tics and computer science

11
 Informatics and Computer Science Informatics Informatics is one branch of science that is quite broad because nearly covers and associated with a variety of things about this l ife, where a collection of disciplines include science and informatics techniques to specifically process data into information by optimally utilizing information technology or computer. In the Indonesian language, the term Informatics INFORMATIQUE derived from French, which in German is called Informatik. Actually, the word is synonymous with the term computer science in the United States and computing sciences in the UK. In terms pendefenisian informatics, according to Philippe D reyfus (1962) and l'Academie Francaise (1967) that defines informatics as follows: Disciplines collection (scientific discipline) and the Discipline of Engineering (Engineering discipline) are specifically related to the transformation / processing of the "Facts Symbolic" (data / information), which mainly use the facilities automatic machines / computers. In English the meaning is slightly different, ie more emphasis on aspects of information processing in a systematic and rational. If taken as a whole from an understanding of informatics at the top, logic and systematic approach is a fairly dominant feature of Informatics, given the approach is key in terms of getting a solution in solving various problems. In principle this scholarship emphasizes how a data and information can be processed in such a technology-assisted automated. The technology terautomatisasi is not only in one machine, but can involve multiple machines. This machine is more commonly called a computer. Computer as the main ingredient in this scientific field has a very high, so simply peeling informatics ranging from how the machine works, how the data is processed in a manner that is understandable by the machine while the information is well understood by humans, to how the machine is able to communicate with other machines. Computer Science In general, Computer Science, better known by the name of Computer Science (Computer Science, CS) is the study o f computing, both hardware (hardware) and software (the software). Computer science covered topics related to computers, from abstract analysis of algorithms to more concrete subjects like programming languages, software, and hardware. As a discipline, Computer Science is different from computer programming, software engineering and computer engineering, even if the third term is often misunderstood. Church-Turing thesis states that all the computational tools that have been commonly known dalah actually the same thing what they can do, albeit with different efficiencies. This thesis is sometimes regarded as the foundation of computer sc ience pronsip. Computer science experts usually emphasize the von Neumann computers or Turing machines, because that was the most computers in use today. The computer scientists are also studying other types of machines, some of which are practical (such as parallel and quantum) and some are quite theoretical (such as random and oracle).

Upload: yuda-zeta

Post on 13-Jul-2015

33 views

Category:

Documents


0 download

TRANSCRIPT

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 1/11

Informatics and Computer Science

Informatics

Informatics is one branch of science that is quite broad because nearly covers and associated with a

variety of things about this life, where a collection of disciplines include science and informaticstechniques to specifically process data into information by optimally utilizing information technology

or computer.

In the Indonesian language, the term Informatics INFORMATIQUE derived from French, which in

German is called Informatik. Actually, the word is synonymous with the term computer science in

the United States and computing sciences in the UK.

In terms pendefenisian informatics, according to Philippe Dreyfus (1962) and l'Academie Francaise

(1967) that defines informatics as follows:

Disciplines collection (scientific discipline) and the Discipline of Engineering (Engineering discipline)are specifically related to the transformation / processing of the "Facts Symbolic" (data /

information), which mainly use the facilities automatic machines / computers.

In English the meaning is slightly different, ie more emphasis on aspects of information processing in

a systematic and rational.

If taken as a whole from an understanding of informatics at the top, logic and systematic approach is

a fairly dominant feature of Informatics, given the approach is key in terms of getting a solution in

solving various problems.

In principle this scholarship emphasizes how a data and information can be processed in such atechnology-assisted automated.

The technology terautomatisasi is not only in one machine, but can involve multiple machines. This

machine is more commonly called a computer. Computer as the main ingredient in this scientific

field has a very high, so simply peeling informatics ranging from how the machine works, how the

data is processed in a manner that is understandable by the machine while the information is well

understood by humans, to how the machine is able to communicate with other machines.

