:16)
vickyday对大家说: What should
we learn from the Model? The methodology or all the details?(2002/01/10 22:16)
qlw5对rogerpressman说: Thanks,
rogerpressman, Can you tell me why CMMI is still not a offcial
version?(2002/01/10 22:16)
zik进入聊天室.(2002/01/10 22:16)
rogerpressman对大家说: Altho I
have a high bandwidth connection, my repsonse time is degrading. Not sure why.
Please understand delayed responses.(2002/01/10 22:16)
csliuyb进入聊天室.(2002/01/10 22:18)
csliuyb对rogerpressman说: what
kind of method is suitable for a small~middle project, no business critical,
mainly for research and interests ?((2002/01/10 22:18)
john long进入聊天室.(2002/01/10 22:18)
qlw5对大家说: Dear rogerpressman, do not feel so aggressive :-)(2002/01/10
22:18)
netd对rogerpressman说: I think SE
just can REDUCE rather than SOLVE the problems in software development, how do
you think about that?(2002/01/10 22:18)
vickyday对大家说: The system is abnormal I think.(2002/01/10 22:18)
csliuyb对大家说: :((2002/01/10
22:19)
qlw5对大家说: To my opinion,
For small Org. the XP is much better than CMM(2002/01/10 22:19)
rogerpressman对大家说: I agree.
SE is not a silver bullet. It reduces the problems of software development, it
cannot completely eliminate them.(2002/01/10 22:19)
【SE不是银弹，它降低问题而非消除问题】
lf168进入聊天室.(2002/01/10 22:19)
john long微笑着对大家说: Dear rogerpressman, nice to meet you(2002/01/10
22:19)
fcx123进入聊天室.(2002/01/10 22:19)
fcx123对rogerpressman说: there
many good open source about framework of CORBA,ACE+TAO is one of the best
frameworks. Can you give us some others ? And how I choose from them?(2002/01/10
22:19)
fsbat离开了聊天室.(2002/01/10 22:20)
huhj_uml离开了聊天室.(2002/01/10 22:20)
rogerpressman进入聊天室.(2002/01/10 22:20)
rogerpressman对大家说: Why not
combine the best proactices of XP with CMM ideas. Therefore you get good
responsiveness and creativity and good quality and discipline.(2002/01/10 22:20)
【为什么不结合XP和CMM的优点。这样你会具有好的责任心、创造性、高质量和规则】
qlw5对大家说: Dear netd, SE can
only solve the problem when there are more support than Finite state
machine(2002/01/10 22:21)
goaha离开了聊天室.(2002/01/10 22:21)
netd对rogerpressman说: So is
there any ways to completely eliminate the problems?(2002/01/10 22:21)
qlw5对大家说: XP + CMM +
adaptive software development is probably a kind of wonderful
approach:)(2002/01/10 22:22)
javaor对rogerpressman说: Software
is powerful today!What is the future of SE?Maybe the end user can customer
software NOT the engineer.:-)(2002/01/10 22:22)
john long微笑着对大家说: what's the most important thing in Object
Oriented Software(2002/01/10 22:23)
rogerpressman对大家说: Sadly,
buiding complex systems is difficult. I do not believe we will ever completely
eliminate the problems. If we did, none of us would have jobs :-))(2002/01/10
22:23)
【很可惜，构造复杂系统是困难的。我不认为我们已经解决了全部问题，如果确实如此，在座各位都失业了:-】】
coolie2k进入聊天室.(2002/01/10 22:23)
forestinnorway离开了聊天室.(2002/01/10 22:23)
rogerpressman对大家说: End user
computing is good for some things, but SEs will be the ones to build complex
systems.(2002/01/10 22:23)
qlw5对大家说: Dear fcx123,
ACE+TAO is a good solution, but have you really understand it?(2002/01/10 22:23)
jameshwoo对rogerpressman说:
wonderfull.(2002/01/10 22:23)
skyoldx进入聊天室.(2002/01/10 22:23)
skyoldx对大家说: what kind of
method is suitable for a small~middle project, no business critical, mainly for
research and interests(2002/01/10 22:23)
fcx123进入聊天室.(2002/01/10 22:24)
fcx123对qlw5说: Not
Yet!(2002/01/10 22:24)
atlas_tao进入聊天室.(2002/01/10 22:24)
atlas_tao对大家说: which
approach will be selected to solve low-level project?(2002/01/10 22:24)
rogerpressman对大家说: Shyoldx:
It sounds like an evolutionary process model would be best. The methods depend
on the application type and the tools available to you.(2002/01/10 22:24)
【听上去演化模型是最好的。但是其依赖于应用类型和可得工具】
