FEAST Workshop II
24 - 25 October 1994
Department of Computing
Imperial College of Science, Technology and Medicine
180 Queen's Gate
London SW7 2BZ
+44 (0) 171 594 8213/4
fax +44 (0) 171 594 8215
Organisational Matters
General Information (2)
Program (3)
Attendees (4)
Briefing Papers
Some Characteristics of S-, E- and P-type Programs (7)
Manny Lehman, ICSTM
Charts for the above (11)
Evolution and Learning in the Software Process (27)
Aidan Ward, Coherent Technology Ltd.
FEAST Papers
F1 - Five Position Papers32 F2 - Various Comments41 F3 - Three-way Discussion62 Jose Fiadeiro, Vic Stenning, Manny Lehman
F4 - A Three way email Discussion66 Wlad Turski, Dewayne Perry, Manny Lehman
F5 - Manifesto and Minutes78 Dewayne Perry, Wlad Turski, Manny Lehman
F6 - Comments on Manifesto82 Bob Snowdon
?The workshop will be held on the 24/25 October in the Department of Computing, Imperial College, 180 Queen's Gate, London, SW7 2PH
?Registration fee of ￡55, cash or UK cheque or traveller's cheque, payable to Imperial College.
?The workshop sessions will begin both mornings at 9 45 in room 418 of 180 with coffee availble from 9 15 in room 433.
?Hotel accomodation for those requesting it has been reserved at the Embassy House Hotel, 32 Queens Gate., SW7. Room charge ￡65 per night single.
?The nearest underground station to the hotel are Gloucester Rd. and South Kensington both on the District, Circle and Picadilly Lines. The College is equidistant from them both.
?Sessions will be in room 418 on the 4th floor.
?Tea, coffee breaks and lunches in room 433 on the same floor.
?Dinner on Monday evening will be at 6 30 for 7 00 at Physics Common Room, Level 8, Blackett Lab in Huxley Building.
?Barbara Claxton, room 554, front corridor, 5th floor, will be available to assist attendees as required.
?In her absence assistance can be obtained from the Department general office, room 437.
?Incoming telephone calls direct to +44 (0) 171 594 8213 (Barbara)
?Departmental fax number is +44 (0) 171 594 8301
?email messages marked for your attention to bpw@https://www.doczj.com/doc/1a4197927.html,.
?Outgoing email can be sent from Barbara's machine or from mml's in room 556.
Monday, 24 October 1994
09 15 - 09 45Registration and Coffee433
Morning Chair Nazim Madhavji
09 45 - 10 00Opening remarks Manny Lehman418
10 00 - 11 15Characteristics of S-type and E- type programs Manny Lehman418
11 15 - 11 45Coffee433
11 45 - 13 00The FEAST Manifesto and a Preliminary Model Dewayne Perry418
13 00 - 14 00Lunch433
Afternoon Chair Vic Basili
14 00 - 15 30Experimentation and Measurement in FEAST Larry Votta418
15 30 - 16 00Tea433
16 00 - 17 30General Discussion418
6 30 -
7 00Cocktails
7 00 - 9 30Dinner Physics Common Room
level 8, Blackett Tuesday, 25 October 1994
09 15 - 09 45Coffee433
Morning Chair Colin Tully
09 45 - 10 15ICL Model Bob Snowdon418
10 15 - 10 45AT&T Model Dewayne Perry418
Larry Votta
10 45 - 11 15Coffee433
11 15 - 12 45Discussion418
12 45 - 14 00Lunch433
Afternoon Chair Bob Bishop
14 00 - 15 30Issues, Challenges and Opportunities Bob Balzer418
15 30 - 16 00Tea433
16 00 - 17 30Commitments and Funding418 NOTE:All topic titles and assignments are tentative and subject to the agreement of the individual concerned.
Alan Ainsworth,
77 High Road
Ickenham, Uxbridge,
Middlesex
tel: -+44 0895 630529,
aa5@https://www.doczj.com/doc/1a4197927.html,
Dr. Bob Balzer,
Information Sciences Inst
4676 Admiralty Way, Suite 1001, Marina del Rey, CA 90292-6695 tel:+1 310 822 1511
balzer @https://www.doczj.com/doc/1a4197927.html,
Professor Vic Basili,]
Dept. of Computer Science, University of Maryland,
College Park, MD 20742,
tel: +1 301 405 2668,
fax +1 301 405 6707
basili@https://www.doczj.com/doc/1a4197927.html,
Professor Keith Bennett
Durham University
Science Labs
South Road
Durham
DH1 3LE
Tel: 091 374 2632
keith@uk.ac.dur.easby
Bob Bishop,
18 Aggisters Lane
Wokingham, Berks, RG11 4DN tel: +44 (0)734 774017
fax: +44 (0)734 894 254
rb@https://www.doczj.com/doc/1a4197927.html,
Anthony Brigginshaw, ICSTM,
180 Queens Gate, London, SW7 2BZ tel: +44 (0)71 589 5111 ext 58248 fax: +44 (0)71 581 8024
awb@https://www.doczj.com/doc/1a4197927.html,
Dr Bill Curtis
TeraQuest Metrics & SEI
3644 Ranch Creek
Austin, Texas 78730-3701, USA Tel: 1 512 346 2435 :
Fax: 1 512 346 4677
email: curtis@https://www.doczj.com/doc/1a4197927.html,
Mark Dowson,
Marleston Software Tech. Inc.,
525K East Market Street 303 Leesburg, VA 22075,
USA tel: +1 703338 3951
Fax: +1 703 771 8413
dowson@https://www.doczj.com/doc/1a4197927.html,
(2031 Chadds Ford Drive)
703 476 9006 Fax. 703 264 1427 evenings 703 264 1427
Dave Homan,
Northern Telecom Europe Ltd, Oakleigh Road South,
London, N11 1HB
tel: +44 (0)81 945 3097
fax: +44 (0)81 945 3960
homan@https://www.doczj.com/doc/1a4197927.html, City University,
School of Informatics, Dept of Bus. Comp. Northampton Square, (Sep 94)
London EC1V OHB
tel: 477 8000 ex 3725
z.levy@https://www.doczj.com/doc/1a4197927.html,
Or: 48 Haparsa St,
Tel-Aviv 69085 (After Sep 94)
Israel
tel: + 972 3 6471 373
fax 972 3 566 1667
Prof. Meir M Lehman, ICSTM
180 Queens Gate,
London, SW7 2BZ
tel: +44 (0)71 594 8214
fax:+44 (0)71 594 8215
mml@https://www.doczj.com/doc/1a4197927.html,
Dr. Pertti Lounamaa
Nokia
P.O.Box 45, (Heikkil?ntie7) FIN-00211 Helsinki, Finland
tel: +358 0 437 6594
fax:358 0 455 2091
lounamaa@research.nokia.fi
Professor Nazim Madhavji,
McGill University
School of Computer Science
McConnell Engineering Bldg.,
3480 University St, Montreal, Quebec Ca. H3A 2A7
tel: 514 398 3740/514 284 6730
fax: 514 398 3883
madhavji@opus.cs.mcgill.ca
Dr. Bashar Nuseibeh,
ICSTM,
180 Queens Gate,
London, SW7 2BZ
tel: +44 (0)71 589 5111 ext 58286
fax:+44 (0)71 581 8024
ban@https://www.doczj.com/doc/1a4197927.html,
Dr Dewayne Perry,
AT & T Bell Laboratories
Room 2B 431,
600 Mountain Avenue,
Murray Hill,
New Jersey 07974, USA
tel: +1 908 582 2529,
fax: +1 908 582 7550
dep@https://www.doczj.com/doc/1a4197927.html,
David Redmond-Pyle
LBMS
Evelyn House
62 Oxford Street
London W1N 9LF
tel: 071 636 4213
fax: 071 636 2708
drp@https://www.doczj.com/doc/1a4197927.html,
Suzanne Robertson,
Atlantic Systems Guild Inc.,
11 St. Marys Terrace,
London, W2 1SU
tel: +44 (0) 262 3395
fax: +44 (0)71 262 1378
100065.2304@https://www.doczj.com/doc/1a4197927.html,
STC,
West Avenue
Kidsgrove,
Stoke on Trent, ST7 1TL
tel: +44 (0) 782 771000
fax: +44 (0) 782 776667
ras@https://www.doczj.com/doc/1a4197927.html,
Prof. Vic Stenning,
Warilda, Soames Lane, Ropley,
Hants SO24 OER
tel/fax: +44 (0)962 772531
vs6@https://www.doczj.com/doc/1a4197927.html,
Colin Tully
Colin Tully Associates
6 Meadow Hill Road Tunbridge Wells, Kent. TN1 1SQ Tel and Fax: 44 892 539125 Prof.Wlad M Turski,
Institute of Informatics University of Warsaw, Banacha 2,
00-903 Warsaw 59,
Poland
tel: +48 2 658 3522,
fax: +48 2 658 3164
wmt@https://www.doczj.com/doc/1a4197927.html,.pl