Computer Science

In general, Computer Science, better known by the name of Computer Science (Computer Science,CS) is the study of computing, both hardware (hardware) and software (the software). Computer

science covered topics related to computers, from abstract analysis of algorithms to more concrete

subjects like programming languages, software, and hardware. As a discipline, Computer Science is

different from computer programming, software engineering and computer engineering, even if the

third term is often misunderstood.

Church-Turing thesis states that all the computational tools that have been commonly known dalah

actually the same thing what they can do, albeit with different efficiencies. This thesis is sometimes

regarded as the foundation of computer science pronsip. Computer science experts usually

emphasize the von Neumann computers or Turing machines, because that was the most computers

in use today. The computer scientists are also studying other types of machines, some of which arepractical (such as parallel and quantum) and some are quite theoretical (such as random and oracle).

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 2/11

 

Computer Science studying what can be done by the program, and what is not (computing and

artificial intelligence), how the program should evaluate the results (algorithm), how programs

should store and retrieve specific bits of an information (data structures), and how the program and

users to communicate (the user interface and programming language).

Computer science is rooted in electronics, mathematics and linguistics. In the last three decades of 

the 20th century, computer science has become a new discipline and has developed his own

methods and terminology.

The first computer science department at Purdue University was founded in 1962. Almost all

universities now have departments of computer science. The highest award in computer science is

the Turing Award, the winner of this award are all pioneers in their fields.

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 3/11

Bhs indonesia

Informatika dan Ilmu Komputer

Informatika

Informatika merupakan salah satu cabang keilmuan yang cukup luas karena hampir

mencakup dan terkait dengan berbagai hal tentang kehidupan ini, dimana kumpulan disiplin

informatika ini meliputi sains maupun teknik yang secara spesifik mengolah data menjadi informasi

dengan memanfaatkan seoptimal mungkin teknologi informasi atau komputer.

Dalam bahasa Indonesia, istilah Informatika diturunkan dari bahasa Perancis informatique,

yang dalam bahasa Jerman disebut Informatik. Sebenarnya, kata ini identik dengan istilah computer

science di Amerika Serikat dan computing science di Inggris.

Dalam pendefenisian istilah informatika , menurut Philippe Dreyfus (1962) dan l’Academie Francaise(1967) yang mendefenisikan informatika tersebut sebagai berikut:

Kumpulan Disiplin Ilmu (scientific discipline) dan Disiplin Teknik (engineering discipline) yang secara

spesifik menyangkut transformasi/ pengolahan dari “Fakta Simbolik” (data/informasi), yang

terutama menggunakan fasilitas mesin-mesin otomatis/komputer.

Dalam bahasa Inggris memiliki makna yang sedikit berbeda, yaitu lebih menekankan pada aspekpengolahan informasi secara sistematis dan rasional.

Jika dilihat secara menyeluruh dari pengertian informatika di atas, pendekatan logika dan

sistematika merupakan ciri yang cukup dominan dari Informatika ini, mengingat pendekatan

tersebut merupakan kunci dalam hal mendapatkan solusi dalam menyelesaikan berbagai masalah.

Pada prinsipnya keilmuan ini lebih menekankan bagaimana suatu data dan informasi dapat diolah

sedemikian dengan berbantuan teknologi yang terotomatisasi .

Teknologi yang terautomatisasi tersebut tidak hanya dalam satu mesin , namun bisa melibatkan

beberapa mesin. Mesin ini lebih umum disebut dengan komputer. Komputer sebagai bahan utama

dalam bidang keilmuan ini memiliki peranan yang sangat tinggi, sehingga informatika secara

sederhana mengupas mulai dari bagaimana mesin tersebut bisa bekerja, bagaimana suatu data

diolah dengan cara yang dimengerti oleh mesin sedangkan informasinya dimengerti juga oleh

manusia, sampai bagaimana mesin tersebut mampu berkomunikasi dengan mesin lainnya.

