|
thunk
THX
TI
tier 1, tier 2, tier 3 ISP (Internet service provider)
TIFF
tiger team
time bomb
time out
time sharing
time to live (TTL)
Time-Division Multiple Access (TDMA)
Timex thunk
When a program needs to execute a subroutine, either its name can be explicitly stated in the program or the program can execute an expression in which the evaluation results in its name. In the latter case, the subroutine name is evaluated dynamically while the program is running. The technique for dynamically handling subroutines, called a thunk, was developed by P. Z. Ingerman.
Basically, the program evaluates the expression and figures out the name of the subroutine to execute and then stores that name in a specific location in memory. The program then looks in that specific location to grab the name of the subroutine. Ingerman picked the term "thunk" after realizing that some of the information about the expression's name had been figured out ahead of time by the compiler. In other words, it had "already been thought of" or "thunk."
Today, the term is used in reference to address changing. Older Intel 16-bit microprocessors use an addressing scheme called segmented memory. To access a particular memory location, a program must specify both the segment number and the location within that specific segment. Newer Intel 32-bit microprocessors, starting with the 80386, use a flat address space. In a flat address space, there are no segments, and each memory location has only a single memory address. This difference in addresses prevents older 16-bit programs from directly executing newer 32-bit code. Likewise, it prevents newer 32-bit programs from directly executing older 16-bit code. Because of the way Windows uses DLL (dynamic-link library) files, many 32-bit programs depend on dialog boxes and other features that Windows provides using 16-bit DLLs.
Windows has a thunking mechanism that lets 32-bit programs execute 16-bit code. This thunk is called a flat thunk. Windows also has a thunking mechanism that lets older 16-bit programs execute 32-bit code. This is called a universal thunk.
Although all versions of Windows since Windows 95 have supported flat addressing and 32-bit code, many of the 32-bit DLL files in Windows rely heavily on 16-bit code. Thus, thunking can be required even when there is no obvious usage of 16-bit code. Thunking does slow down the operation of a program.
The Macintosh operating system has always used a flat address space, so thunking is not required. Back to top
THX
There is a lot of confusion about THX. THX is not a sound format. It's not a special way of encoding sound. It's not an image format. Rather, THX is a Lucasfilm company and a set of quality specifications. The company THX certifies films, equipment, and computer software and hardware that meet certain given standards. Movies and equipment that carry the THX seal are not necessarily better than those without the seal; they've just been certified to meet the technical standards.
The concept was developed in 1982 while George Lucas was producing "Return of the Jedi." Lucas was interested in upgrading the quality of film presentation. The technical director of Lucasfilms identified many problems related to the way sound was played at movie theaters, noting that audiences were not hearing all the sounds recorded in the studio, specifically the more subtle sounds.
His approach to resolving this problem was to develop an innovative performance standard that included not only the sound equipment but also the acoustics of the theater itself. The name, THX, comes from Lucas' first feature film, "THX 1138."
In 1990, certification was introduced for home theater components. In 1993, a standard was introduced for the laser disc. (The laser disc was a 12-inch optical disc used for movies. The format was first used in the late 1970s and became obsolete in the 1990s with the development of the DVD.) In 1996, the laser disc program was expanded to include VHS and DVD, and the name was changed to THX Digital Mastering Program. In 1999, THX developed Surround EX in conjunction with Dolby Laboratories.
THX EX is the name for a new Dolby Digital sound format typically used for home theater. THX EX uses 6.1 channels to provide a better sound than the previous 5.1 channel setup. THX EX does not require additional equipment to listen to the audio as it is backward compatible with the current Dolby Digital standard. Back to top
TI
See Texas Instruments. Back to top
tier 1, tier 2, tier 3 ISP (Internet service provider)
The ISP field has three distinct tiers of firms that offer different services, target different customers, and have different network structures.