Dr Larry Vota
AT&T Bell Laboratories Napierville, IL
votta@https://www.doczj.com/doc/1a4197927.html,
Aidan Ward,
Coherent Technology Ltd., Suite 10, Challenge House, Sherwood Drive
Bletchley,
Milton Keynes, MK3 6DP
tel: +44 (0)908 274591
fax: +44 (0)908 274693
Aidan@https://www.doczj.com/doc/1a4197927.html,
Dr. Alan Whitfield
163 Sycamore
Wilnecote, Tamworth
Staffs, B77 5HF maintain their interest in FEAST
Dr Tarek K Abdel-Hamid,
Administrative Sciences,
Naval Postgraduate School, Monterey, CA 93943 tel: +1 408 656 2686,
fax: +1 408 656 3407
3991P@NAVPGS.EARN
Barry Boehm CS Dept.,
University of Southern Cal.,
LA CA 90089
USA
Tel: 213 740 8163
boehm@https://www.doczj.com/doc/1a4197927.html,
Valerie Downes, Ovum Ltd.,
1 Mortimer St., WlN 7RH
tel: +44 (0)71 255 2670
fax: +44 (0)71 255 1995
100350.1627@https://www.doczj.com/doc/1a4197927.html,
Dr Anthony Finkelstein,
ICSTM,
180 Queens Gate, London, SW7 2BZ
tel: +44 (0)71 589 5111 ext 48261
acwf@https://www.doczj.com/doc/1a4197927.html,
David Freestone, BT,
Section B81, 160PP27
Martlesham Heath, Ipswich 1P5 7RE
tel: +44 (0)473 646433
fax: +44 (0)473 640897
freestone_d@https://www.doczj.com/doc/1a4197927.html,
Professor Carlo Ghezzi
Dipartimento di Elettronica e
Informazione, Politecnico di Milano
Piazza L. da Vinci, 32
20133 Milano, Italia
Tel: 010 39 2 239 93411
ghezzi@elet.polimi.it
Dr. Peter Gibbins,
Sharp Labs of Europe Ltd.
Edmund Halley Road,
Oxford Science Park
Oxford, OX4 4GA
tel: +44 (0)865 747711
fax: +44 (0)865 747717
gibbins@https://www.doczj.com/doc/1a4197927.html,
Professor Peter Henderson
Dept. of Computer Science
Southampton University, University Road Highfield, SO17 1BJ
Southampton University
tel:01703 595000
Dr. Marc Kellner,
Software Engineering Institute
Carnegie Mellon University
Pittsburgh, PA 15213, USA
tel: +1 412 268 7721
Fax: +1 412 268 5758
mik@https://www.doczj.com/doc/1a4197927.html,
Prof. Peter Lee,
Department of Trade and Industry
151 Buckingham Palace Road,
London, SW1W 9SS
tel: +44 (0)71 215 1967
Prof. Stuart Madnick
MIT Sloan School of Management Room E53-321
20 Wandsworth Street Cambridge, MA 02139, USA email:smadnick@https://www.doczj.com/doc/1a4197927.html,
Prof. Tom Maibaum, ICSTM
180 Queens Gate, London, SW7 2BZ tel: +44 (0)71 594 8283/84
fax:+44 (0)71 594 8285
tsem@https://www.doczj.com/doc/1a4197927.html,
Jean Pierre Moularde,
SEMA (France)
16, rue Barbes
F-92126 Montrouge Cedex,
France
tel: +33-1-40 92 43 16
fax: +33-1-46 56 96 53
moularde@metra.fr
Secr phone: 33-1-40924218
Secr. fax: 33-1-40924241
Brian Oakley,
120 Reigate Road
Ewell, Epsom,
Surrey, KT17 2BX
tel: +44 (0)81 393 4096
fax: +44 (0)81 393 9046
Dr Bob Rockwell,
Softlab Gmbh.,
120, D-81677 Munich,
tel: +49 89 93 00 1475,
fax:+49 89 938 281
Home: Milloeckerstr. 34a,
D-81477 Munich, Germany
tel: 049 89 784 9002
email: roc@softlab.de
100317.1353@https://www.doczj.com/doc/1a4197927.html,
Paul Rogoway
26 Haneseim St.
Petach Tikva 49550, Israel
Tel: 010 972 3 388 950
h - 3 922 7744
car 5206666 sub. 50351
Dr. Berc Rustem, ICSTM
180 Queens Gate, London, SW7 1BZ tel:+44 (0) 71 594 8345
fax:+44(0) 71 5890 8024
br@https://www.doczj.com/doc/1a4197927.html,
Graham Samuel
39 Laurier Road, London, NW5 ISH Tel: 071 485 7916
email: graham@https://www.doczj.com/doc/1a4197927.html,
Gordon Scarrott, 34 Parkway Welwyn Garden City, Herts. AL8 6HQ tel: +44 (0)707 323 073
Owen Shallcross
19 Grahame Avenue Pangbourne, Berkshire
RG8 7LF
Tel: 0734 843107
Fax: 0734 842126 Northern Telecom Europe Ltd., Oakleigh Road South,
New Southgate, London, N11 1HB tel: +44 (0) 81 945 3199
fax: +44 (0) 81 945 3454
deps@https://www.doczj.com/doc/1a4197927.html,
Carolyn Story,
BNR Europe,
London Road
Harlow, Essex, CM17 9NA
tel: +44 (0) 279 403648
fax +44 (0) 279 437830
carrie@https://www.doczj.com/doc/1a4197927.html,
M M Lehman
1Introduction
The S-type, E-type and P-type classification for programs1(that is for software and computer applications) was first defined in the early seventies and has been further refined ever since. Most recently, as a result of discussions in the FEAST project the question of the classification and, in particular, of the distinctive properties of E-type systems has been raised. This note represents an initial statement of the currently recognised properties of the three types of program
2Definition
2.1S-type Program
An S-type program is a computer program that is required to satisfy some pre-stated specification.
That is, the specification is the sole, complete and definitive determinant of program properties. It is the only arbiter of correctness. Once a program has been developed correctness relative to that specification, in the strict mathematical sense, is the only criterion of success. The completeness, validity, relevance or ?sthetics of the specification and any program properties not explicitly included in the specification nor (formally) inferable from it, are extraneous issues. S-type programs are mathematical objects and artifacts. They must, at all times, be developed and processed with the rigour, precision and discipline that this implies and permits. In their context correctness has an absolute meaning. It is a specific, formally demonstrable, relationship between a program and its specification.
2.2E-type Program
A n E-type program is a computer program that realises (models) a computer application in some real world domain.
For real world applications it is the consequences of execution as, for example, the information conveyed to human observers or the behaviour induced in attached or controlled artifacts that determines the acceptability of the program. Ultimately, one is concerned with the value or level of satisfaction execution of the program yields. The concept of correctness that defines the acceptability of S-type programs expresses a precise relationship based on calculable equivalence between a program or other formal representation and some higher level specification. For E-type programs the real world is both specification and arbiter. Because the real world domain is unbounded calculable equivalence with it and, therefore, correctness relative to it, are meaningless concepts. Moreover, once installed, the system and its software become part of the application domain. Together they comprise a feedback system that cannot, in general, be precisely or completely known or represented. The specification is intrinsically volatile.
For E-type systems the notions replacing the Boolean concept of correctness are, therefore, essentially fuzzy, based on both qualitative and quantitative measures. System acceptability depends on a process of subjective human judgment. It is the detailed behaviour under operational conditions that is of concern.
2.3P-type Program
A P-type program is computer program that is required to produce an acceptable solution to some stated problem.