Ilmu Komputer

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 4/11

Secara umum, Ilmu Komputer yang lebih dikenal dengan nama Komputer Sains (Computer

Science, CS) adalah ilmu yang mempelajari tentang komputasi , baik perangkat keras (hardware )

maupun perangkat lunak (software ). Ilmu komputer mencakup beragam topik berkaitan dengan

komputer, dari analisa abstrak algoritma sampai subyek yang lebih konkret seperti bahasa

pemrograman , perangkat lunak, dan perangkat keras. Sebagai suatu disiplin ilmu, Ilmu Komputer

berbeda dengan pemrograman komputer, rekayasa perangkat lunak dan teknik komputer, sekalipunketiga istilah tersebut sering disalahartikan.

Tesis Church-Turing menyatakan bahwa semua alat komputasi yang telah umum diketahui

sebenarnya sama dalah hal apa yang bisa mereka lakukan, sekalipun dengan efisiensi yang berbeda.

Tesis ini terkadang dianggap sebagai pronsip dasar dari ilmu komputer. Para ahli ilmu komputer

biasanya menekankan komputer von Neumann atau mesin Turing , karena seperti itulah kebanyakan

komputer digunakan sekarang ini. Para ahli ilmu komputer juga mempelajari jenis mesin yang lain,

beberapa diantaranya praktikal (seperti paralel dan kuantum) dan beberapa diantaranya cukup

teoritis (seperti random and oracle).

Ilmu Komputer mempelajari apa yang bisa dilakukan oleh program, dan apa yang tidak

(komputasi dan intelegensia buatan ), bagaimana program harus mengevaluasi suatu hasil

(algoritma), bagaimana program harus menyimpan dan mengambil bit tertentu dari suatu informasi

(struktur data ), dan bagaimana program dan pengguna berkomunikasi (antarmuka pengguna dan

bahasa pemrograman).

Ilmu komputer berakar dari elektronika , matematika dan linguistik . Dalam tiga dekade

terakhir dari abad 20, ilmu komputer telah menjadi suatu disiplin ilmu baru dan telah

mengembangkan metode dan istilah sendiri.

Departemen ilmu komputer pertama didirikan di Universitas Purdue pada tahun 1962.

Hampir semua universitas sekarang mempunyai departemen ilmu komputer. Penghargaan tertinggi

dalam ilmu komputer adalah Turing Award , pemenang penghargaan ini adalah semua pionir dibidangnya.

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 5/11

A. INTRODUCTION

A defect is a characteristic that reduces the usefulness or the price of an item; or some kind of 

weakness, imperfection, or deficiency [1]. Software defect is any defect or imperfection in the

product software (computer programs, planning, related documentation, or data) or the software

process (activities, and transformation methods are used to develop and manage softwareproducts). Software defect is a manifestation of human error (software manufacturer); after all, not

all human error is a defect, and not all defects are the result of human error. When found in

executable code, a defect is more commonly called a fault or bug. A fault is a step program, process,

or the wrong data in a computer program. Fault is a permanent defect in the software until the

software is executed.

Another term associated with software defects is sotware problem. Software problems are

something that encountered humans from software that causes the trouble, doubt, or uncertainty in

the use or inspection software. In a dynamic environment (operational), some of the problems /

issues may be caused by the failure. A software failure occurs during program execution. A failure

caused by a fault, in which defects are found in the executable code. In a static environment (non-operational), such as code inspections, a problem may be caused by the defect. Among the dynamic

and static environments, munbkin problems caused by misunderstandings, errors or the use of a

number of other factors not related to software products that are being used.

B. 10 THINGS ABOUT defects SOFTWARE

Here is a list of the top 10 things in software defect reduction [2] is adapted from "Industrial Metrics

Top 10 List" (B. Boehm, IEEE Software, Sept. 1987, pp. 84-85)

1.

Find and fix software problem after delivery is often 100 times more expensive than finding and

fix it during the requirements phase and design. As Boehm observed in 1987, "this knowledge has

become a major driver in the software industry focusing on design practice and a thorough needs

analysis, verification and validation from the beginning, as well as prototyping and simulation in the

early to prevent problems beruntut the cost." The use of the word "often" because the software

system smaller and closer to 5:1 noncrtical than 100:1.