qlw5对大家说: Dear fcx123, I
just read some part of the code, I found it is one the open src prj with godo
doc.(2002/01/10 22:25)
fcx123进入聊天室.(2002/01/10 22:25)
qlw5进入聊天室.(2002/01/10 22:25)
brook0344对大家说: HI，问一下，我们这次讨论有一个主题的吗(2002/01/10
22:25)
csliuyb进入聊天室.(2002/01/10 22:25)
rogerpressman进入聊天室.(2002/01/10 22:25)
rogerpressman对大家说: The most
important aspect of OO is to understand the concept of the class and to build
meaningful class hierarchies for the problem to be soleved.(2002/01/10 22:25)
【Oo的重要方面是理解类解决问题所需的有意义的类层次关系。】
csliuyb对rogerpressman说: What is
the future development of SE?(2002/01/10 22:25)
vickyday对rogerpressman说: How do u expect for the industrialization of the
Software industry? And what's ur oppion about the role of the SE
professionals?(2002/01/10 22:25)
vickyday对rogerpressman说: How do
u expect for the industrialization of the Software industry? And what's ur
oppion about the role of the SE professionals?(2002/01/10 22:26)
qlw5对大家说: Dear
rogerpressman, how do you think the role-based process man. approach rather than
activity-based?(2002/01/10 22:26)
rogerpressman对大家说: Many
years ago, people talked about "software factories. SE is a human endeavor.
It's unliley that it will be "industrialized" any time
soon.(2002/01/10 22:26)
【多年前，人们讨论“软件工厂，SE是个人努力”看上去短期很难出现这种情况】
fcx123对qlw5说: I hope to use it
in my network management system.(2002/01/10 22:26)
dearlan进入聊天室.(2002/01/10 22:27)
rogerpressman对大家说: qlw5: I
don't liker to make those distinctions. Activities must be assigned and roles
must be played for effective management. I like both :-)(2002/01/10 22:27)
【qlw5，我不想区分它们，活动必需分配、角色完成有效管理。偶两个都喜欢】
fcx123对qlw5说: Can you give me
some advices about that?(2002/01/10 22:27)
qlw5对大家说: It's major a
Corba sys, which can across OS platform(2002/01/10 22:27)
qlw5对大家说: I currently
rading a book by Highsmith, it say that WBS can never used in new complex
project, why?(2002/01/10 22:28)
luv离开了聊天室.(2002/01/10 22:29)
yin_feng对qlw5说: what is
wbs?(2002/01/10 22:29)
vickyday对qlw5说: Maybe for the inovince of the system, U can hardly develop
the WBS properly(2002/01/10 22:29)
john long进入聊天室.(2002/01/10 22:30)
john long微笑着对大家说: sorry ,Dear rogerpressman,I means OOST,In fact,
In my company,we are in the primary moment of Object Oriented Software Test,And
Do you have some wonderful advice for me ?(2002/01/10 22:30)
qlw5对大家说: Dear
rogerpressman, I think CMM v 1.1 will be replaced be by CMMi 1.02 technically,
but DoD do not want CMMi, right?(2002/01/10 22:30)
vickyday对大家说: yin_feng:Work
Breakdown Structure(2002/01/10 22:30)
qlw5对大家说: Work breakdown
structure(2002/01/10 22:30)
fcx123进入聊天室.(2002/01/10 22:31)
fcx123对qlw5说: I will contact
with you later,Ok?(2002/01/10 22:31)
qlw5对大家说: Dear vickyday,
WBS is derived by experienced team(2002/01/10 22:31)
rogerpressman进入聊天室.(2002/01/10 22:31)
rogerpressman对大家说: Sadly,
OOST is beyond the scope of this chat. There are a number of good books that
have come out recently, Check amazon.com or other sources.(2002/01/10 22:31)
【OOST超出讨论范围，关于其话题，有很多好书，到amazon.com或其他地方看看】
vickyday对大家说: Dear qlw, u
just now mentioned the project is new and complex.(2002/01/10 22:32)
rogerpressman对大家说: I dont
agree with highsmith on that, His contention is that all projects must be
adaptible and that a WBS is not adaptible.(2002/01/10 22:32)
【我不同意highsmith的这个观点。他认为所有项目都必需自适应，因此WBS就没有用了】
kenjiang进入聊天室.(2002/01/10 22:33)
brook0344对大家说: HI，请问这次讨论的主题是什么？(2002/01/10
22:33)
qlw5对大家说: Dear
rogerpressman, Thanks(2002/01/10 22:33)
seeseax对rogerpressman说: How do
you think about UXF (UML eXchange Format ), will it benifit interoperability
much as expected?(2002/01/10 22:33)
rogerpressman对大家说: I
disagree. You can create a WBS and then adapt it as the project progresses.
