Going From “Hello!” to ? to “Merhaba!”
The purpose of translation is to enable speakers of a language to understand another language of communication. This may or may not entail that the techniques, rather hermeneutic methods, they are using to understand their native tongue also be translated. If a given method does not seem to be compatible with language B as much as it seems to be with language A, then the meaning of some exemplary phrase must first be assimilated from its syntactic depiction in A and then studied together with its syntactic depiction in B.
When a method (of interpretation) needs to be compatible with two languages, irrespective of whether they have the same typology or not, there are a number of issues that arise. In order to simply the discussion, the following nomenclature will be in use:
- Processes – step-by-step evaluations used to extract meaning from syntax
- Machine – summation of the human mind and our linguistic faculties
- Operation – an executed action
All processes are worked on by machines, which also handle the operations performed on or by the process itself. When a process needs to be ported, what needs to be checked is whether or not the machine supports the target language (defined by the ability to understand a specific typology).
When the machine does not support the target language, the process will have to be converted to a compatible language and then ported. If the machine supports the target language, then the process is ported first and then converted to the required language.
The bigger issue, which necessitates the creation of interoperability tools, is machine loading (defined by ease of use). Processes are executed in machines in a serial manner. Even in parallel processes, when there are multiple threads running, each thread will have a series of processes that have to be executed one after another. When a process queue is long, the machine becomes overloaded. If the target machine is overloaded, the process cannot be ported unless and until it is made concise.
The biggest problems faced in language conversion are that of syntactic and semantic preservation.
Syntactic preservation refers to the preservation of a syntax structure. However, since a conversion from one language to another language is what is being discussed, syntactic preservation is not a necessity.
As for semantic preservation, it stands for the maintenance of the meaning of each line of text. This means that, if a method is conceived to do something with one particular language, a language-conversion tool must allow the conception to do the same thing with the other language. If the semantics of a phrase are not preserved, then the same words will stand for different encapsulations of meaning in two different languages.
In order to maintain the semantics, it is necessary that every piece is broken down to smaller and smaller parts until only nuclear components remain. Next, these components need to be stored in a suitable format that, essentially, presents a way to recall meaning independently of syntax (often misinterpreted as “the language we think in”) – or at least independent of the two languages under study.
Process control is the science of controlling the way a process behaves or functions. Processes are created within machines in order to perform specific tasks over a period of time. There may be multiple processes operating within the same machine – doing different things. Also, the tasks that each process performs may be all grouped together to achieve a greater goal, or they may be required by other processes to function.
There are three types of process control techniques.
- Discrete – In discrete process control, a process is used to perform extremely specific tasks over a short span of time. A very good example is that of pausing or stopping when one encounters a comma or a period.
- Batch – In batch process control, processes are designed to perform a combination of tasks over an elaborate span of time, but otherwise being repetitive. This is often encountered in mathematical problem solving and logical reasoning.
- Continuous – In continuous process control, the process is regulated by different variables whose values changes continuously with time. Such changes may or may not be mathematically regulated, or they may or may not have pattern. Due to a certain degree of unpredictability, continuous process control is a more complex affair than discrete or batch process controls. A practical example would be that of reading a piece of text and continuously finding meaning while handling variables such as context.
For example, the negative feedback system of hormones in the human body is shown below.
If an algorithm were to be written for this process, how would it be controlled? How would the programmer know at which point to make a required change if the algorithm goes wrong? In order to know such things, the process must be evaluated in a series of steps. Each step must place a weight on the operations being executed at that time. As the process continues, it gains weight incrementally. Wherever the increment is more than normal, or more than expected, the algorithm needs to be corrected there. This is an apt example to also explain the concept of recursion.
When concepts like recursion are being translated, care must be taken to avoid a semantic infinite loop or a semantically null point in the target language owing to a failure to maintain the syntax or the semantics. In such scenarios, the implications of the process control techniques must also be ported as part of the syntax.
The goal of a process control system is to reduce the frequency of occurrence of errors as well as prevent information redundancy.
The most prominent way to go about this is to adopt a method called statistical process control (SPC). In SPC, data is collected about a particular process while it is simulated (through hermeneutic correlations). The construction of sentences must be practiced for hundreds or thousands of times until various statistical averages concerning its performance can be obtained. Then, using such averages, the future performance of the process can be predicted to an acceptable degree (compatibility evaluation). SPC also allows the speaker to keep track of changes in the process center while simultaneously monitoring the variations about that center (cognition).
- Productivity gains are permanent, Performance losses are temporary (frankmccabe.wordpress.com)