2.

Software project currently spend their efforts around 40-50% on doing the job again (rework)

that actually work can be prevented. Reducing "rework that actually can be prevented" can

significantly improve the productivity software. A rework consists of effort spent to fix any softwareproblems that could have been discovered earlier. The implication, an attempt is also made up of 

"rework that can not be avoided", an observation that increases credibility by increasing realization

that generates user-interactive system that is better than the processes that emerge. In these

processes, the needs that arise from the prototyping and multistakeholder activity-shared-learning,

then bring to the design and coding practices. The process appears to indicate a change in the

definition of a system that makes it more cost-effective does not need to be classified as a defect

that can be prevented.

3.

Approximately 80% of rework (rework) is derived from the 20% defect. Value of 80% may be

lower for small systems and higher for very large systems. Two main sources of "rework that can beprevented" (avoidable rework) includes set design as well as the need and cost of development in a

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 6/11

hurry, which causes damage to code, design, and architecture major. A tracking system for reporting

software problems that records an attempt to repair any defects, helps you easily analyze data to

determine the main additional source for rework.

4.

Approximately 80% of the defects come from 20% of the modules, and about half of the moduleis free from defects. Studies from different environments in recent years have shown, with very

consistent, that between 60 to 90% 20% defect arising from the module, with a median around 80%.

5.

About 90% of downtime comes from, at most, 10% of the defect. However, some defects do not

really affect downtime and system reliability. For example, a historical analysis of software failure

from nine large software products IBM revealed that about 0.3% defect count for about 90% of 

downtime.

6.

Peer review found 60% of the defect. If you find and fix most of the defects in the projectdevelopment cycle is more cost effective than to find it later, we find a technique that can find

defects as early as possible. Various studies suggest that peer review provides an effective technique

that can find defects up to 31-93%, with a median of about 60%. Thus, 60% of the value of the so-

called in 1987 is a reasonable estimate. Factors affecting the percentage of defect to be discovered

is including the amount and type of peer review that is formed, the size and complexity of the

system, and the frequency of defects found by better execution, such as algorithms and concurrency

defects. The study points to the fact that peer review, analysis tools, and testing catch different

defect classes at different points in the development cycle. We need empirical research to help

choose the best combined strategy for investment-reduction defects.

7.

Perspective-based review found 35% more defects than nondirected reviews.

8.

Practicing personal discipline may reduce the level of initial defect up to 75%. Some personal

discipline process has been brought into practice. These include Harlan Mills's Cleanroom software

development process and Watts Humphrey's Personal Software Process (PSP). Data from the use of 

cleanroom at NASA have demonstrated 25-75% reduction in error rate during testing. The use of 

cleanroom also shows a reduction in rework effort (rework) so that only 5% of the repairs that take

more than an hour, where the standard process provides more than 60% of repairs require more

time. While the PSP berfous on the root causes of software defects individually, on the development

of personal checklist, as well as the repetition penceghan back, had significantly reduced levels of personal defect.

9.

If all things equal, any instruction sources will spend 50% more to develop high-dependability

software products than to develop low-dependability software products. After all more than worth

the investment if the project includes significant operating expenses and management fees.

10.

Approximately 40-50% user programs contain defects that are not trivial / nontrivial. A study in

this area found that 44% of 27 spreadsheet program produced by experienced spreadsheet

developers contain nontrivial defects / not trivial - the majority of errors in spreadsheet formulas (PSBrown and JD Gould, "An Experimental Study of People Creating Spreadsheets," ACM Trans. Office

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 7/11

of Info. Sys., July 1987, pp.258-272). Lab experiments further reported that the defect rate

spreadsheet was 35-90%. Analysis revealed a spreadsheet which runs between 21-26% defect rate;

low defect rate of correction may be made during operation.

C. EXAMPLE defects SOFTWARE

Here is an example of a software defect that has ever happened

1.