Without it, you have no reasonable raodmap.(2002/01/10 22:33)
【我不同意，你可以先设计WSB，随着项目进行而修订它。如果没有这个东西，你就没有方向】
qlw5对大家说: brook0344, on
software process(2002/01/10 22:33)
brook0344对大家说:
哦，知道了(2002/01/10 22:34)
csliuyb进入聊天室.(2002/01/10 22:35)
csliuyb对大家说: 了解了(2002/01/10
22:35)
rogerpressman对大家说: UXF will
help. but first, we ned to have more people and organizations using UML on a
regular basis.(2002/01/10 22:35)
【UXF有帮助，但首先我们需要更多的人和机构经常性地使用UML】
qlw5对大家说: Dear
rogerpressman, However if the requirement are changing much, a static WBS will
never be applicable(2002/01/10 22:35)
john long进入聊天室.(2002/01/10 22:35)
john long微笑着对大家说: oh, My God, Dear rogerpressman, Do you think
that I can buy a book of OOST from amazon.com? In fact I can't for I have no
dollar,and any e-books?(2002/01/10 22:35)
csliuyb进入聊天室.(2002/01/10 22:35)
qlw5进入聊天室.(2002/01/10 22:35)
csliuyb对大家说: :-)(2002/01/10
22:35)
javaor进入聊天室.(2002/01/10 22:36)
rogerpressman进入聊天室.(2002/01/10 22:36)
rogerpressman对大家说: qlw5: I
agree. But there is no need for a Static WBS. It can adapt as requirements and
the project changes.(2002/01/10 22:36)
【qlw5,偶同意你地观点，静态WBS没有用处，随着需求和项目地发展，WBS应该是自适应的】
oysterlqd进入聊天室.(2002/01/10 22:36)
oysterlqd对大家说: yap,e-books,hoho(2002/01/10 22:36)
luxf001进入聊天室.(2002/01/10 22:36)
qlw5对大家说: Dear
rogerpressman, do u think CMM can be used in small projects less than 6 people,
and less than 6 months?(2002/01/10 22:36)
rogerpressman对大家说: John
long: Sorry. I suspect there is much free info on the internet. Visit
www.rspa.com/spi and look under OO testing.(2002/01/10 22:37)
netd对rogerpressman说: Some bad
habits of the staff can have negative effect on software process improvement,
how can we avoid that or change that?(2002/01/10 22:37)
qlw5对大家说: www.rspa.com is
good place for you to find good templates(2002/01/10 22:37)
zik进入聊天室.(2002/01/10 22:38)
rogerpressman对大家说: qlw5:
Yes, but only if you recognize that it must be applied in a way that will NOT
slow things down. Therefore, it must be adpted to the smaller
project.(2002/01/10 22:38)
【qlw5:可以，但是必需在一种不影响进度的前提下使用。必需自适应】
umlchina进入聊天室.(2002/01/10 22:40)
anon离开了聊天室.(2002/01/10 22:40)
qlw5对大家说: I my experience,
small project can SPTO a project less than a large project, something like size
can be tracked longer than a month, sometimes the size is tracked
only(2002/01/10 22:40)
umlchina对大家说:
test(2002/01/10 22:40)
john long微笑着对大家说: thanks, Dear rogerpressman, And I am in trobule
that when I use Rose, I find it hardly works well,maybe I can't catch uml's
soul(2002/01/10 22:41)
qlw5对大家说: once in a small
project, do you think it's probably? or do you track size in your
project?(2002/01/10 22:41)
rogerpressman进入聊天室.(2002/01/10 22:41)
rogerpressman对大家说: Sorry,
SPTO?(2002/01/10 22:41)
qlw5对大家说: Software Project
tracking and oversight ( a CMM term)(2002/01/10 22:42)
javaor对rogerpressman说: Dr
Pressman.Any advise for a practitioner?Hard work,think deeply?:-)(2002/01/10
22:42)
qlw5对大家说: Dear
rogerpressman, Which is the direction of SE, Management of
Techniques?(2002/01/10 22:43)
vickyday对大家说: If only it
helps, u can choose the frequency and the method I think.(2002/01/10 22:43)
netd对rogerpressman说: Some bad
habits of the staff can have negative effect on software process improvement,