This was the definition of P-type programs as originally formulated. It was based on the observation that when acceptability of a problem solution rather than a relationship of correctness relative to a specification is of concern the program should not be classified as of type S even when though the problem statement can be formalised, its program instantiation rigorously developed and it can be proven correct. With this view if, when the program is analysed or executed, undesirable properties (side effects) not addressed in the problem statement are observed the problem statement must be modified to exclude them. A new version of the program may then be created to provide an acceptable solution. That is, this P-type program definition allows for changes in the specification to achieve a satisfactory outcome when the program is executed and the solution is applied.
More recently, an alternative view has emerged. This considers that if a complete and unambiguous statement of the problem has been provided so that the program may be completely specified it should be classified as of type S. despite the fact that program correctness relative to that specification is a means to the real end, achievement of a satisfactory solution. It requires one to accept that if, after implementation, the program product is unsatisfactory, for whatever reason, one produces a new problem statement (possibly by revision of the old) and infers a new program (possibly by reasoning based on the unacceptable program).
Which of these alternative views is the more useful, that is whether one adopts a two or a three member classification scheme requires further consideration. In any event,P-type programs are intermediate to the other two-types and it is not self evident that they possess any special properties. If such a program is completely specified it may and should be treated as an S-type otherwise as of type E.. The binary classification schema that results from this later viewpoint bisects the universe of programs into those for which correctness, as defined above, is sufficient and those for which (at the whole system level) it is not. The present analysis is restricted to S and E types.
3Some Characteristics
3.1S-type Programs
?An S-type program is completely defined by a specification
?Even if, initially, informally expressed that specification can always be transformed into a formal notation
?By definition the specification is fixed- it may not, for example, be changed because of dissatisfaction with the derived program ?Where circumstances call for revision of the specification, the revision is viewed as a new specification, and the resulting program as a new program- even though it may have been derived from the earlier program
1For convenience the term "X"- type is normally used to catagorise programs. It is equally meaningful to refer to "X"- type software or applications or to programs, software or applications of type X.
new program is, by definition, a process not coupled to the mathematical process of program derivation (development)
?The program may be derived from its formal specification by mathematical reasoning applying the logic associated with the notation- the specification is an extralogical axiom of which the program is a provable theorem
?Sole criterion of successful implementation is the program’s correctness- with respect to the specification
?Correctness is absolute- the program will have all the properties required by the specification independently of the individual(s) who derived it
?The starting point of the S-type development process is uniquely defined and fixed
?The S-type development process is an instance of logical and mathematical inference- it is well established though the specific techniques used may vary from instance to instance
?It may have additional properties not addressed by the specification- as determined by the derivation process and the individual(s) executing it
?The specification / program relationship is recursive- that is, a specification may, itself, be a program that is derived from (is a theorem of) a higher level specification (some more abstract extralogical axiom)
?Similarly, a program may be taken as a specification (extralogical axiom) from which a lower level program may be derived -for example an executable program compiled from a high level language program
?S-type programming elements are appropriate as the unit of individual programming assignment - no individual should ever be given an implementation assignment that is not fully specified
3.2E-type Programs
?By definition the real world is the operational domain- the application concept and its operational domain are initially unbounded and encompass the entire universe; any entity, process or phenomenon in that domain may influence execution
?User and usage satisfaction (acceptability?) determines the value of the system, it is the ultimate criterion of successful completion
?Any E-type system is a model of the application in its the real world domain
?As such, the the solution system contains a- possibly implicit - model of itself
?The domain of operation is, conceptually, unbounded
?It is also essentially continuous
?So is the application concept itself- there is no limit to the number or variety of the concepts, phenomena and behaviours that can be associated with it
?At least in so far as humans are involved in the processes and phenomena of the application and its computerised model, precise and complete theories/models of domain and application properties- including behaviour - cannot exist
?Neither application nor its operational domain can be precisely or completely represented/modelled by finite and discrete models-that is by software
?E-type software model can, therefore, not be complete - or precise
?To bridge the consequent gap between the real world application in its domain and its software model assumptions are made and embedded in the system- in the form of theories, algorithms, structures, parameters, sequencings etc
?The real world undergoes continuous change
?Since it involves human observation, interpretation, action, reaction, creative decision and so on such change is unpredictable ?Some of the assumptions embedded in the system will become invalid
?The consequences under execution of such invalidity are unpredictable
?However often an E-type program has been successfully, that is satisfactorily executed, it may fail to do so as a consequence of a, now, invalid assumption
?The existence, location or nature of such invalid assumptions cannot be known
?The outcome of any execution of an E-type program is uncertain,- unpredictable
?Fixing bounds for a program and its operational domain is an integral part of the evolution process2- elements contained in or excluded from the application or its domain will be referred to as application/domain included or excluded respectively
?Some bounds will be fixed explicitly others implicitly
?The fixing of the bounds includes pragmatic considerations- available expertise, limits of knowledge and understanding, financial resources, time to completion, available hardware, etc.
?As a consequence of changes in the real world- growth, extension, societal, technological, economic, perception, etc - included elements may become redundant and excluded elements may need to be included
?During execution of the evolution process the humans involved will acquire additional insight into the the application and its underlying theories, methods and techniques for its implementation and desirable characteristics of the solution system
?Such insights trigger changes to- generally, extensions of - the application and/or the domain and/or decisions reached in earlier stages of the evolution process
?The evolution process is a learning process with the acquired information applied to activities in the forward path of the process or used to modify earlier decisions
?Installation and operation (use) of an E-type system each change the operational domain- a further feedback effect
?All these phenomena/properties lead to pressure for continuing change
software development and software maintenance as
applied
?Learning and feedback driven evolution is an intrinsic property of E-type systems
?Those involved in the evolution process are essentially multidisciplinary representing at the very minimum application expertise and software development skills
?For any but the smallest applications/systems, organisational, financial and marketing and other problems imply strong involvement and project influence of agencies responsible for such areas
?Evolution of E-type systems, in general, requires interactions between many human populated agencies involving a wide variety of knowledge, understanding, experience and authority
?This implies a need for learning, elucidation, clarification, compromise
4Laws of Evolution
Seven laws of program evolution as originally formulated were inferred from observation and measurement of OS/360 as confirmed by measurements on other systems. Their interpretation was based on organisational and sociological behaviour. It is now possible to link them to the properties listed above with each law being related to certain of the properties. The first law Continuing Change, for example has to do inter alia, with closed loop nature of the process and the bounding of the domain. The sixth law, on the other hand while also associated with the closed nature of the process arises directly from the bounding of the application itself. Thus identifying and examining the E-type properties should prove a powerful means for achieving a sound underpinning for the laws and, perhaps, identifying additional laws. This work has, however, not yet been done in detail. For the present the laws are simply listed and some preliminary observations added.
I Law of Continuing Change
An E-type program that is used must be continually adapted else it becomes progressively l e s s satisfactory
- universal experience
- intrinsic property of computer application
- though all systems evolve, software is distinct in its rate of evolution and in that both individual artifacts and generations of artifacts evolve
- software must be regarded as an organism
- evolution Implemented in feedback driven and controlled maintenance process
II Increasing Complexity- the second law
As a program is evolved its complexity increases unless work is done to maintain or reduce it
-Change upon change upon change
- Spreading interactions and dependencies
- Balance between progressive and anti-regressive activity determined by feedback
III Fundamental Law of Program Evolution - self regulation
Program evolution is self regulating with statistically normal distribution of measures of product and process attributes
- Pseudo independent managerial and implementation decisions
- Feedback stabilisation
IV Conservation of Organisational Stability - invariant work rate
Average global activity rate invariant over product life time
- Feedback regulation
V Conservation of Familiarity- perceived complexity
During active life of an evolving program, content of successive releases is statistically invariant
- Limited by rate of learning and acquisition
- Feedback controlled
VI Continuing Growth
The functional content of a program must be continually increased to maintain user satisfaction
- Specification bounds become performance bottleneck, user irritants
- Feedback generated pressure for change
VII Declining Quality
E-type programs will be perceived as of declining quality unless rigorously maintained
- Existing assumptions become invalid
- New assumptions, constraints become necessary
- Users become more perceptive and expectant
- Acceptance criteria change
- Application demands adaptation
- Feedback generated pressure for change
General Remark
Laws derive from organisational behaviour, describe E-type evolution dynamics and constrain management freedom
- Maintenance increasingly difficult, costly, error prone as program ages
- Management freedom increasingly constrained by system
- Complexity control significant maintenance concern
- Life time expenditure high relative to initial development costs
- Limit to growth of E-type systems via classical r o u t e
Remarks
Since the characteristics described by the laws derive from the behaviour of people and organisations and not from the technological specifics of computer science and software technology they must be regarded as Laws from within those disciplines and taken into account in activities such as process design and process improvement. To overcome limitations implied by laws, the organisation within which and process whereby software is developed must be changed. Organisational behaviour and management are important considerations in software engineering
Uncertainty Principle
In addition to the laws behavioural analysis, now confirmed by identification of the characteristics of E-type systems leads to the statement of an Uncertainty Principle as already identified in the characteristics listing. The principle is simply stated here together with its currently identified sources. Linking this phenomenon with the identified characteristics in detail has not yet been undertaken.