On a Windows XP computer, users can not hibernate if the computer has 1GB or more RAM, or

when the computer is running multiple processes that lead to high-stress conditions (Article ID 330

909). Microsoft has made a patch update to SP2 [3].

2.

Computers running Microsoft Windows XP there is a chance to stop me-respon/nge-hang when

the message "applying local settings" is displayed after login. This problem occurs when the filesrvsvc.dll cause access violation error. This error stops the svchost.exe process that load services

such as workstations and servers. Consequently winlogon.exe stops responding after you log on to

Windows (Article ID 823 830). Microsoft has made this problem patch update on SP2.

3.

On the website of social networking site Friendster (www.friendster.com) in 2008, found a lot of 

bugs and defects. In 2008, the site does not have a checking Friendster the perfect comment. Users

can insert javascript in the HTML-formatted comments. The result, a user can insert javascript that

can cause other users can not access the profile, even theft can happen the other user's web

cookies. Today, Friendster has improved the site that it is more secure.

4.

In Microsoft Office Excel 2007, if you do the calculations at 850 x 77.1 100 000 A1 will be shown

the results of which should result 65535. Of course this will result in calculation errors

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 8/11

A. PENDAHULUAN

Sebuah defect adalah suatu karakteristik yang mengurangi kegunaan atau harga suatu item;

atau semacam kelemahan, ketidaksempurnaan, atau kekurangan[1]. Software defect merupakan

segala cacat atau ketidaksempurnaan didalam produk software (program komputer, perencanaan,

dokumentasi terkait, atau data) atau proses software (aktivitas, metode dan transformasi yang

digunakan untuk mengembangkan dan mengelola produk software). Software defect merupakan

perwujudan dari kesalahan manusia (produsen software); biar bagaimanapun tidak semua kesalahan

manusia merupakan defect, dan tidak semua defect merupakan hasil kesalahan manusia. Ketika

ditemukan di dalam executable code, sebuah defect lebih sering disebut fault atau bug. Sebuah fault

adalah langkah program, proses, atau data yang salah di dalam program komputer. Fault merupakan

defect yang menetap di dalam software sampai software tersebut dieksekusi.

Istilah lain yang berhubungan dengan software defect adalah sotware problem. Softwareproblem adalah sesuatu yang ditemui manusia dari software yang menyebabkan kesulitan,

keraguan, atau ketidaktentuan dalam penggunaan atau pemeriksaan software. Dalam lingkungan

dinamik (operasional), beberapa problem/masalah mungkin disebabkan oleh failure. Suatu software

failure terjadi selama eksekusi program. Sebuah failure disebabkan oleh fault, yang mana defect

ditemukan dalam executable code. Dalam lingkungan statis (non-operasional), seperti inspeksi kode,

suatu problem mungkin disebabkan oleh defect. Diantara lingkungan dinamik dan statis, problem

munbkin disebabkan oleh kesalahpahaman, kesalahan penggunaaan atau sejumlah faktor lain yang

tidak berhubungan dengan produk software yang sedang digunakan.

B. 10 HAL TENTANG SOFTWARE DEFECT

Berikut ini daftar 10 hal teratas dalam software defect reduction [2] yang disesuaikan dari

“Industrial Metrics Top 10 List” (B. Boehm, IEEE Software, Sept. 1987, pp. 84-85)

1.

Menemukan dan memperbaiki software problem setelah delivery seringkali 100 kali lebih

mahal daripada menemukan dan memperbaikinya selama fase requirements and design.

Sebagaimana yang diobservasi Boehm di tahun 1987, “pengetahuan ini telah menjadi penggerak

utama di dalam memfokuskan praktek industri software pada desain dan analisis kebutuhan yang

teliti, verifikasi dan validasi sejak permulaan, serta prototyping dan simulasi di awal untuk mencegah

masalah beruntut yang menghabiskan biaya.” Penggunaan kata “seringkali” dikarenakan pada

sistem software yang kecil dan noncrtical lebih mendekati 5:1 daripada 100:1.

2.