Tier 1 ISPs are large national or international ISPs. They are directly connected to the Internet backbone and can be considered part of the backbone itself. They have the highest speed connections and very reliable networks. Their customers are either lower-tiered ISPs or large companies that are looking for a very reliable and fast access to the Internet. A major advantage of purchasing service from a tier 1 ISP is if there is a problem with access, only one company is involved, so solving the problem is that much easier. Sprint is a tier 1 ISP.
Tier 2 ISPs purchase their Internet service from a tier 1 ISP. Tier 2 ISPs tend to cover a specific region. They focus on business customers and have lower quality networks and slower access than tier 1 ISPs.
Tier 3 ISPs also purchase their Internet service from tier 1 ISPs. Tier 3 ISPs tend to focus on the retail market, and they also tend to cover a specific region. Network quality and access speed are relatively low. Prices are much lower than for tier 2 or tier 1 ISPs. Back to top
TIFF
See Tagged Image File Format. Back to top
tiger team
The term "tiger team" was originally U.S. military jargon for any team whose purpose was to try and penetrate security measures or gain access to a secure location. Its purpose was to test the security measures in place to see if they were adequate. If the tiger team successfully penetrated a facility, it would leave a sign that read "bomb" or a note on top of a code book saying it had been stolen. No bombs were ever left and no secret material was ever really stolen. Some of these tiger teams included computer specialists who would try to break into military computers.
Some of the early commercial computer security measures that were taken included hiring trusted hackers to try and break into a computer network. The hacker would report how he was able to gain access so the network administrators could fix that problem. Later, companies and consultants put together teams of skilled individuals for this purpose. It was only natural that they would be called tiger teams.
Recently, the term has taken on the more generic meaning of any official inspection team. It can also be used to refer to any special problem-solving (fire fighting) group. Back to top
time bomb
Computer code that is designed to activate after a given period of time or after some event has occurred is called a time bomb. A time bomb is a subset of a type of malicious code called a logic bomb.
For example, an employee of Omega Engineering placed malicious code on its network. The code was timed to activate after the employee was terminated. When that happened in 1998, the code erased all of the company's research, development, and production programs. It also destroyed the tape backup. The company expected it to take years to recover.
Time bombs are particularly dangerous because insiders who have detailed knowledge of the systems and who have the clearance to have direct access to critical data and programs usually create them. This combination of knowledge and access lets the person write code that will do the most damage. The most common time bomb is designed to go off after an employee has been terminated, similar to the time bomb at Omega Engineering. The people who usually place mail bombs are people fearful of losing their jobs and want a way of extracting revenge if they are terminated.
The time bomb generally does not "know" that its author has been terminated, although this is possible. Rather, the perpetrator generally designs the time bomb to periodically check and see if some suppressing action has taken place. The perpetrator then periodically performs this suppressing action so that the time bomb does not go off. Once the perpetrator is terminated, he cannot perform the necessary suppressing action. The next time the time bomb runs, it sees that the suppressing action has not been performed, so the time bomb performs the nefarious actions it was programmed to perform. The actions taken by the time bomb are limited only by the knowledge and access of the perpetrator.
Creating a time bomb requires specialized programming knowledge, as well as knowledge of the system to be bombed. Unlike a virus or worm, this specialized knowledge makes time bombs very difficult to spot before they blow up. As a result, the time bomb is one of the computer crimes with the greatest potential for damage. In addition to blowing up when the perpetrator is terminated, time bombs can be timed to go off when they will cause the most damage to the system or the company.
Not all time bombs are nefarious. It is not uncommon for shareware programs to stop running after the trial period has expired. In addition, leased software will generally stop running after the lease has expired unless an extension is purchased. Back to top
Time-Division Multiple Access (TDMA) TDMAcombines three conversations into one channel, giving the system three times the capacity of a system not using TDMA. | Time-Division Multiple Access (TDMA)
TDMA is one of the three main methods used by digital cellular phone networks for transmitting telephone calls. The other two are FDMA (Frequency-Division Multiple Access) and CDMA (Code-Division Multiple Access).