how can we avoid that or change that?(2002/01/10 22:43)
rogerpressman对大家说: qlw5: In
some ways, its more importahnt to SPTO a smaller project. Reason, small
variances in timeine schedule can throw the project off scehdule in
unrecoverable ways.(2002/01/10 22:43)
rogerpressman对大家说:
Therefore, it's necessary to know when things are slipping asap.(2002/01/10
22:43)
【qlw5：在某些场合，SPTO对小型项目更重要。理由：小的变动就可能让项目无法接受地改期，因此应该及时了解出现地问题】
wu_hao进入聊天室.(2002/01/10 22:44)
vickyday对rogerpressman说: How do
U think of ISO9000 2000version?(2002/01/10 22:44)
rogerpressman进入聊天室.(2002/01/10 22:46)
rogerpressman对大家说: vicyday:
each of the process/quality models have merits and problems. The ISO 9000
approach is documentation heavy, but does requires continual auiting (a good
idea).(2002/01/10 22:46)
vickyday进入聊天室.(2002/01/10 22:46)
vickyday对rogerpressman说: Do u
run into this kind of occasion: the schedule the customs ask for is far beyond
the actual baseline u can afford? and how do u manage it?(2002/01/10 22:46)
goldarcher离开了聊天室.(2002/01/10 22:46)
rogerpressman对大家说: We don't
see it applied too often in the software realm in the USA.(2002/01/10 22:46)
seeseax对rogerpressman说: Dr
Pressman, I guess PSP and TSP would be help to SPTO,right? ?(2002/01/10 22:47)
qlw5对大家说: Dear rogerpressman, In US, is there more small projects or
large projects?(2002/01/10 22:47)
rogerpressman对大家说:
javour" the practitioner should work to understand more about SE. The
average practitioner is not familar with good methods for analysis, design and
testing.(2002/01/10 22:48)
qlw5对大家说: Dear seeseax, PSP and TSP will do good to SPTO, but it is the
responsibiulity of project manager(2002/01/10 22:48)
rogerpressman对大家说: The most
important(2002/01/10 22:49)
derta进入聊天室.(2002/01/10 22:50)
vickyday对大家说: sometimes I got puzzled by the techiniques learned from
the books about the SE. I think if it's too delicate, it will be hard to
populize for the engineers wont have too(2002/01/10 22:50)
rogerpressman进入聊天室.(2002/01/10 22:50)
rogerpressman对大家说: PSP and
TSP use a different approach to SPTO. It is generally the individuals
responsibility to measure and track progress. However, these metrics can be
reported for cen(2002/01/10 22:50)
【PSP和TSP使用不同的SPTO方案。通常是个人的责任来测量和跟踪问题。但这些数据必需被…】
3328749进入聊天室.(2002/01/10 22:50)
derta进入聊天室.(2002/01/10 22:50)
vickyday对大家说: much time to get familiar with it(2002/01/10 22:50)
vickyday对大家说: sometimes I
got puzzled by the techiniques learned from the books about the SE. I think if
it's too delicate, it will be hard to populize for the engineers wont have
too(2002/01/10 22:51)
vickyday对大家说: much time to
get familiar with it(2002/01/10 22:51)
emoto进入聊天室.(2002/01/10 22:51)
qlw5对rogerpressman说: Will formal method be the future of the software
engineering?(2002/01/10 22:51)
qlw5对rogerpressman说: Will
formal method be the future of the software engineering?(22:51)(2002/01/10
22:51)
rogerpressman对大家说: Vickyday:
I agree. SEs have little time to learn new things and less time to apply them
rigorously. Yet, to be successful, we should apply effective methods.(2002/01/10
22:52)
【我同意，SE很少有时间来学习新东西，并严格地使用。然而，为了成功，我们必需采取更有效地方法】
vickyday对rogerpressman说: How do
think about this?(2002/01/10 22:52)
seeseax对rogerpressman说: I agree