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 9/11

Proyek software saat ini menghabiskan upaya mereka sekitar 40  – 50% pada

mengerjakan kembali pekerjaan (rework) yang sebetulnya pekerjaan tersebut dapat dicegah.

Mengurangi “pengerjaan ulang yang sebetulnya dapat dicegah” dapat secara signifikan

meningkatkan produktivitas software. Suatu pengerjaan ulang terdiri atas upaya yang dihabiskan

untuk memperbaiki berbagai kesulitan software yang bisa saja telah ditemukan lebih awal.

Implikasinya, suatu upaya juga terdiri atas "pengerjaan ulang yang tidak dapat dihindarkan", sebuahobservasi yang meningkatkan kredibilitas dengan meningkatkan realisasi yang menghasilkan sistem

user-interaktif yang lebih baik dari proses-proses yang muncul. Dalam proses-proses tersebut,

kebutuhan yang muncul dari prototyping dan aktivitas multistakeholder-shared-learning, kemudian

membawa ke praktek desain dan coding. Proses muncul yang mengindikasikan perubahan pada

definisi sistem yang membuatnya lebih cost-effective tidak perlu diklasifikaskan sebagai defect yang

dapat dicegah.

3.

Sekitar 80% dari pengerjaan ulang (rework) berasal dari 20% defect. Nilai 80% tersebutmungkin lebih rendah untuk sistem yang kecil dan lebih tinggi untuk sistem yang sangat besar. Dua

sumber utama dari "pengerjaan ulang yang dapat dicegah" (avoidable rework) meliputi menetapkan

kebutuhan dan biaya desain serta pengembangan dengan tergesa-gesa, yang mana menyebabkan

kerusakan code, desain, dan arsitektur utama. Suatu sistem tracking untuk laporan permasalahan

software yang mencatat upaya untuk perbaikan setiap defect, membantu anda menganalisa data

dengan mudah untuk menentukan sumber tambahan utama untuk rework.

4.

Sekitar 80% defect berasal dari 20% dari modul, dan sekitar separuh dari modul adalahbebas dari defect. Studi dari environment yang berbeda-beda pada beberapa tahun telah

menunjukkan, dengan sangat konsisten, bahwa antara 60 sampai 90% defect timbul dari 20% modul,

dengan median sekitar 80%.

5.

Sekitar 90% dari downtime berasal dari, paling banyak, 10% dari defect. Namun,

beberapa defect tak begitu mempengaruhi downtime dan reliability sistem. Sebagai contoh, sebuah

analisis sejarah software failure dari 9 produk software besar IBM mengungkapkan bahwa sekitar

0,3% defect dihitung untuk sekitar 90% downtime.

6.

Peer review menemukan 60% dari defect. Jika menemukan dan memperbaiki sebagian

besar defect di dalam siklus pengembangan proyek adalah lebih cost-effective daripada

menemukannya kemudian, kita mencari teknik yang dapat menemukan defect seawal mungkin.

Berbagai studi mengemukakan bahwa peer review menyediakan teknik efektif yang dapat

menemukan defect hingga 31-93%, dengan median sekitar 60%. Sehingga, 60% nilai yang disebut

pada tahun 1987 tersebut merupakan estimasi yang beralasan. Faktor yang mempengaruhipersentase defect untuk ditemukan adalah termasuk jumlah dan tipe peer review yang dibentuk,

ukuran dan kompleksitas sistem, dan frekuensi defect yang lebih bagus ditemukan oleh eksekusi,

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 10/11

seperti defect algoritma dan concurrency. Studi menunjukkan fakta bahwa peer review, tool analisis,

dan testing menangkap kelas defect yang berbeda pada poin yang berbeda di dalam siklus

pengembangan. Kita membutuhkan riset empiris untuk membantu memilih strategi gabungan

terbaik untuk investasi defect-reduction.

7.

Perspective-based review menemukan defect 35% lebih banyak daripada nondirected

review.

8.

Mempraktekkan disiplin personal dapat mengurangi tingkat awal defect hingga 75%.