To understand TDMA, it is first necessary to understand the operation of an analog cellular phone network. In 1983, the FCC (Federal Communications Commission) approved an analog cell phone standard called AMPS (advanced mobile phone service). To encourage competition, the FCC requires two carriers in each market. Each carrier was assigned 790 frequencies for voice and 42 for data. It takes a pair of frequencies for a phone call, one for transmitting and one for receiving. This pair of frequencies is called a channel. So, each carrier was assigned 395 voice channels and 21 data channels. (The data channels are used for housekeeping activities, not for carrying computer data.)
In a cellular system, each service area is divided into many small cells. Each cell uses about one-seventh of the available channels, so each cell has about 56 voice channels available. That is, only 56 simultaneous calls can be made within each cell. Nonadjacent cells may use the same channel.
Digital cell phones convert the analog phone call into digital before transmission. This enables greater compression and greatly increases the capacity of each cell. TDMA takes each channel and slices it into three time slots. Each phone conversation gets the radio signal for one-third of the time, with the system rapidly switching between phones. This is referred to as time-division multiplexing. Because the digital signal is highly compressed, the switching between three different phone conversations is accomplished with no loss of information. The net result is that the system has three times the capacity of an analog system using the same number of channels without TDMA. A cell using TDMA can handle roughly 168 simultaneous calls.
TDMA is also used in the GSM (Global System for Mobile Communications) that is the basis for PCS (personal communications services). With PCS, the channel is sliced into eight time divisions rather than three.
In addition to increasing the capacity of the system, TDMA offers several other advantages over standard cellular technology:
•It is easily adaptable to carrying data, as well as voice.
•Because users are separated by time, they do not experience interference from other simultaneous transmissions.
•Because the phone is only transmitting one-third of the time, battery life is greatly extended.
•The tower equipment is smaller and easier to maintain.
TDMA has two main disadvantages. First, the use of a predefined time slot makes it harder to hand off the call to the next cell, increasing the chances of a call being disconnected as it moves between cells. The second disadvantage is that TDMA is subject to multipath distortion, which is caused when snippets of conversation bounce around buildings and other obstructions in such a manner that they arrive at the phone out of sequence. Back to top
Ttime out When Windows tries to print but cannot find a printer, it waits and then issues an error message, similar to the one shown here, after it times out. | time out
When a computer must interact with another device, it usually will not just assume that the interaction has taken place. For example, printing requires that the computer interact with the printer. The computer could just send the information to the printer and assume it was being printed, but this would waste a lot of computational time if the printer were not working. Additionally, the user might not notice the problem with the printer.
When the computer is directly connected to the device with which it is interacting, such as a local printer or a scanner, the two devices can easily talk with one another so the computer knows the status of the device. This is how your computer knows when your local printer needs more paper or your scanner finishes its scan.
When the computer is connected to a remote device, such as a network printer or Web site, the communication is more problematic. If the network goes down, the computer may not receive its signals from the remote device. The remote device needs a method to alert the computer of a problem. That method is a time out.
A time out is a period of time after which a computer assumes an error condition if some event has not occurred. For example, when you establish a dial-up connection to the Internet, your ISP (Internet service provider) starts a timer when that connection goes idle. After a preset period of time, say five minutes, the ISP disconnects you. Your connection is said to have timed out. Likewise, in printing to a network printer, the computer will assume there is a problem if the printer does not confirm the receipt of the job within a specific period of time.
Time outs are common for Web sites that require you to log in, for communications in general, and with printers. For each situation, there is a default length of time before a time out is initiated. Sometimes, either the user or the network administrator can change the default length, but other times it is fixed. Back to top
time sharing
Time sharing is a technique that lets more than one user simultaneously interact with a powerful computer. The technique was originally developed as a convenience for program debugging, but it rapidly gained importance as a way for an entire company to share a single mainframe computer.
Time sharing was developed in the late 1950s and early 1960s, just a few years after the development of reliable commercial mainframe computers, which were very expensive and required a large amount of space, as well as a dedicated maintenance and operating staff. They were so expensive to run that the primary concern in operating them was that they be used as efficiently as possible. This led to the development of batch processing.