with you Dr. Rogerpressman(2002/01/10 22:52)
rogerpressman对大家说: Maybe the
best approach is to learn and few good methods and then apply them
well.(2002/01/10 22:52)
vickyday对rogerpressman说: So
there must be someone to learn and choose for the SEs?(2002/01/10 22:53)
mhuang对大家说: what is the most important thing to apply PSP/TSP in a
application development?(2002/01/10 22:53)
rogerpressman对大家说: qlw5: No,
FM will not be the future. Altho FM are very powerful, they are simply too
difficult for most SEs to apply effectively.(2002/01/10 22:53)
【qlw5: FM不会占据未来。虽然FM非常强大，他们对于多数SE来说还是太难了，难以有效实施】
rogerpressman对大家说: However,
they may be used in particularly critical apps.(2002/01/10 22:53)
mhuang对大家说: what is the
most important thing to apply PSP/TSP in a application development?(2002/01/10
22:54)
rogerpressman对大家说: vickyday:
SE should learn and choose for themselves.(2002/01/10 22:54)
qlw5对rogerpressman说: Dear
rogerpressman, may I ask you that if you are asked to design a MIS system which
method will you use, UML?(2002/01/10 22:54)
xchyj离开了聊天室.(2002/01/10 22:55)
seeseax对mhuang说: time control ,
I guess(2002/01/10 22:55)
3328749进入聊天室.(2002/01/10 22:55)
rogerpressman对大家说: qlw5:
Today, UML has become the analysis/design approach of choice. However, it would
depend on the characteristics of the MIS system(2002/01/10 22:55)
【qlw5: 今天UML已经是分析和设计地选择。但是它依赖于MIS系统地特征】
rogerpressman进入聊天室.(2002/01/10 22:55)
rogerpressman对大家说: Remember,
the method is important, but a discipline approach to analysis and design is
more important.(2002/01/10 22:55)
【记住，方法是重要地，但规范的分析设计方案更为重要】
qlw5对rogerpressman说:
Thanks(2002/01/10 22:55)
javaor对rogerpressman说: Is
system threoa important to SE and practitioner?(2002/01/10 22:56)
mhuang对大家说: to seeseax: why
not the estimation of time and size? it is most difficult(2002/01/10 22:56)
rogerpressman对大家说: javour:
threoa?(2002/01/10 22:57)
qlw5对rogerpressman说: Dear Rogerpressman, Why do you think FM can not used
widely, the tools, IDE or something else?(2002/01/10 22:57)
qlw5对rogerpressman说: Dear
Rogerpressman, Why do you think FM can not used widely, the tools, IDE or
something else?(22:57)(2002/01/10 22:58)
jameshwoo对rogerpressman说: Dr.
Rogerpressman, how do you mean the characteristics of the MIS system?(2002/01/10
22:58)
rogerpressman对大家说: qlw5:
primary reason is that most SEs do not have sufficient background and are
unwilling/unable to learn. Please understand that FM are VERY
powerful,(2002/01/10 22:58)
rogerpressman对大家说: but I
honestly do not thing they will ever dominate our SE work.(2002/01/10 22:58)
【qlw5: 主要原因是多数SE缺乏相关背景，也不愿意学习。请记住FM非常强大。但我个人认为它不会统治SE工作】
yin_feng对qlw5说: what is FM?
thanks(2002/01/10 22:59)
javaor对rogerpressman说: Is
knowage of system important to SE and practitioner?((2002/01/10 22:59)
vickyday对rogerpressman说: Aren't
the priciples built and asked by the organization? It's said No changes will be
taken automaticly.(2002/01/10 22:59)
qlw5对rogerpressman说: So , if
the Tools are powerful enough, do you agree that we can use FM other
tranditional Progamming Language?(2002/01/10 23:00)
seeseax对mhuang说: yes, including
time estimation,plan and track.... you can read <Introduction to the personal
software process>, Chinese version has come out. writen by Watts S.
