1
00:00:00,140 --> 00:00:02,930
To cover the intrinsic problem of eliciting requirements,

2
00:00:02,930 --> 00:00:05,670
many different techniques have been proposed. So here

3
00:00:05,670 --> 00:00:08,260
I list some of most traditional techniques for

4
00:00:08,260 --> 00:00:11,290
requirement elicitation and as I present those, please keep

5
00:00:11,290 --> 00:00:13,140
in mind that these techniques can be used

6
00:00:13,140 --> 00:00:17,230
separately or combined. A first technique is called background

7
00:00:17,230 --> 00:00:20,740
reading. And, this technique involves collecting information by

8
00:00:20,740 --> 00:00:24,840
reading existing documents such as company reports, organizational charts,

9
00:00:24,840 --> 00:00:29,280
policy manuals, job descriptions, documentation of existing systems and so

10
00:00:29,280 --> 00:00:32,400
on. And, this technique is especially appropriate when one Is

11
00:00:32,400 --> 00:00:35,370
not familiar with your organization for which the requirements are

12
00:00:35,370 --> 00:00:38,950
being collected. So you want to get some background before interviewing

13
00:00:38,950 --> 00:00:41,240
actual people. And one of the main imitations of these

14
00:00:41,240 --> 00:00:43,930
kinds of approaches is that written documents might be out

15
00:00:43,930 --> 00:00:46,320
of sync and they often are out of sync with

16
00:00:46,320 --> 00:00:49,970
reality. Tend to be long winded. It may contain many irrelevant

17
00:00:49,970 --> 00:00:51,850
details, so you may have to look at a lot

18
00:00:51,850 --> 00:00:54,810
of materials to extract enough information. The hard data and

19
00:00:54,810 --> 00:00:58,650
samples techniques consist in deciding which hard data we want

20
00:00:58,650 --> 00:01:01,680
to collect and choosing the sample of the population for which

21
00:01:01,680 --> 00:01:04,750
to collect such data and hard data includes facts and

22
00:01:04,750 --> 00:01:09,790
figures such as forms, invoices, financial information, survey results, marketing

23
00:01:09,790 --> 00:01:12,300
data, and so on. And the sampling of this data

24
00:01:12,300 --> 00:01:15,170
can be done in different ways. For example, the typical ways

25
00:01:15,170 --> 00:01:19,200
to do random selection. Interviews are another typical approach for

26
00:01:19,200 --> 00:01:21,870
requirement solicitation, and this is the approach that we use for

27
00:01:21,870 --> 00:01:24,910
the first project in this course, for instance. Interviews can be

28
00:01:24,910 --> 00:01:27,720
structured in which case there is an agenda of fairly open

29
00:01:27,720 --> 00:01:30,450
questions or they can be open ended in which case there

30
00:01:30,450 --> 00:01:32,770
is no preset agenda and the interview is more of a

31
00:01:32,770 --> 00:01:36,500
conversation. On the positive side, interviews can collect a rich set

32
00:01:36,500 --> 00:01:40,260
of information because they allow for uncovering opinions as well as

33
00:01:40,260 --> 00:01:43,230
hard facts. Moreover, they can probe in depth through follow

34
00:01:43,230 --> 00:01:46,790
up questions. On the more negative side, interviewing requires special

35
00:01:46,790 --> 00:01:50,400
skills that are difficult to master and require experience. And

36
00:01:50,400 --> 00:01:52,890
it is not enough to collect a lot of information. If

37
00:01:52,890 --> 00:01:55,520
this information is hard to analyze or even irrelevant, it

38
00:01:55,520 --> 00:01:58,250
might become useless. So you need to know how to conduct

39
00:01:58,250 --> 00:02:01,180
an interview in order to take advantage of these techniques.

40
00:02:01,180 --> 00:02:05,440
Surveys can also be extremely useful for gathering new requirements because

41
00:02:05,440 --> 00:02:08,550
they can quickly collect information from a large number

42
00:02:08,550 --> 00:02:11,520
of people. Moreover, they can be administered remotely. For

43
00:02:11,520 --> 00:02:13,610
example, by email, through the web. On the other

44
00:02:13,610 --> 00:02:16,750
hand, surveys tend to severely constrain the information that

45
00:02:16,750 --> 00:02:19,430
the user can provide and might miss opportunities to

46
00:02:19,430 --> 00:02:24,460
collect unforeseen, relevant information. Finally, meetings are generally used

47
00:02:24,460 --> 00:02:27,690
for summarization of findings and collection of feedback, so as

48
00:02:27,690 --> 00:02:30,500
to confirm or refute what has been learned. So the

49
00:02:30,500 --> 00:02:32,660
only additional thing I want to mention about meetings

50
00:02:32,660 --> 00:02:34,920
is the fact that it is fundamental that have clearly

51
00:02:34,920 --> 00:02:37,970
stated objectives and are planned carefully. This is something that

52
00:02:37,970 --> 00:02:40,730
should be quite obvious, but doesn't always happen in practice.