Principle
The real world outcome of E-type software execution inherently uncertain with precise area of uncertainty also not knowable
- Outcome of any execution not absolutely predictable
- No guarantee of satisfactory result even if previous operation has been impeccable
Types of Uncertainty
Three sources of uncertainty have been identified of which one may be further subdivided.
Embedded
- validity of embedded assumption set cannotbe known nor responsively maintained
Sub-categorisation
Textual
- as sumptions embedded in program text about:
- operational environment
- solution system and its environment,
-algorithms, procedures, data etc.
Procedural
-assumptions about validity of development proces s,its steps, methods, tools
Heisenberg-like
- system development, installation and use changes application, solution and perception of both
- the more precise user knowledge and understanding of an application and its solution the more will views, expectations and criteria of satisfaction change
- validity of totality of assumptions about expectations or criteria of satisfaction cannot be known
G?del-like
- An E-type program is a model of a model……of a model of a real world application
- each model pair can be interpreted as a theory and a model of that theory or as a specification and its implementation
- hence every program is G?del incomplete
- G?del-like application uncertainty is reflection of G?del incompleteness
5Conclusion
The contents of this note are a summary of the observations, thoughts and conclusions of many years. As a first systematic attempt at detailed analysis and listing they must be taken as the input to a discussion rather than, in any way, conclusive. The note constitutes a working paper for the second FEAST workshop and, as modified by the discussion, for the FEAST activity beyond that. There is clearly there is scope or further thought. The last word has not been said.
Manny
mml535[papers/FEAST]
21 October 1994
Some Characteristics of S-type and E-type Software
Second FEAST Workshop
DoC
24 October 1994
M M Lehman
Department of Computing
Imperial College of Science Technology and Medicine
London SW7 2BZ
+44 71 589 5111 xt 5009
fax. +44 (0)71 581 8024
mml@https://www.doczj.com/doc/1a4197927.html,
?S-type-Program defined by fixed specification
-Starting point of development process absolutely defined
-Development open loop by definition
-Unit of individual programming assignment
-Criterion of successful implementation is correctness-
wrt specification
?E-type-Realises real world application
-Unbounded starting point
-Intrinsic closed loop development
-Multi-disciplinary, many person development
-Criterion of successful implementation user/usage satisfaction
?P-type-Solves specific problem
-Intermediate between S and E-types
Is it the characteristic properties of E-type systems that FEAST should be investigating?
?First attempt at detailed listing of both S-type and E-type characteristics provided in preprints
?Have also included statements of the Laws as derived in the past from observation, measurement, modelling of OS/360 growth as subsequently confirmed for other systems
?No systematic attempt yet to link the laws to the identified characteristics
?The first relationships have been identified
?Detailed analysis can greatly strengthen the theoretical foundations of FEAST
?And of software technology/process studies in general
A framework for process improvement
?Completely defined by a specification
?Specification can always be transformed into a formal notation
?Specification is fixed
?Where revision is required it is viewed as a new specification, and the resulting program as a new program- even though derived from earlier program
?Derivation (construction?) process is open loop- revision process is, by definition, not coupled to the mathematical process of program derivation
?Program derived from its formal specification by mathematical reasoning- specification an extralogical axiom of which program is a provable theorem
?Sole criterion of successful implementation is the program’s correctness- with respect to the specification
?Correctness is absolute- program will have all the properties required by specification independently of individual(s) who derived it
?Starting point of S-type development process uniquely defined
?The S-type development process an instance of mathematical inference- well established though specific techniques used may vary
?Program may have additional properties not addressed by specification - determined by derivation process and individual(s) executing it
?The specification / program relationship is recursive-specification may, itself, be a program that is derived from (is a theorem of) a higher level specification (some more abstract extralogical axiom)
?Similarly, a program may be taken as specification for derivation of a lower level program- as in compilation for example
?S-type programming elements are appropriate as the unit of individual programming assignment
?By definition the real world is the operational domain
?User and usage satisfaction (acceptability?) determines value and acceptability of system
?Any E-type system is a model of the application in its the real world domain
?As such, the the solution system contains a- possibly implicit - model of itself
?Domain of operation is, conceptually, unbounded
?It is also essentially continuous
?So is application concept - no limit to number or variety of concepts, phenomena and behaviours that can be associated with it
?Human involvment in application processes and phenomena and its computerised model, excludes precise and complete
theories/models of domain and application properties
?E-type applications and their operational domains cannot be precisely or completely modelled- by finite and discrete models,by software
?E-type software model can, therefore, not be complete - or precise
?Assumptions bridge gap between real world application in its domain and its software model are made, embedded in system
?The real world undergoes continuous change
?Human involvement makes such change unpredictable
?Some embedded assumptions in the system will become invalid
?Consequences under execution are unpredictable
?However often E-type program has been successfully executed, it may still fail as a consequence of a, now, invalid assumption
?Existence, location or nature of invalid assumptions not known
?Outcome of any execution of an E-type program uncertain
?Fixing bounds for program and operational domain an integral part of evolution process
?Some bounds will be explicit others implicit
?Design of bounds includes pragmatic considerations
?Changes in the real world may make included elements redundant, excluded elements may need to be included
?Humans involved in evolution process will acquire additional insight- application and its underlying theories, methods and techniques for its implementation, desirable characteristics of the solution system, etc.