Humphrey(2002/01/10 23:00)
qlw5对yin_feng说: Formal
method(2002/01/10 23:00)
rogerpressman对大家说: FM =
formal methods(2002/01/10 23:00)
shenqw进入聊天室.(2002/01/10 23:01)
rogerpressman进入聊天室.(2002/01/10 23:01)
rogerpressman对大家说: javoar:
Yes! understanding the system put the software into context, defines appropeiate
external interfacts, and often dictate architecture.(2002/01/10 23:01)
vickyday对rogerpressman说: Aren't
the priciples built and asked by the organization? It's said No changes will be
taken automaticly(2002/01/10 23:01)
mhuang对大家说: to seeseax:
very thanks(2002/01/10 23:01)
rogerpressman对大家说: Vicyday:
that does not mean that an individual cannot learn new things and apply them
within the principles set down by the organization.(2002/01/10 23:02)
【意思不是说个人不能学习新东西，并在机构规则范围内使用它们】
3328749对rogerpressman说: Mr rogerpressman,there are so many concepts and
methods about SE. To learn SE, what should I learn first?(2002/01/10 23:03)
qlw5对rogerpressman说: Dear
Rogerpressman, How do you think about Code constructuon and Detail design? are
they the same for some project?(2002/01/10 23:03)
seeseax对mhuang说: you are
welcome, it's Chinese name: 个体软件过程，人民邮电出版社(2002/01/10
23:03)
qlw5对rogerpressman说: Dear
mhuang, I prefer your reading the English version(2002/01/10 23:04)
rogerpressman进入聊天室.(2002/01/10 23:06)
rogerpressman对大家说: qlw5: In
the modern context, coding and DD are often one in the same. An older view is to
consder coding as a mechansitic process, but today, code and DD are often
combin(2002/01/10 23:06)
【qlw5：从现代的观点看，编码和DD常常是一件事情。老式观点认为编码是机械的，但今天编码和DD通常被合并】
mum进入聊天室.(2002/01/10 23:06)
vickyday对大家说: Why english version? Dont U think it's quick to get
understood?(2002/01/10 23:06)
seeseax对qlw5说: Is there English
version of that book in China now?(2002/01/10 23:07)
qlw5对大家说: Dear Drogerpressman, Thanks, do you mean the modern design
based on RAD?(2002/01/10 23:07)
qlw5对大家说: Dear
Drogerpressman, Thanks, do you mean the modern design based on
RAD?(23:07)(2002/01/10 23:07)
netd对rogerpressman说: how do you
think about the communication in software process? Is talking more important
than document?(2002/01/10 23:07)
qlw5对大家说: seeseax, I think
www.sei.cmu.edu will provide you a good tutorial(2002/01/10 23:08)
netd对rogerpressman说: Which is
more important, talking directly or document?(2002/01/10 23:08)
seeseax微笑着对qlw5说: thanks
so much(2002/01/10 23:08)
brook0344对大家说: or a
formal meeting for communication?(2002/01/10 23:08)
rogerpressman对大家说: netd:
Both are important. Today, we demphasize paper documents and try to build
documentation into the workproducts =that are created. Talking (via effective
meeting and(2002/01/10 23:09)
rogerpressman对大家说: reviews
is very important. SE involves people and people must communicate to be
effective.(2002/01/10 23:09)
【评审是非常重要的，SE要求人与人间有效勾通】
john long对大家说: Dear Drogerpressman, though it's out of scope of this
chat, but i still very long for asking a question followed like this:How Can We
Use OO Method In A System Test?(2002/01/10 23:10)
rogerpressman进入聊天室.(2002/01/10 23:11)
rogerpressman对大家说: qlw5: No.
what I mean is the modern view of SE. The design approach can encompass any
recognized method.(2002/01/10 23:11)
【不是，我的意思是从现代的观点看待SE。设计可以通过多种途径】
seeseax进入聊天室.(2002/01/10 23:11)
go4all2000进入聊天室.(2002/01/10 23:11)
dzxing进入聊天室.(2002/01/10 23:12)
vickyday对rogerpressman说: Could u recommand us some material about FP
estimate?(2002/01/10 23:12)
jameshwoo进入聊天室.(2002/01/10 23:13)
rogerpressman对大家说: john
long: There are technical called intercall test case design that are part of
system testing. I can't describe here but they make use of usecases, behavioral
models a(2002/01/10 23:13)
rogerpressman对大家说: and other
mechanisms.(2002/01/10 23:13)
qlw5对大家说: Thanks
Rogerpressman, How do you think about aspect-orientated Programming(AOP), do you
think it's important to some extend?(2002/01/10 23:13)
siralos进入聊天室.(2002/01/10 23:14)
3328749进入聊天室.(2002/01/10 23:14)
rogerpressman对大家说: vickyday.
