Dear Friends,
Researchers of computer science and software engineering committed huge mistake, such mistake was not committed by researchers of any other scientific or engineering field for past 2000 years. The mistake is: Evolving a complex paradigm of computer science and software engineering for nearly 45 years that resulted in a huge body of knowledge, which is rooted in flawed 48 years old hidden beliefs. Only other example I could find is geocentric paradigm that evolved for 1600 years (rooted in flawed belief “the Earth is static”) and was replaced by heliocentric model nearly 400 years ago.
It is a monumental mistake to have undocumented or unproven beliefs (or assumptions) at the root (i.e. core) of any body of knowledge of any scientific or engineering discipline. The scientific rules (i.e. processes and principles) were created during the first greatest scientific revolution (i.e. when error at the root of geocentric paradigm was exposed by enduring huge pain and suffering) and the rules has been refined over the centuries to avoid repeating of such kind of shocking mistake again at any cost. Of course, researchers may rely on documented beliefs (or assumptions) for proposing hypothesis to investigate or explore scientific or technological frontiers to discover new facts to expand the body of knowledge, but it is a blatant violation of the scientific process to not documenting any such belief as an assumption.
Likewise, even a fact (i.e. widely accepted to be self-evident) must be treated as an assumption, until it is proven by documented proof backed by evidence. The proof must be clearly documented and publicly available, so that the proof can be falsified (e.g. any time in the future as technology advances over time), if and when new demonstrable anomaly or contradicting evidence is surfaces or discovered. We all know what happened when mankind relied on a belief “the Earth is static” by considering it to be a self-evident fact. Researchers of no scientific or engineering discipline can afford to repeat such monumental mistake again and again.
Unfortunately, few experts confuse between these two completely different kind of beliefs or assumptions (1) having (or relying) on documented beliefs or assumptions in (or by) a hypothesis for exploring un-chartered frontiers for expanding the boundaries of body of knowledge, and (2) having and relying on undocumented or unproven beliefs at the very core or root of a mature body of knowledge, which has been evolving for decades or even centuries (e.g. by considering the unproven beliefs to be self-evident facts). It is foolish to justify the second kind of beliefs by referring to the first kind of beliefs.
Relying on documented beliefs for exploring new un-chartered frontiers is not a violation – it is a proven method for research. But having or relying on undocumented or unproven beliefs at the very core/root and as a foundation of a mature body of knowledge makes the whole body of knowledge flawed and invalid. For example, the whole geocentric paradigm that evolved for 1800 years ended up invalid. Likewise, existing paradigm for CBSD that has been evolving for decades end up useless for achieving real CBD for software products, if the beliefs at its root or core are flawed.
The basic rules (e.g. processes and principles) for any kind of research to advance any scientific or engineering discipline are devised to prevent this kind of mistake at any cost. The researchers of computer science and software engineering violated even such basic common sense rules by relying on undocumented and unproven (or untested) beliefs, when defining nature and properties for so called software components for CBSD (Component Based Design for Software products).
The definitions for so called software components or many concepts or aspects of existing CBSD has no basis in reality. But in fact those definitions and concepts are in clear contradiction to the facts and reality we know about the physical components that are capable of achieving the CBD of physical products. For example, almost every expert readily admits the facts that the nature and properties of so called software components are in clear contradiction to the reality. If the beliefs (that lead to such contradictory definitions or concepts) are documented, what are those beliefs? If those 45 years old beliefs are facts, where can I find the proof (e.g. to investigate if the proof can be falsified using latest 21st century knowledge and tools)?
There are many unproven and undocumented beliefs at the root of software engineering, such as: (i) Many software experts give baseless excuses such as software is different or unique, or (ii) believe that it is impossible to invent real-software-components for achieving real CBD for software, which is equivalent to the CBD of physical products. Is there any proof that it is impossible to achieve such real CBSD? It is impossible to find, if any one ever even tried to verify such baseless beliefs.
I can demonstrate proof that it is possible to achieve such real CBD for software applications or products. In light of this kind of evidence and examples, it is impossible to find a valid reason, why software products can’t be designed as engineers design and build large one-of-a-kind physical products such as an experimental spacecraft or prototype of a next generation fighter-jet (completely free from spaghetti design/code).
Best Regards,
Raju
Dear Friends,
What I realized must be shocking to any real scientist. I got to share this: What is the only difference between the hard sciences and the soft sciences? The hard sciences deal with objective facts and reality such as nature and properties of physical things or demonstrable facts or concepts about physical phenomena. The soft sciences deal with subjective things and trying hard to see, if they can find a way to measure and/or quantify the subjective things, for example to improve our ability to predict the subjective nature or phenomena.
'Measure what can be measured and make measurable what cannot be measured.' … Galileo Galilee
The hard sciences deal with objective things and reality that are measurable, repeatable, demonstrable and predictable. The soft sciences deal with subjective things and unpredictable realities (e.g. influences of religions, culture, charismatic leaders, technological progress and human instincts such as greed, fear, love, hate, or perceptions etc., on economics, societies, regions or politics etc.), which are hard to measure, predict or repeatable. But goal is to employ proven scientific methods, principles and processes to acquire knowledge such as to gain insights and understanding, for example, improving the probability of predictions (increase chances or degree of repeatability) or improving the approximation of measurements.