?Such insights trigger changes to application, the domain and/or decisions reached in earlier stages of evolution process
?Evolution process a learning process- acquired information applied to activities in forward path of the process or used to modify earlier decisions
?Installation and operation (use) of an E-type system changes operational domain- a further feedback effect
?All these characteristics lead to pressure for continuing change
?E-type process is a multi-loop feedback process- inherently unstable unless adequate negative feedback controls are applied
?Learning and feedback driven evolution perhaps the intrinsic characteristic of E-type systems
?Multidisciplinary involvment essential in evolution process - at the very minimum, application expertise and software development skills
?For any but the smallest systems, organisational, financial, marketing and other problems imply strong involvement and
project influence of agencies responsible in such areas
?E-type system evolution, in general, requires interactions between many human populated agencies involving a wide variety of
knowledge, understanding, experience and authority
?Demands lea rning,elucidation,clarification,compromise
Systematic characteristics may be formalised
as Laws
信息科学与技术学院实验报告 课程名称: 计算机网络应用技术教程实验项目: WWW服务器配置 实验地点:指导教师: 日期: 2013/10/29 实验类型:验证性实验（验证性实验综合性实验设计性实验）专业: 班级: 11级姓名: 学号: 一、实验目的及要求 1.实验目的： 1.正确理解WWW服务的运行机制，了解常用的wed服务器软件。 2.掌握IIS服务器的安装和管理，创建wed站点利用IIS在一台服务器上运行多个网站。 3.掌握虚拟机主机和虚拟目录的创建删除。 2.实验要求： 1.理解IIS服务的概念及其所具有的功能。 2.掌握IIS服务的安装方法。 3.掌握WWW服务的配置包括IP地址、端口号、默认文档、安全等设定，以及如何应用WWW服务的方法。 4.了解虚拟目录服务的作用。 二、实验仪器、设备或软件 1.实验仪器：电脑一台 三、实验内容及原理 1.实验内容： (1).学会安装IIS。 (2).掌握www服务器的配置和使用。 (3).创建虚拟目录。 2.实验原理： 万维网WWW（World Wide Web）服务，又称为Web服务，是目前TCP/IP互联网上最方便和最受欢迎的信息服务类型，是因特网上发展最快同时又使用最多的一项服务，目前已经进入广告、新闻、销售、电子商务与信息服务等诸多领域，它的出现是TCP/IP互联网发展中的一个里程碑。 WWW服务采用客户／服务器工作模式，客户机即浏览器（Browser）,服务器即Web服务器，它以超文本标记语言（HTML）和超文本传输协议（HTTP）为基础，为用户提供界面一致的信息浏览系统。信息资源以页面（也称网页或Web页面）的形式存储在Web服务器上（通常称为Web站点），这些页面采用超文本方式对信息进行组织，页面之间通过超链接连接起来。这些通过超链接连接的页面信息既可以放
大学英语四六级写作体裁 1. Argumentation 2. Report 3. Caricature Writing 4. Proverb Commenting Argumentation 1. 文章的要求：On Students' Evalation of Teachers A:近年来，很多大学把学生对教师的评价纳入评估体系。B：就其利弊而言，人们之间存在争议。C: 你的看法。 On Students' Evaluation of Teachers With the reformation and development of higher education, many universities have gradually shifted their management from the traditional system of academic year to the system of credit, and from teacher-centered to student -centered teaching. As a result,teachers are now managed on the basis of students' evaluation, which has aroused heated disputes. Many people agree that students' opinion and appraisals are crucial to both the teaching process and the teacher-student relationship. For one thing, the teacher may reflect on his teaching, and then change accordingly.For another, mutual undersatnding is enhanced. However, many experts worry that some of the judgements are rather subjective, and even strongly biased. To make things worse, some students take it as a game--" anyway, my ideas do not count at all". Consequently, nobody will benefit from the valueless feedback. As far as I am concerned, students' views should be taken into consideration, but the views should not be taken as decisive factors when it comes to evaluating a teacher. The improvement of the mechanism requires the change of notion about teaching. We used to relate the evaluation to the employment of a teacher, but if we take advantage of the evaluation as a guide,it might be another story. Marriage upon Graduation 1. 越来越多的大学生一毕业就进入婚姻的殿堂。 2.产生这种现象的原因。 3. 我的看法。 A growing number of college students get married almost at the very moment of their graduation. As for this issue, we cannot help wondering what account for this trend. To begin with, there is no denying that the intensive employment competition is a major factor for this phenomenon. Getting married means the couple can support each other in the harsh environment and equally new couples can get the financial assistance from both families. Furthermore, compared to some instant marriages relying on web love or money worship, marriages upon graduation make a more romantic and reliable choice.
----Web服务器配置 实验目的 （1）根据实验的目标和要求画出网络拓扑图，查找IP地址，制定域名规划，提出各种服务器配置方案； （2）掌握在windows server 2008环境下架构web网站； （3）掌握测试web服务器的方法。 实验环境 实验分两人一组，每组每人分配一台PC，一个windows server2008安装ISO文件。 实验要求 假设每位同学成立了自己的公司，现要求在公司内部安装windows server 2008，用自己的姓名拼音作为域名，发布自己公司的网站，建立内部的web服务，可以通过客户机访问和实现web服务。 实验步骤 1、步骤一：安装Web服务 windows server2008上安装Web服务器步骤 1）点击“开始”->“管理工具”->“服务器管理器”，进入服务器管理工具界面。
2）在服务器管理工具界面时面点击“角色”->“添加角色”，进入到添加角色界面 3）在选择服务器角色对话框中将“Web服务器（IIS）勾选上，点击下一步。 4）在Web服务器（IIS）对话框中，直接点击下一步 5）在选择"Web服务器"角色是，会弹出如图所示窗口，点击"添加必需的功能"，然后点击"下一步"。如图所示： 6）在"Web服务器简介（IIS）"界面，直接点击"下一步"。 在"选择角色服务"界面，默认设置，直接点击"下一步"。如图所示： 7）点击安装，进入安装界面。 8）安装完成后，点击 9）web服务器安装成功界面