With batch processing, the program and its data were prerecorded onto magnetic tape or punch cards. Once everything was ready and time was available on the mainframe computer, the job began. Once started, the job had the full attention of the computer, and it ran uninterrupted until the job was finished or until a bug in the program caused the computer to stop processing the job.
If a bug developed, the programmer would have to fix the program and then resubmit the job. Once resubmitted, the job would have to wait in the queue until time again became available on the mainframe computer. If the programmer encountered another bug, the process of debugging, resubmitting, and waiting for computer time repeated. This process continued until either the program ran successfully or the programmer gave up. This was particularly frustrating to users at universities and research institutes where the computers were usually saturated with jobs and the waiting time to rerun a program could be very long. As programs grew larger and larger, the number of errors encountered grew even faster and the time required for debugging increased.
Time sharing was originally designed to speed the program debugging process by letting a program share the computer resources while being debugged so that each attempt at debugging did not put it once again at the end of the queue. By not having to wait between debugging cycles, programmers could debug their programs much faster.
On April 7, 1964, IBM introduced the System/360. This was the first OS (operating system) to include time-sharing features. Time sharing has been central to mainframe OSes ever since.
Conceptually, time sharing is straightforward and relatively simple. The mainframe computer works on one job for a period of time, then puts it aside to work on a second job, and so on. Once it reaches the last job, it starts again with the first job. The brief burst of time that each job receives during the round-robin rotation is called a quantum or time slice.
Time sharing requires one software and three hardware features. The software required is a controlling program, called a supervisor, that manages the switching between programs, as well as all of the resources needed to support those programs.
The first hardware feature needed is an internal timer that the supervisor uses to interrupt programs at the end of their time slice. Time sharing also requires a privileged operating mode for the supervisor. This super-status is not available to the individual programs. It lets the supervisor control the hardware and file services and shut down programs that misbehave. The remaining required hardware feature is a way of controlling memory to make sure that none of the programs try to access memory not allocated to that program. These hardware features combine to make sure that the supervisor never loses control of the computer, no matter how badly the individual programs misbehave.
Central to the concept of time sharing is the sharing of the computer by multiple users. Thus, while PCs can multitask, rapidly switching between multiple programs, they generally are not considered time-sharing devices.
Time sharing does not increase the amount of computer power available. Therefore, a well-behaved program without bugs will take longer to run in a time-sharing environment than in a batch environment because it must share the computer with other programs. Additionally, the overhead associated with the supervisory program and the switching between programs reduces the overall computer power that is ultimately available to run user jobs.
Because time sharing was implemented on expensive mainframe computers, it was not uncommon for users to be billed for the time their programs ran on the mainframe. In a corporate environment, it was not uncommon for departments to be charged for part of the computer department budget based on their usage of the mainframe. Thus, an important part of the supervisory program's responsibilities was to track the time used by each program and charge it to the appropriate user or department. Ultimately, these charges would become one of the main justifications that departments would use to bring PCs into the corporate environment. Back to top
time to live (TTL)
On the Internet, data is sent out in IP (Internet Protocol) packets. Not all of these packets reach their destination in a reasonable length of time. For example, the destination could be improperly specified, causing a packet to loop endlessly. A method was needed to make sure that old packets were removed from the network. TTL provides that method.
TTL is a value in the IP packet that tells a network router if it should pass a packet on or discard it. The basic idea is that a packet is given a certain period of time to reach its destination, or the router discards it.
As implemented, the TTL is an 8-bit header field in every IP packet. Its value is usually around 30. Each time a packet passes through a router (each pass is called a hop) at least one is subtracted from this value. Thus, the TTL value functions as the number of router hops a packet is allowed to make. When the count reaches zero, the router detecting this condition discards the packet under the assumption that it must be undeliverable because it has not been delivered by now. In addition to discarding the packet, the router sends an ICMP (Internet Control Message Protocol) message back to the sender so the sender can decide how to handle the situation.