The fact is: soft sciences call themselves sciences, because their goal is to employ proven scientific principles and processes to gain knowledge and insights about the subjective things and unpredictable phenomena or reality (e.g. complex influences of religions, cultures and human instincts such as greed, fear, love, hate, or perceptions etc., on economics, societies, regions or politics etc.). https://en.wikipedia.org/wiki/Hard_and_soft_science
One thing in common for hard sciences and soft sciences is: Both are trying to gain insights and knowledge of nature of things and reality by employing proven basic scientific principles and processes. The difference is: hard sciences deal with objective things and phenomena, while soft sciences have to deal with subjective things and unpredictable phenomena or reality (e.g. complex influences and interactions of religion, culture and instincts such as greed, fear, love, hate, or perceptions etc., on economics, societies, regions or politics etc.).
On the other hand, computer science not only ignored reality and facts but also blatantly violated basic scientific principles and processes when dealing with quintessential objective things such as nature and properties of components for CBD or neurons for AI. Isn’t it insult to soft sciences, if any one insists that computer science is a soft science, because soft sciences neither ignore reality and facts nor blatantly violate basic scientific principles and processes.
If untested and undocumented beliefs were to be detected at the very core/root of the body of knowledge of any hard science, it would be shocking and scrabble for answers to find out how could such error slipped through and could stay unexposed for decades. Even soft sciences must follow scientific processes and principles. It is a violation of the scientific process to rely on undocumented or untested belief (opinion or assumption). The soft sciences must document each of the beliefs (i.e. well informed opinions or educated assumptions) only after testing each educated belief (or well informed opinion) by making sure that the well informed belief doesn’t contradict known reality or obvious facts.
Today it is an obvious fact: the existing definitions for so called software components and concepts for so called CBSD paradigm are in clear contradiction to the reality and facts we know about the physical components and concepts of CBD (Component Based Design) of physical products. How could computer science be a real science, if it has such flawed beliefs at the very core and has been evolving for nearly 45 years on such foundation?
Any research effort for advancing engineering or technology must have sound scientific foundation (i.e. a body of knowledge). The research effort would be wasted and no meaningful technological progress is possible, if there are flawed beliefs at the core (or root) of the scientific foundation (e.g. if the body of knowledge has been evolved or created by relying on hidden untested beliefs).
P.S: Isn't it Proof That Computer Science is Not Even Soft Science?
Best Regards,
Raju
There is a good paper that deals with some of the ideas you expose,
Hamlet, D. Science, Mathematics, Computer Science, Software Engineering. The Computer Journal, 2012, 55, 99-110.
Kinds,
Dear Dr. Gomez,
Thank you for referring to an interesting paper. I read abstract and registered for oxfordjournals.org to access full paper. But I got an email saying that it may take up to 36 hours to activate my login.
Based on the abstract, this paper looks in the perspective of existing relation between computer science and mathematics. But I am looking in more pragmatic perspective. What are our objective and goals?
The objective of researchers of any scientific discipline is to acquire new useful knowledge and insights or wisdom. The objective of engineering researchers is to invent useful solutions and processes to address problems or issues (by relying on the existing body of knowledge). If existing body of knowledge cannot address an engineering problem or issues, it is essential to acquire necessary knowledge.
Obviously researchers of computer science are using mathematics as a tool to accumulate body of knowledge. This knowledge is enough to address many problems and issues. However, researchers of computer science are not using science as a tool to accumulate body of knowledge. Hence fail to address many problems such as software crisis or machine intelligence (i.e. real Artificial Intelligence).
My contention is software must use every available tool to gain necessary knowledge for addressing such huge outstanding problems and issues. Please kindly review attached PDF ToolsForKnowledge.pdf for understanding my perspective.
P.S: Science is the most powerful tool known to mankind for acquiring knowledge. Whatever the excuses, not using such power tool is a monumental mistake, because necessary knowledge to easily address certain huge unaddressed problems can be acquired by using science as a tool (i.e. by employing proven scientific processes and methods).
Best Regards,
Raju
Dear Dr. Gomez,
Thank you for useful reference. The paper (you mentioned) states current state of (software) computer science. The paper is saying (software) computer science is not science at all. It implies, software does not use science as a tool at all to gain knowledge (i.e. scientific knowledge).
It is true. I agree. But I think, it is a monumental mistake not to use science as a tool to gain knowledge. Mathematics alone can’t provide knowledge essential for addressing certain software or computer science problems.
Each scientific discipline (e.g. botany, virology or bacteriology) is nothing but a body of scientific knowledge acquired by using science as a tool (i.e. by employing proven scientific processes and methods). Science is a tool for acquired such scientific knowledge.
Isn’t it astonishing that the researchers of (software) computer science ignored the most powerful and the greatest tool known to mankind for acquiring knowledge and deeper insights? Unfortunately, software researchers refusing to even consider science as a tool for acquiring useful knowledge.
Is there any evidence to show that it is impossible to use science as a tool for acquiring useful knowledge for addressing unsolved problems (e.g. software crisis or Artificial Intelligence) in software? I can't find any evidence. Any such scientific knowledge and insights never hurt any one.
My question is how to expose this monumental mistake. For example, I achieved real CBD (Component Based Design) for software products by using science as a tool to gain necessary knowledge about the nature and properties of physical components.
Best Regards,
Raju
@Raju, if the purpose of science is to explain a process or phenomena, then the method we use, i.e., top down approach vs bottom up, does not really matter, so long as it is applicable to the problem being solved.
Take my experience in economics, for instance. I used a top down approach, where the researcher devises new hypotheses and builds empirical evidence along the way.
This not to say that this approach works for everything. An empiricist would always pick bottom up (start with facts, proven results, proven theorems, etc), because the problem is different. Contrary to many of the natural sciences and most of the social sciences, the purpose here is not to explain, but to develop or validate what has been developed.
- Badges
- Science topic
- Similar topics
- Computer Science