Data

1
Data
2
Where does data come from?

• Acquisition of information
• The acquisition of information is how the data is
  provided to the system.
• It can be done either
• In batch mode
• At source
• Electronically

3
Form filling

• Every time you fill in and return a form, you are
  preparing a batch data entry.
• Generally for batch data entry, form filling is
  used, where the information provider responds in
  a set way to a series of questions or prompts.
• The data is then entered by either key-to-disk or
  key-to-tape or some type of optical reader (e.g.
  bar-code reader (e.g. supermarkets, survey forms,
  etc), magnetic character recognizer (e.g. cheque
  validation) or voice recognizer (experimental
  only).

4
Preparing a form

• It is important in this situation to ensure that
  questionnaires are sufficiently unambiguous (e.g.
  multiple choice) and that the mechanism is in
  place for collecting the forms and entering them
  into the system.

5
Entry at source

• In a service situation, one of two things will
  happen
• An employee will act as go-between between the
  customer and the machine or
• The customer will enter data directly.

6
Designing an interface for an employee

• Used where a member of the public will answer
  questions orally and an employee or
  representative of the company running the system
  will enter the details at source (e.g. airline
  reservations, banking transactions, etc.).
• The employee must be well-versed in the needs of
  the system and must be able to interpret less
  than precise answers to a form acceptable to the
  system.
• The response time of the system is also a big
  factor, as is availability of input devices.

7
Designing an interface for a member of the public

• Where the customer will enter the data directly
  over the internet or workstation.
• The format of questioning must be extremely
  precise.
• Quite often localization software is required
  to enable the customer to interact with the
  system.
• E.g. Enter ZIP code.

8
Analog devices

• Information may be provided to the system in
  response to other processes occurring - e.g.
  telephone systems, signalling systems, etc.
• This type of data acquisition is used where data
  is logged or monitored for billing or emergency
  interruption purposes.
• In this situation, the system response time is
  highly critical. Every source will be an input
  device. Trapping of communication failure is
  important. (e.g. parity and protocol checks).
• Up-time of input devices should also be
  monitored.
• Modern-day devices are electronic.

9
Examples of modern-day devices?

• What electronic devices are in use every day?
• How does the data get in to them?
• Does it stay in them?
• Do you ever transfer data?

10
Data sampling

• In some circumstances, data sampling is used.
• This is a technique where an analysis of data is
  done, taking only a sample of the data (e.g.
  polls).
• Various different algorithms are available to
  produce what could be interpreted as
  representative sample data.
• The algorithm will depend on the context. This
  is used where the volume of data is too large or
  is coming in too fast to be able to use all of
  it.
• It is also used for monitoring remote devices to
  see if they are on-line - i.e. spot checks.

11
Presentation of information

• The way in which information is represented to
  the users of the system can vary considerably
• Screen / Print.
• Audio.
• Digital / Analog.

12
Screen/Print

• The most common output devices are screens and
  printers.
• Using these devices, information can be presented
  in report form, graph form, chart form or picture
  form.

13
Audio.

• Other output devices produce voice or sound.
  These can be
• voice synthesizers or
• digital to audio signals (as can be heard on a
  phone line using a modem).
• Streamed audio signals.

14
Digital/Analog.

• Signals can be sent out by a system to cause
• e.g. a traffic signal to change,
• a clown to nod and speak or
• a monitoring device to increase the heat in an
  air-conditioning system.
• Digital signals are now used for audio purposes,
  with digital music and streamed digital radio.

15
Storage of information

• Information can be stored
• Core memory
• Secondary memory
• On-line devices
• Off-line devices

16
Core memory

• Allows for short-term (i.e. program load time)
  storage of information while a process is
  running.
• A memory management unit manages which data is
  held in core memory at any one time.

17
On-line magnetic/optical devices

• Disks (optical or magnetic), memory sticks or
  memory units.
• Used to store large volumes of information which
  must be readily accessible.
• These must be backed up regularly, to a
  well-defined schedule.
• The backup should be stored both on another disk
  and on serial magnetic or optical storage.

18
Off-line devices

• Magnetic serial devices, such as tapes, can be
  used on-line, but the response time is very slow,
  due their serial nature.
• They are, however, perfect for batch transaction
  storage and for backup of on-line devices.
• Due to the decrease in price of magnetic disks,
  they are currently in widespread use for backups.
  E.g. rw dvds.
• What do you need to worry about in relation to
  backup devices?

19
Data security and protection from error.

• Physical security
• Software security
• Protection from hardware error.

20
Physical Security

• In order to ensure that data is secure, then all
  input, storage and output devices must be secure.
  
• This may mean that computer rooms need to be
  properly air-conditioned and heat-controlled
  tapes must be kept in a tape safe
• disks must be kept in an appropriate environment
• 1980s CDs hailed as indestructible damaged by
  their containers and rendered unusable!
• Remote devices must be kept free from tampering
  and out of danger from dust, smoke and other
  hazards (such as leaking roofs, etc).

21
Software security

• All sensitive information should be password
  protected, where the passwords are issued and
  controlled centrally and are monitored for
  possible break-in.
• Viruses or deliberate changes to data introduced
  by unauthorized users are undesirable and cannot
  be afforded.
• Firewalls must be in place to prevent harm.
• Protection from hardware error.
• This is done by using protocols and parity bits,
  backups and journalling.

22
Data Validation

• Edits on transaction data
• Proper range of values
• Presence of required fields
• Proper types of values, e.g numeric, etc
• Accuracy of self-checking field values e.g. check
  digits, etc
• Presence of related fields (e.g. cannot add a
  record if no key field is present).
• Control totals for batches
• Counts of records
• Check sums on nonsensical fields (e.g. sum all
  employee numbers in batch to match input total)