A default TTL setting of 30 has worked for years, but the Internet is growing so rapidly that it is not uncommon for it to take 30 or more hops to reach a destination. The default value in Windows 98 is 32 hops. You can raise this default if you often have problems reaching certain sites. The maximum possible value is 255. Although the user can increase the TTL setting on her end, she is still limited by the TTL setting her destination is using. If the destination uses too low of a TTL setting, the acknowledgements and information sent back to the user will never be received because they will be discarded en route.
The TTL field is used in the Traceroute utility (software that lets you follow the path of a data packet on the Internet) to trace the route to a host and to time connections. By setting the TTL to one, the first router will discard the packet and notify the user. Increasing the TTL to two will identify the second router, and so on. In this fashion, the software can trace the entire route. By timing the period between sending and receiving these packets, the software computes network response time along the route. Back to top
Timex Watches with Timex's Data Link
technology interface with a computer. | Timex
Timex is a privately held company with 7,500 employees and sales of more than $850 million in 1998. When you think of Timex, you think of its watches. However, Timex has sold a lot of computers. Although they are no longer in the computer industry, it still markets the Data Link watch, which interfaces with a computer.
Although Commodore contacted Timex regarding a merger, Timex decided to market a computer created by British inventor Clive Sinclair. In 1980, Sinclair founded Sinclair Research. The company soon released a computer called the ZX80. For about $100, the purchaser received a case, circuit board, and lots of individual parts, which he then had to use to build the computer. The ZX80 had 1KB of RAM, 4KB of ROM, and weighed 12 ounces. In March 1981, a ZX81 kit was released with 2KB of RAM.
In July 1982, Timex began selling the TS 1000 (Timex Sinclair 1000), a preassembled version of the ZX81. Like the Sinclair ZX81 on which it was based, the TS 1000 was very limited. The only software it came with was BASIC (Beginner's All-Purpose Instruction Language) so, out of the box, it could do nothing but let the user write programs. Rather than a monitor, it used a television. It could only display text in black-and-white. The only way to display graphics was with block characters. The TS 1000 did not have a floppy diskette drive or hard drive. For storage, you purchased a cassette recorder and connected it to the computer. Because a cassette recorder is an analog device and the TS 1000 was storing data and programs digitally, the success rate for writing to and reading from tapes was low.
Even though the TS 1000 was in every way inferior to its Commodore competition, Timex quickly sold 500,000 units. This was more that Tandy, Apple, and Commodore combined. However, Commodore retaliated by engaging Timex in a price war, which prevented Timex from money off the computer. Another problem that hurt profits was poor quality. It is estimated that as many as one out of three TS 1000s were defective. In August 1983, Timex released the TS 1500, an update to the TS 1000 with a new case and 16KB of The TS 1500 was a failure. The public was no longer willing to accept its limitations.
In April 1982, Sinclair Research released the ZX Spectrum in Britain. The ZX Spectrum supported color and had up to 48KB RAM, better graphics, and sound. It would be the basis for the remaining Timex computers. Timex produced two ZX Spectrum-based prototypes (the TS 2000 and TS 2048) that were never marketed.
Timex's fifth, and last, computer was the TS 2068, released in November 1983. The TS 2068 was also based on the Sinclair ZX Spectrum. However, it deviated in several important ways, including offering more memory, a cartridge slot, and a new operating system called the TOS (Timex Operating System). The TS 2068 was not a big success, and by the spring of 1984, Timex was out of the computer business—almost.
In 1995, Timex introduced its Data Link watch with a Windows program for data entry. The user transfers computer data to the watch by holding the watch up to the screen. The software flashes bars of light on the monitor to represent the data and the watch reads the flashing bars of light to store the data. In 1997, Timex introduced an adapter so the Data Link watch could connect to a notebook computer. The Data Link watch is still available from Timex.
Currently, Timex is the worldwide leader in watches. It is active in more than 90 countries worldwide. Back to top
|
Email This
Print This