2.步骤二：设置Web服务 1)运行Internet 信息服务(IIS)管理器 2)选择网站右键添加网站，新建一个“MyWeb站点” 3)填写添加网站信息 4)添加首页文件：转到“文档”窗口，再单击“添加”按钮，根据提示在“默认文档”后输入用户自己网页的首页文件名“”。 5）添加虚拟目录。例如主目录在F:\web目录下，输入“的访问方式就可以调出“F:\web”中的网页文件，这其中的“test”就是虚拟目录。在“MyWeb站点”处单击鼠标右键，选择“新建”-“虚拟目录”，依次在“别名”处输入“test”，在目录出输入“F:\web”后再按照提示操作即可添加成功。 6) 点网站右键—管理网站—重新启动 7）在客户机打开浏览器以IP地址访问测试 8）添加更多的Web站点，因为本实验测试的计算机只有一块网卡，所以采用“一个IP地址
F T P服务器配置实验报告 Prepared on 22 November 2020
实验报告 课程：计算机网络实验 实验名称： FTP服务器配置与管理 系别 : 电子信息工程系 实验日期 : 专业班级 : 03通信师 组别 : 第10组 实验报告日期 : 姓名 : 学号 : (40) (41) 报告退发 : ( 订正、重做 ) 第1页共 12 页 FTP服务器配置与管理 一.题目: FTP服务器配置与管理 二.环境: Sever2000 三.试验目的 1.掌握FTP服务的基本概念与工作原理 2.懂得安装FTP服务器的过程 3.配置与管理FTP服务器 四.试验内容及步骤 1.的安装，具体步骤如下： (1)运行“控制面板”中的“添加或删除程序”，点击“添加/ 删除Windows组件”按钮。 第 2 页共 12页 (2)在出现组件安装向导中，选择“Internet信息服务 （IIS）”，单击“下一步”开始安装，单击“完成”结 束。 第 3 页共 12 页 系统自动安装组件，完成安装后，系统在“开始”/“程序”/“管理工具”程序组中会添加一项“Internet服务管理器”，此时服务器的WWW、FTP等服务会自动启动。 2.设置FTP站点 第 4 页共 12 页 (1)使用IIS默认站点
①将制作好的主页文件（html文件）复制到 \Inetpub\ftproot目录，该目录是安装程序为默认FTP站点 预设的发布目录。 ②将主页文件的名称改为。IIS默认要打开的主页文件是 或，而不是一般常用的。 完成这两个步骤后，打开本机或客户机浏览器，在地址栏 中输入FTP服务器的 IP地址（）或主机的FQDN名字（前 提是DNS服务器中有该主机的记录），就会以匿名的方式 登录到FTP服务器，根据权限的设置就可以进行文件的上 传和下载了。 (2)添加新的FTP站点 ①打开“Internet信息服务窗口”，鼠标右键单击要创建 新站点的计算机，在弹出菜单中选择“新建”/“FTP站 点”，出现“FTP站点创建向导”，单击“下一步”继 续。 第 5 页共 12 页 ②输入FTP站点说明,单击下一步 第 6 页共 12 页 ③ 单击下一步 ④指定FTP输入主目录的路径(如选择新建文件夹),单击下一步 第 7 页共 12 页 ⑤设置访问权限为读取和写入,并单击下一步,完成FTP站点创建向导 第 8 页共 12 页 站点的管理 (1)本地管理 通过“开始”/“程序”/“管理工具”/“Internet服务管理 器”，打开如图9-1的“Internet信息服务”窗口，在要管 理的FTP站点上单击鼠标右键，选择“属性”命令，出现如下 图所示对话框。 第 9 页共 12 页 ①“FTP站点”属性页 IP地址：设置此站点的IP地址，即本服务器的IP地址。 如果服务器设置了两个以上的IP站点，可以任选一个。FTP 站点可以与Web站点共用IP地址以及DNS名称，但不能设置 使用相同的TCP端口。 TCP端口：FTP服务器默认使用TCP协议的21端口，(若端口号21以被配置,则需更改此端口，用户在连接到此站点时，
大学英语四六级作文 2019-11-20 大学英语四级范文：说谎 题目要求： Directions: Write a composition entitled Telling Lies. You should write at least 120 words according to the outline given below in Chinese: 1. 人们对说谎通常的看法; 2. 是否说谎都是有害的?说明你的看法。 3. 总结全文。 参考范文： Sample: Telling Lies Telling lies is usually looked upon as an evil, because some people try to get benefit from dishonest means or try to conceal their faults. However,_I_thinkdespitejits_neg1ative_effects,jsometimes_itjisjessent ialto iemigsinour_dailylifei First, the liar may benefit from the lie by escaping from the pressure of unnecessary embarrassment. Meanwhile, the listener may also feel more comfortable by reasonable excuses. For ex ample, if a little girl’s father died in an accident, her mother would comfort her by saying “farther has gone to another beautiful land”. In such cases, a lie with original goodwill can make the cruel nice. Second^thejsk^lsjofJtellingJlies^jtojsome extent,canbe_regard_asja_capacity_ofcreationandjimagination. Therefore,jtakingjallJthese_factor^jntojconsideration,_wecandefinitel y come to the conclusion that whether telling a lie is harmful depends on its original intention and the ultimate result it brings. (159 words)
在Linux下配置Apache服务器 一、实验目的 完成本次实训，将能够: ●配置基本的Apache服务器 ●配置个人用户Web站点。 ●配置虚拟目录别名功能。 ●配置主机访问控制。 ●配置用户身份验证功能.。 ●配置基于IP地址的虚拟主机. 二、实验环境 1、RedHat Linux4AS. 2、Apache 2.0 三、实验内容 1.配置基本的Apache服务器 2.配置个人用户Web站点。 3.配置虚拟目录别名功能。 4.配置主机访问控制。 5.配置用户身份验证功能.。 6.配置基于IP地址的虚拟主机。 四、实验要求 在Linux操作系统下配置Apache服务器。 五、注意事项 1.在修配置文件下注意区分大小写、空格。 2.在每次重新开机后都必须启动Apachec服务器。 3.在每次修改完主配置文件后保存起来，必须重启Apachec服务器，如果不重启会 导致配置无效，最终导致实验失败。 六、实验步骤 1、检测是否安装了Apache软件包: A、首先为服务器网卡添加一个固定的IP地址。 B、在Web浏览器的地址栏中输入本机的IP地址，若出现Test Page测试页面（该 网页文件的默认路径为var/www/html/index.html）如下图1所示就说明Apache 已安装并已启动。
另一种方法是使用如下命令查看系统是否已经安装了Apache软件包： [root@rhe14~]# rpm –aq | grep httpd Httpd-suexec-2.0.52-9.ent Httpd-manual-2.0.52-9.ent System-config-httpd-1.3.1-1 Httpd-devel-2.0.52-9.ent 出现以上内容表明了系统已安装Apache软件包。 2、安装Apache软件包 超级用户（root）在图形界面下选择“应用程序”|“系统设置”|“添加/删除应用程序”命令，选择“万维网服务器”软件包组，在单击“更新”按钮就可以安装与Apache相关的软件包。 3、Apache的基本配置 （1）打开终端输入[root@rhe14~]# /etc/rc.d/init.d/httpd start //启动Apache 或者 [root@rhe14~]# apachectl start //启动Apache [root@rhe14~]# apachectl stop //停止Apache服务 [root@rhe14~]# apachectl restart //重启Apache服务 [root@rhe14~]# apachectl configtest //测试Apache服务器配置语法（2）在httpd.conf将Apache的基本配置参数修改、将一些注释的语句取消注释，或将某些不需要的参数注释掉。 （3）将包括index.html在内的相关网页文件复制到指定的Web站点根目下（var/www/html/index.html） （4）重启httpd进程 (5) 在Web浏览器下输入配置的ip地址出现如下图2，那表明基本配置成功了:
email营销邮件模板 篇一:电子邮件营销之优秀邮件模板制作秘诀 如何制作一个讨人喜欢的邮件模板，让顾客眼前一亮,是众多邮件营销人员头痛的事。因为水能载舟，亦能覆舟。好的EDM邮件模板可以轻松引导顾客注意你希望表达的重要信息，反之，将导致他们觉得不知所云，让整个电子邮件营销前功尽弃。 U-Mail邮件群发平台总结发现，制作一套优秀的电子邮件模板的秘诀，重点遵从以下规则: 一、格式编码注意事项: 1.邮件内容页面宽度在600到800px(像素)以内，长度1024px内 2.电子邮件模板使用utf-8编码 3.源代码在10kb以内 4.用Table表格进行布局 5.邮件要求居中时，在table里设定居中属性 6.文件不能由WORD直接转换而成 7.不能有外链的css样式定义文字和图片 8.不要有动态形式的图片 1 9.没使用<table</table以外的body、meta和html之类的标签 10.背景图片代码写法为:<table background=background.gif cellspacing=0 cellpadding=0 11.没有出现onMouseOut onMouseOver 12.没使用alt
13. font-family属性不为空 二、电子邮件模板文字规则: 1.邮件主题在18个字以内，没有――～??等符号 2.邮件群发主题和内容中不存在带有网站地址的信息 3.文字内容、版面简洁，主题突出 4.没有敏感及带促销类的文字 5.本次模板发送不超过20万封 三、电子邮件模板图片规则: 1.尽量使用图片，减少文字量 2.整页图片在6张以内，每张图片最大不超过15kb 3.图片地址不是本地路径 4.图片名称不含有ad字符 5.整个邮件模板不是一张图 四、电子邮件模板链接规则: 1.链接数量不超过10个 2 2.链接是绝对地址 3.链接地址的长度不超过255个字符 4.不使用map功能 5.制作了一份和邮件内容一样的web页面，在邮件顶部写: “如果您无法查看邮件内容，请点击这里”，链接指向与邮件营销内容一致的web网页。 篇二:英语外贸邮件模板大全, 各类正式外贸电子邮件开 头模板, 英文商务回复跟进email模板