There is a wide array of info a FP avaliable on the internet. See
www.rspa.com/spi and llok under process metrics and project management as a
starting point.(2002/01/10 23:14)
mum对大家说: Dear rogerpressman, How do you think about the difference
between small project and big project? let's say 3-5 developers on small one a,
and 50 developer in a larger pro(2002/01/10 23:14)
slob进入聊天室.(2002/01/10 23:15)
rogerpressman进入聊天室.(2002/01/10 23:15)
rogerpressman对大家说: mum:
There are substantial differences -- in communication, in planning, in
coordination, in measurement, in team structure, in methodolgy, in SQA, in
comtrol ....(2002/01/10 23:15)
【有微妙的不同－－交流、策划、协调、测量、团队结构、方法学、SQA和控制】
slob进入聊天室.(2002/01/10 23:15)
qlw5对大家说: slob, how are
you?(2002/01/10 23:16)
slob对大家说: hi, i am just back from lab, hehe(2002/01/10 23:16)
rogerpressman对大家说: qlw5: I'm
only beginning my study of AOP, so I'll reserve comment until another time
:-)(2002/01/10 23:17)
【qlw5:我刚刚开始学AOP，所以偶保持沉默，直到以后有机会】
vickyday对rogerpressman说: I once
followed ur book(SE ,4th) to try a FP estimate for one module, but find hard to
decide for the value of actors.(2002/01/10 23:17)
vickyday对rogerpressman说:
factors.(2002/01/10 23:17)
slob进入聊天室.(2002/01/10 23:18)
vickyday对rogerpressman说: And
the assessor of CMM dont agree that we have performed FP estimate for Size
estimate though we finally got a result.(2002/01/10 23:18)
rogerpressman对大家说: vicky:
That is a problem. FP demands some subjective evaluation of the factors. There
are more complete and detailed guides for doing this than my book. See
www.fpug.org(2002/01/10 23:18)
【这是个问题，FP需要主观评估各种因素。我的书中有更详细描述，参见www.fpug.org】
qlw5对大家说: Dear
rogerpressman, If I say SW CMM v1.1 will die in the coming 2~3 years, do you
agree?(2002/01/10 23:18)
dzxing进入聊天室.(2002/01/10 23:18)
seeseax离开了聊天室.(2002/01/10 23:19)
qlw5对大家说: Dear vickyday, FP
is rather hard for everyone to understand(2002/01/10 23:19)
netd对rogerpressman说:
rogerpressman, what is the best way to accumulate the knowledge of
SE?(2002/01/10 23:19)
qlw5对大家说: Slob, how about
Tang yi, are you fine?(2002/01/10 23:20)
dfhprjz进入聊天室.(2002/01/10 23:20)
jameshwoo离开了聊天室.(2002/01/10 23:20)
vickyday对qlw5说: Could u give me
advice where to start from?(2002/01/10 23:20)
xuefg进入聊天室.(2002/01/10 23:20)
umlchina进入聊天室.(2002/01/10 23:20)
rogerpressman进入聊天室.(2002/01/10 23:20)
rogerpressman对大家说: qlw5: In
SE things never die, they simply slowly fade away :-) It is likely that CMM v1.1
will fade as CMMI gains prominence.(2002/01/10 23:20)
【在SE，啥也不会消亡，它们只是淡出而已:-).我想CMM v1.1会随着CMMI的出现而淡出】
qlw5对大家说: Dear vickyday,
the FP method is not free one, you will have to search internet to find some
free material and course, good luck(2002/01/10 23:21)
netd对rogerpressman说: How do you
think the importance of the industrial background when you develop an enterprise
application software?(2002/01/10 23:21)
vickyday对rogerpressman说: I
tried for the fpug , but the web site seems not active.(2002/01/10 23:22)
xuefg对rogerpressman说: how
are thinking for C#?(2002/01/10 23:22)
rogerpressman对大家说: netd:
good books. also try www.swebok.org which cover the SE body of knowledge. Also
try www.rspa.com/spi for resources.(2002/01/10 23:22)
qlw5对大家说: Dear
Rogerpressman, I agree with u, in my opinion SW CMM v 1.1 will be faded
soon:--)(2002/01/10 23:22)
3328749对rogerpressman说: Mr
rogerpressman. What are you working on ? In US software corporations, what SE
methods are used most popularly?(2002/01/10 23:22)
rogerpressman对大家说: vickyday:
Do a search on the FP user group. They should have asite.(2002/01/10 23:23)
qlw5对大家说: vickyday, it is
not free site(2002/01/10 23:23)
rogerpressman对大家说: 3328749:
UNL is the most active methodology area at the moment. in addition
component-based methods and web-engineering are active. I'm working with clients
on process(2002/01/10 23:24)
vickyday对rogerpressman说: Thanks
a lot. Maybe my company will arrange a training for FP for the size estimate is
the main problem in our CMM assessment.:)(2002/01/10 23:24)
rogerpressman对大家说: netd:
very important. Business knowedge is critical. It helps in requirements
gathering and in communication.(2002/01/10 23:24
【非常重要，业务只是是关键性的。它帮助收集需求和交流】
siltboy进入聊天室.(2002/01/10 23:25)
vickyday对qlw5说: Thanks a lot.