Beberapa proses disiplin personal telah dibawa ke dalam praktik. Ini termasuk Harlan Mills’s

Cleanroom software development process dan Watts Humphrey’s Personal Software Process (PSP).

Data dari penggunaan Cleanroom pada NASA telah menunjukkan 25-75% pengurangan tingkat

kesalahan selama testing. Penggunaan Cleanroom juga memperlihatkan pengurangan upaya

pengerjaan kembali (rework) sehingga hanya 5% dari perbaikan yang memerlukan waktu lebih dari

sejam, dimana proses standar memberikan lebih dari 60% dari perbaikan memerlukan waktu lebih

lama. Sementara PSP berfous pada akar penyebab software defect individual, pada pengembangan

ceklis personal, serta pada penceghan pengulangan kembali, telah secara signifikan mengurangi

tingkat defect personal.

9.

Jika semua hal sama, setiap intruksi sumber akan menghabiskan 50% lebih banyak untuk

mengembangkan produk software high-dependability daripada untuk mengembangkan produk

software low-dependability. Bagaimanapun investasi lebih dari nilainya jika proyek meliputi biaya

operasi signifikan dan biaya pengelolaan.

10.

Sekitar 40 – 50% program user mengandung defect yang tidak sepele / nontrivial. Sebuah

studi pada area ini menemukan bahwa 44% dari 27 program spreadsheet yang diproduksi olehdeveloper spreadsheet berpengalaman mengandung defect nontrivial / tidak sepele — sebagian

besar error di dalam formula spreadsheet (P.S. Brown and J.D. Gould, “An Experimental Study of 

People Creating Spreadsheets,” ACM Trans. Office Info. Sys., July 1987, pp.258-272) . Eksperimen lab

selanjutnya melaporkan bahwa tingkat defect spreadsheet saat itu 35-90%. Analisa spreadsheet

yang berjalan mengungkapkan tingkat defect antara 21-26%; tingkat defect yang rendah mungkin

dari koreksi yang dibuat selama pengoperasian.

C. CONTOH SOFTWARE DEFECT

Berikut ini adalah contoh dari adanya software defect yang pernah terjadi

5/11/2018 tics and Computer Science - slidepdf.com

http://slidepdf.com/reader/full/tics-and-computer-science 11/11

 

1.

Pada komputer Windows XP ,user tidak bisa melakukan hibernate apabila komputermemiliki 1GB atau lebih RAM, atau ketika komputer menjalankan multiple proses yang

menyebabkan kondisi high-stres (Article ID 330909). Microsoft telah membuat update patch pada

SP2 [3].

2.

Komputer yang menjalankan Microsoft Windows XP ada kemungkinan berhenti me-

respon/nge-hang ketika pesan “Applying local settings” muncul setelah login. Masalah ini terjadi

ketika file srvsvc.dll menimbulkan error access violation. Error ini menghentikan proses svchost.exe

yang meload layanan seperti workstation dan server. Akibatnya winlogon.exe berhenti merespon

setelah anda log on ke windows (Article ID 823830). Microsoft telah membuat update patch masalah

ini pada SP2.

3.

Pada website situs jejaring sosial Friendster (www.friendster.com) di tahun 2008, banyak

sekali ditemukan bug dan defect. Pada tahun 2008, situs Friendster tidak memiliki pengecekan

comment yang sempurna. Pengguna bisa memasukkan javascript di dalam komentar yang berformat

HTML. Hasilnya, user dapat menyisipkan javascript yang bisa menyebabkan user lain tidak dapat

mengakses profilnya, bahkan bisa terjadi pencurian cookies web milik user lain. Saat ini, pihak

Friendster telah memperbaiki situsnya sehingga lebih aman.

4.

Pada Microsoft Office Excel 2007, jika anda melakukan perhitungan 850 x 77,1 pada A1

akan ditampilkan hasil 100000 dimana seharusnya hasilnya 65535. Tentu saja hal ini akan berakibat

pada kesalahan perhitungan