web服务器的配置实验报告 篇一：计算机网络实验报告——Web服务器的配置 实验2 web服务器配置 一、实验目的： 掌握如何使用windows XX server的IIS5.0配置出web 服务器 二、实验内容： 1、创建一个web站点，并且可以实现在别人的计算机上访问该站点 2、使用不同的ｉｐ建立多个站点 3、在一个站点下建立多个子站点（使用虚拟目录实现） 4、在同一个套接字（即ip地址+端口）上建立多个站点（使用加主机头名方法实现） 5、对站点进行安全管理（如浏览权限、帐号的使用、ｉｐ地址的设定） 三、实验要求： 一定要保证让别人正常的访问你建立的站点，并使实验结果达到预期的目的！ 四、实验步骤： 1. 使用当地IP地址建立web站点 （1）准备工作： ①关闭Windows 防火墙
实验中，为了我们所建的站点能够被成功访问，先将Windows 防火墙关闭。如图： ②IIS功能设置 控制面板\所有控制面板项\程序和功能---“打开或关闭windows所有功能”： 出现了安装Windows功能的选项菜单，在“Internet信息服务”中手动选择需要的功能，如下图： ③下载“花生壳软件”到本地，申请免费域名mqqfhg。 这样，完成了前期的所有准备工作，开始进行web服务器的建设。 （2）开始建立web站点 ①创建web站点“酒窝” 打开“控制面板”——“管理工具”—“ Internet 信息服务(IIS)管理(本文来自：小草范文网:web服务器的配置实验报告)器”——右击“网站——“添加网站——选择“IP地址”及“物理路径”： 篇二：实验六web服务器配置实验报告 XX-XX学年第一学期课程实验报告课程名称：计算机网络 实验名称： 篇三：Web服务器的配置实验报告 实验5 Web服务器的配置
实验10 Web服务器的配置 一、实验目的 1、了解Web工作原理； 2、掌握基于IIS的Web服务器的创建和配置； 二、实验要求 1、设备要求：计算机1台（装有Windows Server 2003操作系统及活动目录、装有网卡）， 集线器、交换机，UTP线（直通、交叉）；Windows Server 2003安装源； 2、每组1人，独立完成。 三、实验基础 Web服务的实现采用B/W（Browser/Web Server）模式，Server信息的提供者称为Web 服务器，Browser信息的获取者称为Web客户端。Web服务器中装有Web服务器程序，如：Netscape iPlanet Web Server、Microsoft Internet Information Server、Apache等；Web 客户端装有Web客户端程序，即：Web浏览器，如：NetscapeNavigator、Microsoft Internet Explorer、Opea等。 Web服务器是如何响应Web客户端的请求呢？Web页面处理大致分3个步骤： 第一步：Web浏览器向一个特定服务器发出Web页面请求； 第二步：收到Web页面请求的Web服务器寻找所请求的页面并传送给Web浏览器； 第三步：Web浏览器接收所请求的Web页面并将其显示出来。 Web应用的基础还包括HTTP和HTML两个协议。 HTTP协议是用于从Web服务器传输超文本到本地浏览器的传输协议。它使浏览器的工作更高效，从而减轻网络负担；它不仅使计算机传输超文本正确、快速，而且可以确定传输文档的哪一部分以及哪一部分的内容首先显示等。HTTP使用一个TCP/IP连接，默认TCP80端口。 HTML 是用于创建Web文档或页面的标准语言，由一系列的标记符号或嵌入希望显示的文件代码组成，这些标记告诉浏览器应该如何显示文章和图形等内容。 四、内容与步骤 1、IIS6.0的安装、配置和测试； （1）从控制面板安装IIS； 第1步单击“开始” →“添加或删除程序” →“添加/删除Windows组件”，显示“Windows 组件向导”窗口，在“组件”列表框中依次选择“应用程序服务器” →“详细信息”，显示“应用程序服务器”窗口，选中“https://www.doczj.com/doc/1a4197927.html,”复选框以启用https://www.doczj.com/doc/1a4197927.html,功能，如图所示。
大学英语四六级考试万能作文模板 在CET4/6试题中，有两个拉分最厉害的部分，一是听力题，一是 作文题。，跟汉语作文的“起承转合”类似，英语作文其实也有固 定模式。 我到那里一看，果然有五个模板，认真拜读一遍之后，不由得 大吃二惊。第一惊的是：我们中国真是高手如林，多么复杂、困难 的问题，都能够迎刃而解，实在佩服之极。第二惊的是：“八股文”这朵中国传统文化的奇葩，在凋谢了一百多年之后，竟然又在中华 大地上重新含苞怒放，真是可喜可贺。 感慨之后，转入正题。这五个模板，在结构上大同小异，掌握 一种即可，所以我从中挑选了一个最简单、最实用的，稍加修改， 给各位介绍一下。这个模板的中文大意是：在某种场合，发生某种 现象，并提供一些相关数据，然后列出这种现象的三个原因，并将 三个原因总结为一个最主要原因，最后提出避免这种现象的两个办法。总的来说，利用这个模板写英语作文，是相当容易的，您只要 将适当的内容，填写到对应的方括号中，一篇通顺的英语作文即可 完成。下面就是这个模板。 Nowadays, there are more and more [某种现象] in [某种场合]. It is estimated that [相关数据]. Why have there been so many [某种现象]? Maybe the reasons can be listed as follows. The first one is [原因一]. Besides, [原因二]. The
third one is [原因三]. To sum up, the main cause of [某种现象] is due to [最主要原因]. It is high time that something were done upon it. For one thing, [解决办法一]. On the other hand, [解决办法二]. All these measures will certainly reduce the number of [某种现象]. 为便于读者理解，我特意用这个模板，写了一篇关于ghost writer(捉刀代笔的枪手)的示范性小作文，请您观摩一下。 Nowadays, there are more and more [ghost writers / 枪手] in [China's examinations / 中国的考场]. It is estimated that [5％ examinees are ghost writers / 5％的应试者是枪手]. Why have there been so many [ghost writers / 枪手]? Maybe the reasons can be listed as follows. The first one is [hirers' ignorance / 雇主无知]. Besides, [hirers' indolence / 雇主懒惰]. The third one is [hirers' obtusity / 雇主迟钝]. To sum up, the main cause of [ghost writers / 枪手] is due to [hirers' low IQ / 雇主智商低]. It is high time that something were done upon it. For one thing, [flagellation / 鞭打]. On the other hand, [decapitation / 斩首]. All these measures will certainly reduce the number of [ghost writers / 枪手].
1.营销的计划和目的 2017年是邮件营销突飞猛进的一年。社交媒体和移动互联网发展迅速，传统的邮件营销方式在与微博、微信、QR 码、人人网等社交网络方式的积极融合中，展现了新的营销活力，邮件营销的可拓展性得到延伸，我们企业把邮件营销列为重要营销工具之一。为推广我们名裳的服装在线营销，吸引广大客户的加入和购买我们制定了一系列计划： 1.先整理出要发送的主题和内容，配置图文最好。 2.甄选客户，最好挑一些有代表性的，可以提供建议的，新老客户都可以。 3.发送邮件，一定要确保邮件到收信人手中。 4.初步统计邮件的开启率及邮件内容链接的跟踪率‘ 5.做好统计记录，这次邮件发送出现的问题及时改进。 目的：1、提升公司及产品品牌知名度和认知度 2、建立庞大的客户信息邮件数据库，可以有计划有组织地开展营销活动 3、与客户建立良好的关系，提高客户的品牌忠诚度 4、让客户第一时间获取公司及相关行业内信息 5.能够探测市场, 找到新的市场机会和提供新产品与新服务 6、可实现一对一的销售模式 7、必须将宣传对象引到网站来，因为网站才能提供详尽的信息，才更有说服力。 2.如何获取邮件列表 1.发动注册会员收集邮件列表 邀请了人同时在网站注册，初次购买可以享受折扣、送积分等等，要顾及注册会员的利益。获得的邮件列表，再拿来群发邮件。 2.做广告吸引用户主动订阅 把邮件地址印在名片、文具、发/票、传真、印刷品上，即使在广播电视广告上也留下邮件地址，这样，便于顾客通过邮件和你联系， 3.在报刊上发布新闻 在报刊上发表人们必须通过邮件才可以接收的免费报告，或者可以通过邮件发送的产品，可以发表一些引起读者共鸣的话题，在读者回应的过程中收集其姓名和邮件地址。 4.现有客户的邮件列表 下次与客户通过邮寄或邮件联系时，可以为通过邮件回复者提供特别的服务，例如一份免费报告或者特别折扣优惠等，你必须尽可能地得到顾客的邮件地址。5.购买邮件列表 现在有很多的公司有邮件列表买，很多人也是选择的这个方法，如果选择的渠道比较好的话，收效也是相当快的，当然花费的费用也不低。这些数据一般都是有比较详细的数据，不单单是一个邮箱，有性别、年龄、职业、地区等，购买的时候最好先分析一下对自己产品感兴趣的人员的一些特征，然后筛选后购买，这样相关性会高很多，自己以后群发邮件的时候效果也会好很多。当然购买邮件列表最好能够选择一个好一些的机构，信誉要高些的，否则大部分邮件地址都无效，也就白白浪费钱了。 6.做好邮件列表管理