and I think we can communicate more if U r free.(2002/01/10 23:25)
qlw5对大家说: Rogerpressman,
What is the relationship between component based Analysis and Design and OOAD
?(2002/01/10 23:25)
qlw5对大家说: vickyday, you
will be charged a lot(2002/01/10 23:26)
vickyday对qlw5说: qlw, U r
scaring me.(2002/01/10 23:27)
rogerpressman进入聊天室.(2002/01/10 23:27)
rogerpressman对大家说: qlw5:
OOAD assumes (at least to some extent that the SE will be creating the
appropriate classes or accessing them from an exsting in house library. CBSE
assume access to(2002/01/10 23:27)
【OOAD假定（至少在某种程度上）SE会创建合适的类或从现存本地库中进行访问，CBSE假定访问商业化部件。但从核心上看，区别是微妙的】
javaor对rogerpressman说: Dr
Pressman.In your view,what is system?(2002/01/10 23:27)
rogerpressman对大家说:
commerically available components. However, at the core there are only subtle
differences.(2002/01/10 23:28)
qlw5对大家说: vickyday, Do not worry! but it's really(2002/01/10 23:29)
rogerpressman对大家说: javaor :
a collection of interrelated elements that work together to meet the needs of
some customer. The customer can be a person, a group, a machine or another
system.(2002/01/10 23:29)
rogerpressman对大家说: Not a
formal defintition :-)(2002/01/10 23:29)
vickyday对rogerpressman说: Thanks
a lot again. For it's ur book of SE(esp. the intro of urself) that light my fire
of interest for SE and now I become a career SE professional. I would like to
go(2002/01/10 23:29)
rogerpressman进入聊天室.(2002/01/10 23:30)
rogerpressman对大家说: All: I do
have another meeting in 15 minutes, so it may be a good idea to end our chat. I
enjoyed it and apologize for my typos. Best of luck to all of you! And
thanks.(2002/01/10 23:30)
【我15分钟后有个会，因此只好说再见了。我非常愿意和各位交流，另外请原谅我的拙劣指法。祝各位好运，并深表谢意】
qlw5对大家说: Thanks Rogerpressman, you give us a lot of new
ideas(2002/01/10 23:31)
umlchina对rogerpressman说: best
wishes to you, dr. pressman(2002/01/10 23:31)
netd对rogerpressman说: Thank
you,Dr rogerpressman.You helped us much!(2002/01/10 23:31)
yin_feng对大家说: Dr.Thank you
very much!(2002/01/10 23:31)
mum进入聊天室.(2002/01/10 23:31)
mum对大家说: thank
Rogerpressman and have a great day!(2002/01/10 23:31)
oysterlqd进入聊天室.(2002/01/10 23:31)
oysterlqd对大家说: thx(2002/01/10 23:31)
brook0344对大家说: thank
you(2002/01/10 23:31)
umlchina对rogerpressman说: the
exit button is at bottom-right(2002/01/10 23:31)
mhuang对大家说: thank you so
much(2002/01/10 23:32)
yin_feng对rogerpressman说: nice
to meet you(2002/01/10 23:32)
mhuang离开了聊天室.(2002/01/10 23:32)
yin_feng对rogerpressman说: 再见(2002/01/10
23:32)
qlw5对大家说: Sorry, umlchina, I come late, because someone want to make me
drunk(2002/01/10 23:32)
javaor对rogerpressman说:
Thanks,Dr Pressman.Nice to meet you.:)(2002/01/10 23:32)
umlchina对qlw5说: never
mind(2002/01/10 23:33)
yin_feng离开了聊天室.(2002/01/10 23:33)
vickyday对rogerpressman说: Thank you and wish to see you again.(2002/01/10
23:33)