实验报告 一、实验目标及要求 1、实习目标：（1）掌握windows环境下web服务器的配置。 （2）学会对当今主流的web开发环境ASP和JSP配置自己 的web服务器，为今后开发网站打下基础。 2、实习要求：（1）通过本次实验，理解WEB服务的工作原理； （2）掌握在Windows系统上实现WEB服务，IIS和TOMCAT 服务器配置，ASP和JSP工作环境的配置。 二、实验步骤 1、实验环境描述（包括硬软件）： IIS是Internet Information Server的缩写，它是微软公司主推的Web服务器，最新的版本是Windows2000里面包含的IIS 5.0，IIS与WindowNT Server 完全集成在一起，因而用户能够利用Windows NT Server和NTFS（NT File System，NT的文件系统）内置的安全特性，建立强大、灵活而安全的Internet和Intranet 站点。 IIS的一个重要特性是支持ASP。IIS 3.0版本以后引入了ASP，可以很容易的张贴动态内容和开发基于Web的应用程序。IIS的安装，管理和配置都相当简单。 IIS包括WEB、FTP、Mail等等服务器。因为IIS的FTP和Mail服务器不是很好用，一般用IIS只用其WEB服务器。 目前运行IIS的最理想平台是Windows2000高级服务器版和服务器版。 Win9x/Me里也有IIS，但只是PWS（个人WEB服务器），功能很有限，只支持1个连接。WinXP里的IIS也只支持10个连接。如果您要使用IIS，请使用Windows 2000高级服务器版，在这里我用Windows 2000高级服务器版给大家做演示。 2、操作步骤： IIS的安装步骤（将WINDOWS 2000的安装光盘放入光驱内）： 1、点击：【开始】->【设置】->【控制面板】->【添加/删除程序】，打开“添加/删除程序”对话框。 2、点击左边菜单栏中“添加删除Windows组件”项，打开“Windows组件向导”对话框（这个过程可能需要花几十秒钟），在对话框“组件”列表中，选中“Internet 信息服务（IIS）”项。 3 、将“Windows 2000”安装盘放入光驱中，然后点击“下一步”。这个步骤可能会花几分钟。 4、点击“完成”按钮，完成IIS的安装。 IIS的配置步骤：
实验报告 课程名称计算机网络基础实验项目 FTP服务器的配置 专业班级 0906603 姓名学号 27 指导教师陈伟宏老师成绩日期2011.11.12 一、实验目的 掌握如何在局域网内配置FTP服务器。 二、实验设备和环境 局域网内多台个人计算机、Windows 2003操作系统。 三、实验内容 1、安装IIS或Serv-U； 2、配置及管理FTP服务器； 3、使用FTP服务。 四、实验过程 1、安装IIS V5.1 for 2003 截图如下： 点“详细信息”
选择Internet 信息服务(IIS),点详细信息.再选择“文件传输协议（FTP）服务” 2、FTP服务器的配置 启动IIS信息管理：控制面板——管理工具——IIS信息管理，选择FTP站点。右键新建FTP站点。
3.右击FTP站点的默认FTP站点的属性设置主目录F:\学习资料 4、设置安全账户为只允许匿名连接
5、测试本地ftp站点：在浏览器中输入ftp://192.168.137.3访问结果如下：
五、实验心得 这次试验为FTP服务器的配置。总的来说，由于上次已经做过web服务器的配置，而ftp的配置跟它大致相同，所以过程相对来说还是比较顺利，出现的问题也不多。 不过在实验过程中，自己只是按照老师的《FTP服务器的配置示例》一步步去做，实验虽然很成功地完成了，但在实验过程中我感觉自己并没有完全的、真正的“消化”理解好其中的含义，于是又反复地理解了一下各个步骤的原理。 通过这次实验理解了FTP服务器的工作基本原理，以及匿名访问和非匿名访问的一些相关设置，文件具有长传和下载的权限，对文件安全性的控制等等。 同时，也让我又学会了一种传送文件的新方法：只需要通过构建局域网，然后通过FTP就可以实现资源共享啦。感觉非常有用。 能把知识用到实处才是真正的学好了知识，这也是我们做实验的真正目的。以后我会继续努力提高自己的动手操作能力，把知识付诸于实践，同时在实践中更加深刻地理解知识。
F T P服务器配置实验报 告 Company Document number：WUUT-WUUY-WBBGB-BWYTT-1982GT
实验报告 课程：计算机网络实验 实验名称： FTP服务器配置与管理 系别 : 电子信息工程系 实验日期 : 专业班级 : 03通信师 组别 : 第10组 实验报告日期 : 姓名 : 学号 : (40) (41) 报告退发 : ( 订正、重做 ) 第1页共 12 页 FTP服务器配置与管理 一.题目: FTP服务器配置与管理 二.环境: Sever2000 三.试验目的 1.掌握FTP服务的基本概念与工作原理 2.懂得安装FTP服务器的过程 3.配置与管理FTP服务器 四.试验内容及步骤 1.的安装，具体步骤如下： (1)运行“控制面板”中的“添加或删除程序”，点击“添加/ 删除Windows组件”按钮。 第 2 页共 12页 (2)在出现组件安装向导中，选择“Internet信息服务 （IIS）”，单击“下一步”开始安装，单击“完成”结 束。 第 3 页共 12 页 系统自动安装组件，完成安装后，系统在“开始”/“程序”/“管理工具”程序组中会添加一项“Internet服务管理器”，此时服务器的WWW、FTP等服务会自动启动。 2.设置FTP站点 第 4 页共 12 页 (1)使用IIS默认站点
①将制作好的主页文件（html文件）复制到 \Inetpub\ftproot目录，该目录是安装程序为默认FTP站点 预设的发布目录。 ②将主页文件的名称改为。IIS默认要打开的主页文件是 或，而不是一般常用的。 完成这两个步骤后，打开本机或客户机浏览器，在地址栏 中输入FTP服务器的 IP地址（）或主机的FQDN名字（前 提是DNS服务器中有该主机的记录），就会以匿名的方式 登录到FTP服务器，根据权限的设置就可以进行文件的上 传和下载了。 (2)添加新的FTP站点 ①打开“Internet信息服务窗口”，鼠标右键单击要创建 新站点的计算机，在弹出菜单中选择“新建”/“FTP站 点”，出现“FTP站点创建向导”，单击“下一步”继 续。 第 5 页共 12 页 ②输入FTP站点说明,单击下一步 第 6 页共 12 页 ③ 单击下一步 ④指定FTP输入主目录的路径(如选择新建文件夹),单击下一步 第 7 页共 12 页 ⑤设置访问权限为读取和写入,并单击下一步,完成FTP站点创建向导 第 8 页共 12 页 站点的管理 (1)本地管理 通过“开始”/“程序”/“管理工具”/“Internet服务管理 器”，打开如图9-1的“Internet信息服务”窗口，在要管 理的FTP站点上单击鼠标右键，选择“属性”命令，出现如下 图所示对话框。 第 9 页共 12 页 ①“FTP站点”属性页 IP地址：设置此站点的IP地址，即本服务器的IP地址。 如果服务器设置了两个以上的IP站点，可以任选一个。FTP 站点可以与Web站点共用IP地址以及DNS名称，但不能设置 使用相同的TCP端口。 TCP端口：FTP服务器默认使用TCP协议的21端口，(若端口号21以被配置,则需更改此端口，用户在连接到此站点时，
大学英语四六级考试作文宝典 Nowadays, there are more and more [某种现象] in [某种场合]. It is estimated that [相关数据]. Why have there been so many [某种现象]? Maybe the reasons can be listed as follows. The first one is [原因一]. Besides, [原因二]. The third one is [原因三]. To sum up, the main cause of [某种现象] is due to [最主要原因]. It is high time that something were done upon it. For one thing, [解决办法一]. On the other hand, [解决办法二]. All these measures will certainly reduce the number of [某种现象]. 为便于读者理解，我特意用这个模板，写了一篇关于ghost writer(捉刀代笔的枪手)的示范性小作文，请您观摩一下。 Nowadays, there are more and more [ghost writers / 枪手] in [China's examinations / 中国的考场]. It is estimated that [5% examinees are ghost writers / 5%的应试者是枪手]. Why have there been so many [ghost writers / 枪手]? Maybe the reasons can be listed as follows. The first one is [hirers' ignorance / 雇主无知]. Besides, [hirers' indolence / 雇主懒惰]. The third one is [hirers' obtusity / 雇主迟钝]. To sum up, the main cause of [ghost writers / 枪手] is due to [hirers' low IQ / 雇主智商低]. It is high time that something were done upon it. For one thing, [flagellation / 鞭打]. On the other hand, [decapitation / 斩首]. All these measures will certainly reduce the number of [ghost writers / 枪手]. 四六级答案规律 1。ABCD分布均匀。 即：20个阅读理解，ABCD基本上是个5个 但近年来出现4664的分布，只有一次出现了8个D 预测今年：ABCD分别为6446 2。不可能出现3个答案相同的，例如：AAA这样的答案一般来说是有问题的。 3。词汇题是ABCD7887，最多的可以达到9个。 如果出现了10个C很有可能有问题。 声明：这种规律只能用来检查，不能用用来作题 4。听力题，前面十个听到什么就不要选什么，后面十个听到什么选什么。 5。作文应该是图表题。 切忌使用“There is................”."I think that............."这样的句子 四六级写作必备（背）句型 （1）用于描写漫画、图表的常用句型 ①As the graph d epicts , … ②From the cartoon /picture , we can see that … ③According to the statistics shown in the first /second graph , ④The table shows / indicates / reveals that … ⑤It can be seen /concluded from the picture / table / figures that … （2）用于句首提出论题或现象的句型 ①Recently , …has become the focus of the society . ②…has been playing an increasingly important role in our day-to-day life . ③Nowadays there is a growing concern for …